Umami taste transduction mechanisms^,

Abstract

l-Glutamate elicits the umami taste sensation, now recognized as a fifth distinct taste quality. A characteristic feature of umami taste is its potentiation by 5′-ribonucleotides such as guanosine-5'-monophosphate and inosine 5′-monophosphate, which also elicit the umami taste on their own. Recent data suggest that multiple G protein–coupled receptors contribute to umami taste. This review will focus on events downstream of the umami taste receptors. Ligand binding leads to Gβγ activation of phospholipase C β2, which produces the second messengers inositol trisphosphate and diacylglycerol. Inositol trisphosphate binds to the type III inositol trisphosphate receptor, which causes the release of Ca²⁺ from intracellular stores and Ca²⁺-dependent activation of a monovalent-selective cation channel, TRPM5. TRPM5 is believed to depolarize taste cells, which leads to the release of ATP, which activates ionotropic purinergic receptors on gustatory afferent nerve fibers. This model is supported by knockout of the relevant signaling effectors as well as physiologic studies of isolated taste cells. Concomitant with the molecular studies, physiologic studies show that l-glutamate elicits increases in intracellular Ca²⁺ in isolated taste cells and that the source of the Ca²⁺ is release from intracellular stores. Both Gα gustducin and Gα transducin are involved in umami signaling, because the knockout of either subunit compromises responses to umami stimuli. Both α-gustducin and α-transducin activate phosphodiesterases to decrease intracellular cAMP. The target of cAMP in umami transduction is not known, but membrane-permeant analogs of cAMP antagonize electrophysiologic responses to umami stimuli in isolated taste cells, which suggests that cAMP may have a modulatory role in umami signaling.

INTRODUCTION

One hundred years ago, Kikunae Ikeda isolated l-glutamate from dried konbu and identified it as a unique taste, different from the tastes of bitter, sweet, salty, and sour. Ikeda called this taste “umami,” from the Japanese word umai, meaning delicious. A characteristic feature of umami taste is its potentiation by ribonucleotides such as inosine 5′-monophosphate (IMP) and guanosine-5'-monophosphate (GMP), which also elicit umami taste on their own. Despite Ikeda's seminal discovery, umami taste was not completely accepted as a unique taste quality until the recent molecular identification of specific G protein–coupled receptors for glutamate taste that exhibited nucleotide potentiation when expressed in heterologous cells. In this review, I will briefly describe the receptors that have been identified, although these will be covered in more detail in other chapters of this volume (1, 2). This review will focus instead on physiologic responses of taste cells to umami stimuli and describe the intracellular signaling events downstream of the umami taste receptors.

UMAMI TASTE RECEPTORS

Several receptors that bind glutamate and/or nucleotides have been identified in taste cells, including the heterodimer T1R1/T1R3 (3, 4), the taste-specific isoforms of metabotropic glutamate receptors mGluR4 (5) and mGluR1 (2, 6), and mGluR2 and mGluR3 (7), and several ionotropic glutamate receptors, including both NMDA and kainate receptors (8). One problem with identifying the potential role of these receptors in taste transduction is that glutamate also serves as a neurotransmitter, and for receptors to be considered taste receptors they must be expressed on the apical membrane of taste cells where they will encounter glutamate in the oral cavity. In this regard, both NMDA and kainate receptors have been identified on basolateral membranes of taste cells, where they likely respond to glutamate as a neurotransmitter (9). The definitive role of a particular receptor in umami transduction requires that the taste is modified when the receptor is genetically ablated. In this regard, the only receptor for which there is genetic data is the T1R1/T1R3 heterodimer. In one study, knockout of either T1R1 or T1R3 completely eliminated the responses to oral glutamate, which suggests that the heterodimer is the only umami receptor (10). However, in another study, knockout of T1R3 only eliminated nucleotide potentiation of glutamate taste responses, with little effect on responses to glutamate alone (11). Although the reason for these discrepancies is not known, the latter data strongly suggest the existence of multiple receptors for umami taste. The metabotropic glutamate receptors are the likely candidates, but knockout will be necessary to confirm a role in taste transduction.

DOWNSTREAM SIGNALING EFFECTORS

The T1R1/T1R3 heterodimer is coupled to a heteromeric G protein, where the Gβγ subunit appears to mediate the predominant leg of the signaling pathway. Ligand-binding activates Gβ3γ13, which results in activation of phospholipase C β2 (PLCβ2) (12, 13), which produces inositol trisphosphate (IP₃) and diacylglycerol. IP₃ activates the type III IP₃ receptor (IP₃R3) (14, 15),which results in the release of Ca²⁺ from intracellular stores and Ca²⁺-dependent activation of a monovalent-selective cation channel, TRPM5 (16–18). TRPM5 is expected to depolarize taste cells, which results in action potential generation and release of ATP, which activates ionotropic purinergic receptors on gustatory afferent nerve fibers (19–21) (Figure 1). Evidence of involvement of this pathway in umami taste transduction comes from several studies. First, all of these signaling effectors are co-localized with the T1R1/T1R3 heterodimer in the type II (receptor) taste cells (17). Second, knockout of PLCβ2 (17), IP₃R3 (22), and TRPM5 (23) all reduce umami taste responses in a manner similar to that of the knockout of T1R3 (11). Third, pharmacologic inhibitors of PLCβ2 and Ca²⁺ ATPase, which maintain intracellular Ca²⁺ stores, virtually eliminate responses to glutamate and nucleotides applied selectively to the taste pore in Ca²⁺ imaging studies of a lingual slice preparation (24).

FIGURE 1.

Model illustrating the signaling effectors downstream of the umami receptor T1R1/T1R3. Receptor binding activates Gβ3γ13, which in turn activates phospholipase C β2 (PLCβ2), to produce the second messengers inositol trisphosphate (IP₃) and diacylglycerol (DAG). IP₃ binds type III IP₃ receptor (IP₃R3) to elicit release of Ca²⁺ from intracellular stores and subsequent Ca²⁺ activation of TRPM5, which results in taste cell depolarization and release of ATP onto gustatory afferent fibers. In fungiform and palate taste buds, Gα gustducin or Gα transducin activates phosphodiesterase (PDE) to decrease intracellular cAMP concentrations. In circumvallate and foliate taste buds, cAMP is also decreased by activation of Gα, but likely by inhibition of adenylyl cyclase (AC) (dashed line) rather than by activation of PDE. The target of the decreased cAMP is not known, but cAMP antagonizes responses to umami stimuli in physiologic studies, which suggests that it may modulate the sensitivity of the PLC signaling pathway.

The Gα subunit that mediates umami transduction varies according to taste field. In fungiform and palatal taste buds, T1R1/T1R3 is almost completely co-localized with α-gustducin, whereas T1R1/T1R3 in circumvallate and foliate taste buds is expressed with a different and as yet unidentified Gα (25, 26). Gα-gustducin is related to Gα-transducin, which is also expressed in taste buds. Both α-gustducin and α-transducin activate phosphodiesterases (PDEs), which results in decreases in intracellular cAMP concentrations (Figure 1). Knockout of either α-gustducin or α-transducin compromises umami taste, which suggests that both Gα-gustducin and Gα-transducin participate in umami transduction (27). Physiologic studies also support a role of cAMP in umami taste. Because the activation of PDEs suppresses cAMP concentrations, cAMP should antagonize responses to umami stimuli. This has been shown in whole-cell patch clamp studies of rat fungiform taste cells, where responses to glutamate, GMP, and the synergistic response to glutamate and GMP were suppressed by membrane-permeant cAMP (28). Furthermore, biochemical studies have shown that glutamate decreases cAMP concentrations in taste buds, and the response is potentiated by 5′-nucleotides (29, 30). These latter experiments were performed on rat circumvallate taste buds, which suggests that cAMP modulates umami signaling in posterior taste fields as well, likely mediated by a Gα subunit other than α gustducin. Gαi-2 is abundantly expressed in taste buds, so this Gα may couple to T1R1/T1R3 in circumvallate and foliate taste buds (31, 32). Alternatively, a different umami receptor, such as taste-mGluR4, may mediate the responses to umami stimuli in posterior taste buds.

The role of cAMP in umami signaling is unclear. Cyclic nucleotide-gated cation channels have been identified in mammalian taste buds (33), but physiologic studies have failed to show conductance changes in response to membrane permeant cAMP analogs in taste cells. It is more likely that cAMP modulates the efficacy of Ca²⁺ signaling. Both IP₃R3 (34) and PLCβ2 (35) are modulated in other tissues by cAMP-dependent phosphorylation, and, in both cases, phosphorylation decreases the Ca²⁺ released from intracellular stores. Additional studies will be required to determine whether decreases in cAMP mediated by α-gustducin, α-transducin, or αi-2 modulate Ca²⁺ signaling in taste buds. (Other articles in this supplement to the Journal include references 1, 2, and 36–62.)