Studies on Umami Taste
J. Agric. Food Chem., Vol. 56, No. 3, 2008 1049
between MSG and ribonucleotides and, as a consequence, the
possibility of L-glutamate concentrations being reduced in foods
by addition of small amounts of IMP and/or GMP (4).
to electrostatic interaction between the positively charged
exocycic NH2 (26, 32) and the phosphate group as well as to a
hydrogen bond between the same amino group (weakly pyra-
2-
midalized) ( (26)) and one oxygen atom of PO3 (N2---O
Inspection of Figure 2 suggests that the synergistic activity
of N2-substituted guanylic acids is related to the chain length
of the alkyl or acyl substituent. In both of these series such
activity reaches a maximum, i.e., about 6 times the activity of
IMP, for a number of terms equal to six (including NH and S
as an equivalent of CH2). It can be noted that the synergistic
efficacy decreases slightly by replacing the CH2 group in the
R-position of the aliphatic chain with a carbonyl group (cf. 6a
vs 9a, 6e vs 9d, and 6b vs 9b). An opposite effect was observed
when the CH2 group in the γ- or δ-position of the alkyl chain
was substituted with an S atom (cf. 6a vs 6d and 6e), while in
the case of acyl chains the effect appears to be noticeable only
for the sulfur substitution in the δ-position (cf. 9e and 9c vs 9f
and 9d and also 9g). The contribution of S atoms present in the
N2-substitutent of GMPs to the synergistic effect is understand-
able, considering the importance of sulfur compounds in the
formation of flavors (12, 30), in particular of meat flavors. This
fact, together with the above SAR data, indicates that the
substituent linked to the exocyclic NH of the guanine moiety
is actively implicated in the sensory perception of MSG, in
agreement with Kuninaka’s hypothesis (3) on the binding mode
of 5′-ribonucleotides to the taste receptor.
distance ) 2.68 Å; N2sH---O angle ) 163.5°) Such factors of
stabilization do not occur in syn conformers of N2-substituted
GMPs either in Vacuo or in water (see compound 6d in Figure
4C′,D′): in fact, in these compounds hydrogen bonding is
precluded by a preferred orientation of the N2HR group coplanar
with the pyrimidine ring (R being an alkyl or acyl substituent).
Such orientation may be the result of steric repulsion (A(1,3)
strain) (33) and/or hydrophobic attractions.
If we assume that the continuity in the synergistic effect we
observed going from IMP to 9b (Figure 2) implies a substantial
similarity in the conformation adopted by all of the nucleotides
while interacting with the receptor protein, then the anti
conformation appears to be the most probable one.
ACKNOWLEDGMENT
We thank Sergio Crippa and Teresa Recca (Dipartimento di
Chimica Organica e Industriale) for recording NMR spectra.
Supporting Information Available: Synthesis of aldehydes
3a-e and acyl chlorides 7a-g; 1H and 13C NMR data of
1
nucleosides 5a-e and 8a-g; MS, H, 13C, and 31P NMR data
A number of theories have been proposed to explain the taste
potentiating phenomenon (29). Most of them are interrelated,
and all invoke combined allosteric interactions between receptor
proteins and flavor enhancement molecules. In particular, it has
been suggested that the function of GMP is to unmask the
receptor site for the L-glutamate ion (as well as that of other
chemical stimuli), making it more available for the attachment
of its specific ligand (29). It is known that thermodynamic and
conformational parameters are of key importance in describing
the structure and stability of nucleotide/protein (enzyme)
complexes. For instance, it was found that the activation of
glycogen phosphorylase b by purine nucleotides requires the
latter to be as an anti conformer (31). Taking account of these
considerations, we have carried out theoretical calculations to
estimate conformational preferences and related energies of IMP,
GMP, and some of the GMP derivatives we have synthesized.
Calculations were performed assuming the mononucleotide to
be present as a dianion both in gas phase and in aqueous
solution, the former condition being useful to gain insight into
the intrinsic features of the molecule.
The data of Table 2 show that the anti conformation is
energetically favored in all of the purine ribonucleotides
examined, both in aqueous solution and in Vacuo, with the
exception of GMP in Vacuo. In addition, the anti conformers
are characterized by very small differences in the geometrical
parameters of the nucleoside moiety. So, a common global
geometry can be assigned to IMP, GMP, compounds 6a,d and
9a,c,d both in gas (ꢁ, ca. -150°; P, ca. 125°) and in condensed
phase (ꢁ, ca. -170°; P, ca. 139°), apart from the orientation of
the 5′-phosphoryl group (ꢀ and γ are g– in Vacuo but t in water)
(see Figures 3 and 4A-D). By contrast, structural dissimilarities
are evident among the 5′-nucleotides in the two sets of syn
conformers. In particular, we can note a marked difference in
the orientation of the phosphoryl group between the couple IMP/
GMP and the other compounds in water and also an anomalous
behavior of GMP in Vacuo. In the latter condition syn-GMP
shows a lower energy than the anti conformer and a unique
global geometry. The intrinsically more stable syn conformation
of the GMP dianion (cf. Figure 4A′) may be explained as due
of nucleotides 6a-e and 9a-g; and calculated minimum
geometries of 1a, 6a, and 9a,c,d. This material is available free
LITERATURE CITED
(1) Yamaguchi, S. Basic properties of umami and its effects on food
flavor. Food ReV. Int. 1998, 14 (2 and 3), 139–176.
(2) Yamaguchi, S.; Ninomiya, K. What is umami? Food ReV. Int.
1998, 14 (2 and 3), 123–138.
(3) Kuninaka,A.Tasteandflavorenhancers.InFlaVorResearchsRecent
AdVances; Teranishi, R., Flath, R. A., Sugisawa, H., Eds.; Marcel
Dekker: New York, 1981; pp 305–353, and references cited
therein.
(4) Nagodawithana, T. W. SaVory FlaVors; Esteekay Associates:
Milwaukee, WI, 1995; pp 297-333.
(5) Roininen, K.; Lähteenmäki, L.; Tuorila, H. Effect of umami taste
on pleasantness of low-salt soups during repeated testing. Physiol.
BehaV. 1996, 60, 953–958.
(6) Maruyama, Y.; Pereira, E.; Margolskee, R. F.; Chaudhari, N.;
Roper, S. D. Umami responses in mouse taste cells indicate more
than one receptor. J. Neurosci. 2006, 26, 2227–2234.
(7) Chaudari, N.; Landin, A. M.; Roper, S. D. A metabotropic
glutamate receptor variant functions as a taste receptor. Nat.
Neurosci. 2000, 3, 113–119.
(8) Toyono, T.; Seta, Y.; Kataoka, S.; Kawano, S.; Shigemoto, R.;
Toyoshima, K. Expression of metabotropic glutamate receptor
group I in rat gustatory papillae. Cell Tissue Res. 2003, 313, 29–
35, and references cited therein.
(9) Imai, K.; Marumoto, R.; Kobayashi, K.; Yoshioka, Y.; Toda, J.;
Honjo, M. Synthesis of compounds related to inosine 5′-phosphate
and their flavor enhancing activity. IV. 2-Substituted inosine 5′-
phosphate. Chem. Pharm. Bull. 1971, 19, 576–586.
(10) Mizuta, E.; Toda, J.; Suzuki, N.; Sugibayashi, H.; Imai, K.;
Nishikawa, M. Structure-activity relationship in the taste effect
of ribonucleotide derivatives. Chem. Pharm. Bull. 1972, 20, 1114–
1124.
(11) Yamaguchi, S.; Yoshikawa, T.; Ikeda, S.; Ninomiya, T. Measure-
ment of the relative taste intensity of some L-R-amino acids and
5′-nucleotides. J. Food Sci. 1971, 36, 846–849.
(12) Kuninaka, A.; Kumagai, M.; Fujiyama, K.; Ogura, M.; Sakata,
S.; Yonei, S. Flavor activity of sulfur-containing compounds