L-Glutamate and 5'-ribonucleotides such as guanosine-5'-monophosphate (GMP) and inosine-5'-monophosphate (IMP) elicit a unique taste called 'umami' that is distinct from the tastes of sweet, salty, sour, and bitter. For umami, like sweet and bitter compounds, taste signaling is initiated by binding of tastants to G-protein-coupled receptors (GPCR) in taste bud cells. To date, several GPCRs for umami compounds have been identified in taste cells, including the heterodimer T1R1/T1R3, and truncated type 1 and 4 metabotropic glutamate receptors missing most of the N-terminal extracellular domain (taste-mGluR4 and truncated-mGluR1). Apparently contradictory data in T1R3 knock-out (KO) mouse models have been reported. One study showed that behavioral preference and taste nerve responses to umami stimuli in T1R3-KO mice were totally abolished, suggesting that T1R1/T1R3 is a sole receptor for umami taste. The other reported reduced but not abolished responses to umami in T1R3-KO mice, suggesting existence of multiple receptors for umami taste. In this paper, we summarized the data from recent studies that further addressed this issue by using different experimental techniques. Some of the studies provided additional evidence for the existence of umami receptor systems mediated by mGluR1 and mGluR4 in addition to T1R1/T1R3. It is proposed that the signal mediated by the pathway involving T1R1/T1R3 may play a different role from that derived from the mGluRs. The former occurs mainly in the anterior tongue, and plays a major role in preference behavior, whereas the latter occurs mainly in the posterior tongue and contributes to behavioral discrimination between umami and other taste compounds.