Mammalian type 2 taste receptors (T2R) are a family of G protein-coupled receptors that mediate bitter signals in taste cells. In the present study, we compared the genomic organization of rodent T2R genes based on the recently completed mouse and rat genomes and examined tissue- and cell-specific expression of T2Rs. Both mouse and rat T2R families consist of 36 intact genes and at least 7 pseudogenes that are mapped to mouse chromosomes 15, 2, and 6 and to rat chromosomes 2, 3, and 4, respectively. All but two T2R genes are clustered on mouse chromosome 6 and rat chromosome 4 with virtually identical genomic organization. The orthologs of the first human T2R gene identified, mT2R119 and rT2R1, are located on mouse chromosome 15 and rat chromosome 2, whereas the novel rodent-specific T2R genes, mT2R134 and rT2R34, are located on mouse chromosome 2 and rat chromosome 3, respectively. Our results, using RT-PCR, demonstrate the presence of transcripts corresponding to the putative denatonium benzoate (DB) and phenylthiocarbamide (PTC) receptors in the antrum, fundus, and duodenum as well as in STC-1 and AR42J cells. The novel rodent-specific T2R gene (mT2R134 and rT2R34) was also expressed in these tissues and cell lines. The addition of DB, PTC, or cycloheximide to AR42J cells induced a rapid increase in the intracellular Ca2+ concentration. The specificity of these effects is shown by the fact that these bitter stimuli did not induce any detectable Ca2+ signaling in many other rodent or human cells that do not express receptors or G proteins implicated in bitter taste signaling. These results demonstrate that mouse and rat T2R genes are highly conserved in terms of genomic organization and tissue expression, suggesting that rodent T2Rs are evolved under similar dietary pressure and share bitter sensing functions in the lingual and gastrointestinal systems.
- calcium flux
- neuroendocrine cells
bitter sensing, one of the five taste modalities in the mammalian lingual system, serves as a central warning signal against ingestion of potentially toxic substances and subsequent aversion of food components with similar taste. In rodents, bitter sensing is highly developed, as demonstrated by the ability of the mouse and rat to taste a wide variety of bitter molecules known to humans (10, 28). Molecular and physiological studies showed that mechanisms underlying type 2 taste receptor (T2R)-mediated signaling transduction in response to specific bitter ligands are similar in rodents and humans (1, 5, 16).
We have previously shown that T2R genes are also expressed in the stomach and small intestine and in enteroendocrine cells of mouse origin (29). Further supporting a role of T2Rs in gastrointestinal (GI) bitter sensing, Gα gustducin (Ggust) and Gα transducin (Gt-2), which mediate sweet and bitter taste signals in taste cells, are also expressed in cells of the stomach and pancreas (13, 14, 29). These findings raised the possibility that in addition to lingual taste cells, GI neuroendocrine cells are also able to detect bitter substances. T2Rs in the GI tract are likely to perceive chemical components of ingested substances including drugs and toxins and function in the regulation of food intake and poison rejection.
Recent completion of the mouse and rat genomes allowed direct sequence comparison and functional analysis of the entire rodent T2R repertoire and thus provided better understanding of structure and function relationships of bitter sensing in mammals. Information of mouse T2R genes has been collected mostly from the earlier releases of genomic assembly in public genomic databases. Mouse T2R genes have been estimated to be in the range of 36–41 members present in 3 chromosomes (9, 24). All but two T2R genes are clustered in known bitter loci on mouse chromosome 6. However, rat T2R genes have not been systematically examined and characterized.
In the present study, we searched T2R-related sequences from public and commercial databases and examined the expression and function of novel and deorphanized T2Rs in GI tissues and cells. On the basis of the latest release of National Center for Biotechnology Information (NCBI; m33 and r2) and Celera (R13i and R1d) databases, we identified and compiled a total of 36 intact T2R genes and 7 pseudogenes present in each rodent species. Here, we report the modified version of mouse and rat T2R gene organization supported by experimental evidence and consistent with overall annotation. Furthermore, we examined the expression of rodent T2Rs that respond to denatonium benzoate (DB) and phenylthiocarbamide (PTC) and a rodent-specific T2R(mT2R134 and rT2R34) in upper GI tissues and in mouse enteroendocrine cell line STC-1 and rat pancreatic acinar tumor cell line AR42J. We then demonstrated the presence of functional T2Rs in these cells by stimulating them with bitter ligands including DB, PTC, and cycloheximide (CYX) to induce rapid Ca2+ signaling in these cells.
Our findings indicate that mouse and rat T2R gene families are highly conserved in their chromosomal localization, number of genes, and sequence homology between the orthologs. In addition, we provide further evidence of an expanded role of T2R in regulating digestive functions by demonstrating the expression of functional bitter taste receptor in rat pancreatic amphicrine cells.
MATERIALS AND METHODS
The NCBI database (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Genome, MGSC http://www.ensembl.org/Musmusculus/, RGD http://rgd.mcw.edu/) and a commerical database [Celera Discovery System (CDS), Applied Biosystems; Foster City, CA] were searched by query sequences STC 9–1 (mT2R110), and results from each database were cross-examined. Potential T2R gene candidates were compared with known T2R sequences using the Blast program at both nucleic acid (BlastN) and protein (TblastN) levels. Multiple sequence alignments were performed with ClustalW using MacVector (version 7.1, Accelerys; San Diego, CA). Protein phylogenic trees were generated based on the alignment of intact mouse and rat T2R proteins using ClustalW and TreeView (19). Domain identification was analyzed based on mammalian T2R domain IPR007960 and pfam05296.
Chromosome mapping and organization of T2R gene clusters.
Mouse and rat T2R maps were constructed based on the analysis of the latest genome information compiled from public and commercial databases (NCBI m33 and r2, CDS R13i and R1d) and experimental results. Two T2Rs are single genes located on distinct chromosomes, whereas the remainder of T2R genes are arranged as three major clusters on chromosome 6 (mouse) and chromosome 4 (rat). A final organization was constructed based on NCBI contigs (mouse: NT_039618, NT_039206, NT_039340, NT_039341, NT_039358, and NT_039359; rat: NW_047623, NW_047655, NW_047689, NW_047690, NW_047696, and NW_043770) and CDS scaffolds (mouse: GA_x6K02T2NNLH, GA_x6K02T2Q125, GA_x6K02T2P3E9, and GA_x6K02T2QD22; rat: GA_x54KRFVTF4J, GA_x9P1GCHQ3LS, GA_x9P1GCHQPBL, and GA_x9P1GCHQ5N).
Cell lines and DNA.
Murine STC-1 (a gift from D. Hanahan, Univesity of California-San Fransisco) and AR42J (CRL-1492, American Type Culture Collection; Manassas, VA) cells were cultured in DMEM-F12 supplemented with 10% fetal bovine serum, l-glutamine, and antibiotics (100 U/ml penicillin plus 50 mmol/l streptomycin and gentamycin) in a humidified atmosphere with 10% CO2 and 90% air at 37°C. Mouse (C57BL/6) and rat (Sprague-Dawley) genomic DNA were purchased from a commercial source (BD Sciences; Palo Alto, CA) or isolated from rodent cell lines (DNAzol, Invitrogen; Carlsbad, CA). Mouse or rat cDNA was synthesized from 0.5–1 μg of DNase I-treated total RNA isolated from upper GI tissues (MasterPure RNA isolation kit, EpiCentre; Madison, WI) or from 50–100 ng of poly A+-RNA (RNeasy, Qiagen; Valencia, CA) from cells using a thermostable RT-PCR kit (Invitrogen). Animal protocols were approved by the Office for Protection of Research Subjects, Chancellor’s Animal Research Committee of University of California, Los Angeles, and the Animal Research Committee of Veterans Administrations Greater Los Angeles Healthcare System.
Genomic PCR and RT-PCR.
Partial or full-length T2R coding sequences were amplified from mouse (C57BL/6) and rat (Sprague-Dawley) genomic DNA or cDNA (with and without RT) using oligodeoxynucleotide primers designed for each specific T2R. The sequences of primers used in the present study are listed in Table 2. PCR was performed in a total volume of 30 μl containing 100 ng of DNA, 300 nM of each primer in ExTaq buffer, and 2.5 units of ExTaq polymerase (TaKaRa; Madison, WI). An initial denaturation step of 94°C for 2 min was followed by 31 cycles of denaturation at 92°C for 40 s, annealing at 57°C for 40 s, and extension at 72°C for 2 min and finished with a final extension at 75°C for 5 min on a thermocycler (PTC-200, MJ Research; San Francisco, CA). RT-PCR of putative T2R was performed under the same conditions, but the total reaction number was increased to 34 cycles. The housekeeping gene acetic ribosomal protein (ARP) was used as a control for cDNA quality and relative abundance. All PCR products were separated on 1% agarose gels and stained with ethidium bromide. Gel images were recorded from the UV illuminator and analyzed by imaging software (1D Image Analysis, Kodak; Rochester, NY). The predicted T2R gene products were cloned into pCR II-TOPO vectors (Invitrogen), and their identities were confirmed by sequencing at least three positive clones.
Real-time PCR was performed by TaqMan analysis using C57BL/6 mouse GI tissues (fundus, antrum, duodenum, and liver) and STC-1 and Swiss 3T3 cell cDNA synthesized by a ThermoScript RT-PCR system (Invitrogen). PCR primers and TaqMan probes were selected using the Primer Express 1.0 software program (Applied Biosystems). Primers and probe sequences were as follows: mT2R108 receptor (amplicon size = 75 bp); forward primer 5′-GCAGAATTGCCTCTCCGGA-3′; reverse primer 5′-GAAAACAACCCCAAAGTCAGGAA-3′; and TaqMan probe 6FAM-5′-AGGATCCTGTTCAGCTTGGCCATCACTA-3′-TAMRA. RodentGAPDH primers and VIC probe (ABI P/N 4308313) were included as references. PCR was performed using the TaqMan Universal PCR Master Mix, 200 nM of both primers, and 100 nM of TaqMan probe and varying dilutions of normalized reverse-transcribed cDNA on the ABI PRISM 7700 Sequence Detection System according to manufacturer’s directions (Applied Biosystems). Data were quantitated using a comparative cycle threshold (Ct) method. For each sample, data were normalized to GAPDH, and this value (2−ΔΔCt) was adjusted so that the fundus had a mean relative mRNA level of 1. Data are presented as means ± SD (n = 3). Statistical analysis of the difference in mRNA levels was performed by two-tailed Student’s t-test.
Intracellular Ca2+ concentration ([Ca2+]i) was measured by calcium fluorometry using fura-2 AM as previously described (6). Cells were grown on 9 × 22-mm glass coverslips in 35-mm dishes. The cells were washed twice with calcium buffer [Hanks’ balanced salt solution supplemented with HEPES (pH 7.4), 1.26 mM CaCl2, 0.5 mM MgCl2, 0.4 mM MgSO4, and 0.1% BSA]. The cells were incubated at 37°C for 15 min in 1 ml of calcium buffer with 1.0 μM fura-2 AM. The cells on the coverslip were then washed three times with calcium buffer and inserted into a quartz cuvette containing 2 ml of calcium buffer, which was placed into a Hitachi F-2000 Fluorospectrophotometer, and the incubation medium was continuously stirred at 37°C. The excitation wavelengths were set at 340 and 380 nm, and the emission wavelength was set at 510 nm. Maximum fluorescence was determined by injecting 100 μ1 of 5 mM digitonin into the cuvette, and minimum fluorescence was measured after an injection of 100 μl of 0.5 M EGTA, pH 8.0. A Kd of 224 nM was used for the calcium dissociation constant from fura-2 in the cells at 37°C. [Ca2+]i was determined automatically by Cation measurement software of the F-2000 Fluorospectrophotometer.
Identification of mouse and rat T2R genes.
Our data mining of multiple genomic databases enabled the coverage of at least five times of the entire mouse and rat genomes. We used both DNA and protein blast search on public (NCBI Build m33 and r2) and commercial (CDS, August 25, 2004) databases. Recent reports by two groups based on earlier releases tabulated 36–41 genes/pseudogenes in the mouse genome (9, 24). Our analysis identified a total of 36 intact T2R genes in both species plus 7 pseudogenes in the mouse and rat, respectively (Table 1).
Chromosomal localization of T2R gene clusters.
Except for two members described below, the rest of the T2R genes and pseudogenes are all located on the same chromosome in separate clusters defined by their physical locations (Fig. 1). Overall, mouse and rat T2R gene clusters are organized almost identically on mouse chromosome 6 and rat chromosome 4 based on our analysis of all available genomic assembly. However, a few minor but notable differences nonetheless exist between the mouse and rat genomes. First, an extensive search has not been able to identify the true orthologs for mT2R115 or for rT2R8, despite the fact that the latter sequence is highly homologous to mT2R102. Second, the rat ortholog of mT2R122 (ps7) and the mouse ortholog of rT2R14 (ps1) are presently designated as pseudogenes. Finally, the location and incidence of the pseudogenes are found to be the least conserved between the mouse and rat (Table 1). So far, at least seven pseudogenes of each rodent species have been identified, mostly resulting from nonsense mutation or by deletion.
Our rodent T2R gene mapping also revealed critical differences in current T2R annotations from the NCBI and CDS databases. First, the orientation of a 140-kb region covering 11 mT2R genes/pseudogenes on mouse chromosome 6 is ambiguous due to an 80-kb gap between the BAC ends. This discrepancy resulted in the assignment of mT2R109 (by CDS) instead of mTR123 (by NCBI) next to mT2R117, thus reversing the orientation of the next 10 T2R members (Fig. 1, boxed area). Another mapping discrepancy is the assignment of rT2R8 and its four neighboring T2R members. The present assignment by CDS and an earlier version of NCBI (NW_043770) are similar (Fig. 1 and Table 1), but, in the present NCBI assembly, it is assigned to the end of the supercontig (NW_047696).
Chromosomal localization of two single T2R genes.
Two T2Rs are single genes located on distinct chromosomes (mT2R119 at 15B3.1 and mT2R134 at 2C1.1; rT2R1 on 2q22 and rT2R34 at 3q12). Each shares the highest sequence homology with its respective ortholog at the syntenic region. The first pair of T2R genes (mT2R119 and rT2R1) are the orthologs to hT2R1, which is a gene originally found in a contig where the PROP locus that controls the detection of the bitter compound 6-n-propyl-2-thiouracil (PTU) in humans is located (16). The second pair, mT2R134 and rT2R34, was identified from the syntenic chromosome regions at which no bitter locus was previously known in rodents. Molecular evidence suggests that it is a novel candidate for the bitter taste receptor based on its sequence homology with other functionally defined T2Rs and possession of signature residues in T2R-specific domains IPR007960 and pfam05296. However, the syntenic region on human chromosome 2q23.3 does not harbor any T2R-related or GPCR-like sequence. Thus mT2R134/rT2R34 can be regarded as a unique rodent-specific T2R, and, interestingly, it shares the highest sequence homology to a human pseudogene (hT2R62) on chromosome 7.
On the basis of previous reports, the human genome contains 15 T2R genes and 5 pseudogenes on chromosome 12, 9 gene and 3 pseudogenes on chromosome 7, and a lone member, hT2R1, on chromosome 5 (8, 11). A comparison between human and rodent T2R genes indicates that significantly higher numbers of rodent T2R genes exist as a result of higher frequency of gene duplication. Thus rodents possesses about 30% more bitter taste receptors and probably wider spectrum of bitter sense detection (Table 1).
Phylogenic analysis of rodent T2R family.
We then conducted phylogenetic analysis of rodent T2R (excluding pseudogenes), and a dendrogram was constructed using ClustalW alignment (Fig. 2). Rodent orthologs were paired consistently with only a few exceptions described earlier. It became more apparent that overall rodent T2R genes are highly conserved. Each ortholog pair has a sequence identity of at least >68% at protein and >80% at nucleic acid coding sequences. The orthologs of mT2R122 (rat ps7) and rT2R14 (mouse ps1) are pseudogenes and share low homology (<50%) with each other. It is intriguing that no rat ortholog has been identified for mT2R115 based on sequence homology and physical location.
Results from phylogenetic analysis also indirectly support our current T2R gene mapping (Fig. 2), which depicts that mT2R117 and mT2R109 (64% identity) and rT2R24 and rT2R21 (58% identity) are two adjacent homologs that may arise from more recent gene duplication (Fig. 2). In agreement with the consensus assembly data, mT2R124 and mT2R102 (orthologs of rT2R24 and rT2R21) and rT2R17 and rT2R26 (orthologs of mT2R117 and mT2R109) are physically associated in their respective genome (Table 1).
Expression of T2R in upper GI tissues.
Recently, we demonstrated the expression of bitter taste receptors of the T2R family in mouse and rat gastroenteric systems. To extend these findings, we determined whether functional T2Rs are expressed in GI tissues and cells. Specifically, we performed RT-PCR to detect transcripts encoding for mT2R108 and mT2R138, whose putative ligands are DB and PTC, respectively. In addition, we also examined the expression of the unique rodent-specific mT2R134 in GI tissues and in the tongue, which is the primary site of expression for all functional taste receptors. As shown in Fig. 3, mT2R138 and mT2R108 were detected in the fundus, antrum, duodenum, and tongue but not found in the liver. In addition, the rodent-specific mT2R134 was similarly expressed in gastric and duodenal tissues as well as in the tongue.
The rat orthologs of putative PTC (rT2R38) and DB (rT2R16) receptors were also expressed in upper GI tissues but not in the liver (Fig. 4). The expression of rodent-specific rT2R34 was also observed, and its full-length coding sequence was confirmed from both cDNA and genomic DNA clones.
To compare the level of expression of a typical T2R in GI tissues, we performed real-time PCR analysis using mT2R108 as an example. As shown in Fig. 5, the relative abundance of mT2R108 mRNA was higher in the fundus and antrum compared with the duodneum, but transcripts encoding this receptor were not detected in the liver.
Expression of T2R in GI cell lines.
As a first step to identify which GI cell types express T2Rs, we examined the expression of known functional T2Rs in rodent cell lines STC-1 and AR42J (Fig. 6). RT-PCR was performed on cDNA prepared from these cells using T2R primers for mT2R108 and rT2R16 and mT2R138 and rT2R38 and for rodent-specific mT2R134 and rT2R34 (Table 2). In addition, RT-PCR analysis for Ggust and Gt-2 was performed to determine whether these essential bitter taste signaling molecules are also coexpressed in these cells.
As shown in Fig. 6, all tested mouse T2Rs and two Gα subunits were detected in the STC-1 cells by primers that generated the partial sequence of mT2R108 and full-length sequences of mT2R134, mT2R138, Ggust, and Gt-2. Similarly, expression of rat orthologs of these T2Rs (full-length sequence for rT2R16 and rT2R34 and partial sequence for rT2R38) and Gα subunits (full length) were also found in AR42J cells. Real-time PCR analysis further confirmed high levels of mT2R108 expression in STC-1 cells (Fig. 5). In contrast, no T2R or Ggust transcripts were detected in mouse Swiss 3T3 and Rat-1 and rat IEC-18 cell lines (Fig. 5 and data not shown).
Calcium response to bitter ligand in AR42J cells.
Having demonstrated that STC-1 and AR42J cells express bitter taste receptors and signaling molecules implicated in taste reception, our next step was to examine whether the addition of bitter taste compounds induces a functional response in these cells. Specifically, activation of bitter taste receptors promotes the synthesis of second messengers leading to the release of Ca2+ from intracellular stores or modulates the gating of ion channels that mediate Ca2+ entry into the cell. Previously, we demonstrated that DB and PTC elicited a rapid and marked increases in [Ca2+]i in enteroendocrine STC-1 cells (29). In the next series of studies, we determined whether the addition of the bitter compounds DB, PTC, or CYX increases [Ca2+]i in AR42J cells. As shown in Fig. 7, the addition of either DB or PTC from 1 to 8 mM induced a dose-dependent increase in [Ca2+]i in AR42J cells. The concentrations of bitter compounds used in these experiments are similar to those used for eliciting second messenger changes and ion channel activity in taste tissues (17).
To determine the specificity of the [Ca2+]i signals induced by bitter compounds in AR42J cells, we determined the effect of DB and PTC on other cell types that neither exhibit neuroendocrine properties nor express receptors or G proteins required for the reception of bitter stimuli. As shown in Fig. 8A, AR42J cells showed a prominent increase in [Ca2+]i in response to the sequential addition of PTC, DB, and bombesin (BBS), an agonist of the gastrin-releasing peptide (GRP) receptor endogenously expressed by AR42J cells. In contrast, the addition of either PTC or DB to undifferentiated rat intestinal epithelial IEC-18 cells neither induced a detectable Ca2+ response nor prevented the marked increase in [Ca2+]i induced by vasopressin in these cells (Fig. 8B). In addition, we also demonstrated that the addition of PTC or DB did not induce any increase in Ca2+ signaling in pancreatic ductal adenocarcinoma cells (BxPc3), whereas these cells responded to the addition of bradykinin, which was used as a positive control (Fig. 8C). Similarly, the addition of PTC or DB to undifferentiated rat intestinal epithelial IEC-6 cells or Swiss 3T3 fibroblasts and PANC-1 cells, another line of ductal adenocarcinoma cells, did not induce any detectable increase in [Ca2+]i (data not shown).
CYX is another compound extensively used in bitter taste studies. As shown in Fig. 8D, the addition of CYX to AR42 J cells induced an increase in [Ca2+]i. In contrast, Rat-1 cells, which do not express T2Rs, did not respond to either CYX or DB but were able to show a marked increase in [Ca2+]i in response to BBS, which was added as a positive control (Fig. 8E). Furthermore, the addition of DB, PTC, or CYX did not induce any detectable change in [Ca2+]i in human HEK-293 cells, a cell line that does not express endogenous receptors involved in bitter taste signaling (Fig. 8F). The fact that bitter stimuli, including DB, PTC, and CYX, did not induce any effect on Ca2+ signaling in a variety of cell lines that do not express T2Rs and G proteins implicated in bitter taste reception, including rat (IEC-18, IEC-6, and Rat-1), mouse (Swiss 3T3), or human (PANC-1, BxPC3, and HEK-293) cells, reinforces the notion that the effects of bitter compounds on second messenger production in AR42J cells are mediated by specific receptors and signal transducers linked to bitter taste and expressed in these cells.
Next, we examined the mechanisms by which bitter stimuli promote Ca2+ signaling in AR42J cells. Removal of extracellular Ca2+ by adding EGTA completely blocked the increase in [Ca2+]i induced by DB (Fig. 9, A and B). However, treatment of AR42J cells with pertussis toxin (200 ng/ml), an inhibitor of Gi function, did not affect the increase in [Ca2+]i induced by DB (Fig. 9, C and D). In contrast, the addition of EGTA only attenuated the Ca2+ signal elicited by PTC and BBS that binds to the endogenous GRP receptor (Fig. 10, A and B). Interestingly, treatment of AR42J cells with pertussis toxin (200 ng/ml) virtually abolished the increase in [Ca2+]i induced by PTC but not by KCl, which induces Ca2+ signaling via activation of voltage-operated Ca2+ channels in these cells (Fig. 10, C and D). These results demonstrate that different T2Rs differentially couple to Gi in the same cell type.
As a first step toward gaining more insight into the molecular basis of taste sensing, we searched the latest rodent genome databases and compared all T2R-related sequences from the mouse and rat. After filtering through the criteria that define T2R family members (e.g., sequence homology, protein motif, site of expression, and ligand specificity), we determined a total of 36 true T2R genes in each species with small variations in pseudogene numbers and locations (Table 1). Rodent genes are unevenly distributed amongst three chromosomes. Two chromosomes each contain a single T2R gene (mT2R119 and mT2R134 and their respective rat orthologs), whereas the third chromosome harbors the rest of the T2R genes and pseudogenes in multiple clusters. Similar to that found in human T2R genes, loss of function of taste receptors by polymorphism or pseudogenization is common in rodents as the selection process relaxed (5, 26, 27).
Our database mining revealed some surprising findings unknown from previous reports. A number of discrepancies in cluster orientation and gene duplication, which affect at least 11 mT2R in mouse chromosome 6 and 5 rT2R genes in rat chromosome 4, were found to exist between NCBI and CDS assemblies. Within the bitter cluster on mouse chromosome 6F3, the orientation of a region that covers 11 mT2R members remained ambiguous from the current releases. As a result, the assignment of mT2R123 or mT2R109 next to mT2R117 is uncertain, but the implication of the final draft will be significant. Direct and indirect evidence support the arrangement of this gene cluster presented here. First, a contig (GA_x6K02T2QD22) from the CDS database showed that mT2R109 is adjacent to mT2R117 in the same orientation and no other T2R-like sequence within this 22-kb intergenic region has been identified. Second, mouse BAC clones that contain three of the affected genes [mT2R123 (STC9–7), mT2R116, and mT2R110 (STC9–1)] that we have previously isolated did not contain any mT2R117 intergenic sequence (data not shown).
Phylogenetic analysis showed mT2R117 and mT2R109 shared the highest sequence homology with each other but not with the others, strongly suggesting a prior gene duplication event (Fig. 2). Indirect evidence supporting a close physical association between these two receptors came from comparable data in the rat. Rat T2R17 and rT2R26, the respective orthologs of mT2R117 and mT2R109, were assigned to the syntenic region next to each other on chromosome 4 in both NCBI and CDS rat genome assembly.
A novel and rodent-specific T2R.
A lone T2R gene identified from mouse chromosome 2 (mT2R134) and rat chromosome 3 (rT2R34) was confirmed from our PCR analysis of rodent GI and tongue tissue cDNA and genomic DNA. Thus far, no human T2R-related sequence can be found in the syntenic region on human chromosome 2. However, the neighboring genes of mT2R134/rT2R34, including Rhoe (upstream) and Nmi (downstream), are conserved among rodents and human. Even a more thorough scanning of the entire region between Rhoe and Nmi of human chromosome 2q23.3 (NT_005403) did not turn up any T2R-related genes or pseudogenes. Interestingly, hT2R, which shares the highest homology with rodent T2R134, is a pseudogene (hT2R62) (9).
Functional conservatism of rodent T2R.
The most conserved T2Rs (e.g., T2R1, T2R4, and T2R38) between human and rodents have known bitter trait or ligands [6-n-propylthiouracil (PROP)/sucrose octaacetate (SOA)/PTC]. DB is regarded as the bitterest substances perceived by human and is widely used in taste warning (4, 25). Potential T2Rs that recognize DB include hT2R4 and mT2R108 and hT2R44 and mT2R120, as demonstrated by their sensitivity to DB in a heterologous expression system (5, 22). On the other hand, the ability to taste PTC is a well-documented Mendelian trait. PTC is bitter tasting to most people but not to some (nontasters) (2). Recently, using a positional cloning approach, the PTC receptor has been identified as a member of the T2R family (hT2R38) from the bitter locus on human chromosome 7q34–35 (15, 21). Three residues account for the genetic variation of PTC receptors in humans: the dominant P88, A262, and V297 as the taster phenotype versus the recessive A, V, and I at the same positions as the nontaster phenotype. In a wider survey of T2R38 in other primates (11, 20) and in the two rodent species reported here, Pro/Ala/Val (PAV) or Pro/Ala/Ile (PAI), which confers the human taster phenotype as recently shown by Bufe (3), is conserved among primates and rodents.
In the present study, we examined the expression patterns of putative T2R to DB (mT2R108 and rT2R16) and to PTC (mT2R138 and rT2R38) in the upper GI tract and in two rodent cell lines. Both T2R transcripts were detectable in the mouse and rat stomach and duodenum but not in the liver (Figs. 3 and 4). The relative abundance of mT2R108 in GI tissues and cells were further measured by real-time PCR analysis. Consistent with results obtained by conventional RT-PCR, the mouse fundus and antrum and STC-1 cells showed higher levels of expression compared with the duodenum (Fig. 5). We also added results measuring the relative mRNA abundance of mT2R108 from mouse cell line Swiss 3T3 fibroblasts, which are not responsive to any bitter tastant tested and thus serve as a negative control. It should be pointed out that the differences in relative abundance of mT2R expression between GI tissues and a cell line has to be interpreted bearing in mind that >90% of STC-1 cells are immune positive for the α-subunit of Ggust (and higher expression of T2R), whereas endocrine cells of the mixed cell population of the GI mucosa represent <5% of the total cells.
The expression of the novel rodent-specific mT2R134/rT2R34 was prominent in GI tissues as well as in the tongue (Fig. 4). Its high overall homology and conserved signature motif to T2R suggest that T2R134 is an orphan taste receptor that recognizes yet another unidentified bitter ligand. Previously, we have demonstrated that the CYX receptor (mT2R105) and its rat ortholog (rT2R9) are expressed in STC-1 cells and rat GI tissues, respectively (27). The putative CYX receptor (rT2R9) was also detected in AR42J cells by RT-PCR (data not shown) and might confer the CYX-stimulated calcium response observed in our study.
Functional T2R in pancreatic AR42J cells.
The immortal pancreatic acinar tumor cell line AR42J is characterized by amphicrine properties, that is, AR42J cells possess both neuroendocrine and exocrine cell actions (7, 23). The neuroendocrine properties include the presence of voltage-activated Ca2+ channels and synaptic-like microvesicles containing neurotransmitters. These cells also possess zymogen granules containing digestive enzymes and can secrete them in response to receptor stimulation (12). Having demonstrated that AR42J cells express bitter taste receptors and α-subunits of taste signaling G proteins (Ggust and Gt-2), our next step was to examine whether the addition of bitter taste compounds induces a functional response in these cells. Similar to our previous report in STC-1 cells, stimulation with DB or PTC elicited a rapid increase in [Ca2+]i through influx and/or mobilization (Fig. 7). We have produced several lines of evidence indicating that these bitter agonists promote Ca2+ signaling through different pathways. For example, removal of extracellular Ca2+ prevented the increase in [Ca2+]i induced by DB but not by PTC. These results suggest that DB induces the opening of Ca2+ permeability pathways, whereas PTC stimulates the mobilization of Ca2+ from intracellular stores. Furthermore, DB and PTC receptors appear to couple to different G proteins. The increase in [Ca2+]i induced by PTC, but not that stimulated by DB, was prevented by treatment with pertussis toxin, which inactivates Gi and Go by catalyzing the ADP ribosylation of α-subunits of these subfamilies of heterotrimeric G proteins, implying that the PTC receptor requires functional Gi to stimulate Ca2+ signaling in AR42J cells (Figs. 9 and 10). It has been shown that bitter tastants activate distinct signaling pathways in the taste cells and neurons (10, 18). Thus our results demonstrate, for the first time, that pancreatic AR42J cells can be used as a cell model for studying multiple T2R-mediated signal transduction pathways initiated by diverse bitter tastants.
In conclusion, overall, rodent T2R genes are highly conserved in terms of chromosomal localization, genomic organization, ortholog sequence homology, ligand specificity, and receptor functions. The identification of bitter taste receptors in the stomach and intestine that perceive chemical components of ingested substances including drugs and toxins has a number of important implications including the design of novel molecules that modify responses initiated by activation of these receptors and extends our concept of the gustatory system. It is likely that the large family of T2Rs expressed in the stomach and intestine play a major role in mediating these responses.
In the present study, we also demonstrated that the addition of compounds widely used in bitter taste signaling (e.g., DB and PTC) to cultures of AR42J cells promoted rapid [Ca2+]i responses in these cells. Given at present the need of cultured cell model systems to study taste receptor-mediated signaling, our findings suggest that AR42J cells together with STC-1 cells, identified by us previously, may emerge as a cell model for studying T2R-mediated signal transduction.
This study was supported by National Institute of Diabetes and Digestive and Kidney Diseases RO1 Grants DK-17294, DK-56930, and DK-55003 and by CURE: Digestive Diseases Research Center Grant DK-41301.
We thank Moon Yang and Yenlin Peng for technical assistance. DNA sequencing service was provided by University of California-Los Angeles Genetics and DNA Sequencing Core. Animal tissue collection and processing were provided by Center for Ulcer Research and Education (CURE) Animal Core service.
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: E. Rozengurt, CURE/Digestive Diseases Research Center, Div. of Digestive Diseases, Dept. of Medicine, David Geffen School of Medicine, Univ. of California, 900 Veteran Ave., Rm 11-115, Los Angeles, CA 90095 (E-mail:).
- Copyright © 2005 the American Physiological Society