FASEB J. Uncover Your Biological Pathway
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.07-100289.
 (The FASEB Journal. 2008;22:2323-2330.)
© 2008 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.07-100289v1
22/7/2323    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nakajima, K.-i.
Right arrow Articles by Abe, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nakajima, K.-i.
Right arrow Articles by Abe, K.

Acid-induced sweetness of neoculin is ascribed to its pH-dependent agonistic-antagonistic interaction with human sweet taste receptor

Ken-ichiro Nakajima*, Yuji Morita*, Ayako Koizumi*, Tomiko Asakura*, Tohru Terada*,{dagger}, Keisuke Ito*, Akiko Shimizu-Ibuka§, Jun-ichi Maruyama{ddagger}, Katsuhiko Kitamoto{ddagger}, Takumi Misaka* and Keiko Abe*,{dagger},1

* Department of Applied Biological Chemistry,

{dagger} Agricultural Bioinformatics Research Unit, and

{ddagger} Department of Applied Biotechnology, Graduate School of Agricultural and Life Sciences, University of Tokyo, Tokyo, Japan; and

§ Department of Nutritional Science, Tokyo University of Agriculture, Tokyo, Japan

1Correspondence: Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, the University of Tokyo, 1–1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan. E-mail: aka7308{at}mail.ecc.u-tokyo.ac.jp


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Neoculin (NCL) is a sweet protein that also has taste-modifying activity to convert sourness to sweetness. However, it has been unclear how NCL induces this unique sensation. Here we quantitatively evaluated the pH-dependent acid-induced sweetness of NCL using a cell-based assay system. The human sweet taste receptor, hT1R2-hT1R3, was functionally expressed in HEK293T cells together with G{alpha} protein. When NCL was applied to the cells under different pH conditions, it activated hT1R2-hT1R3 in a pH-dependent manner as the condition changed from pH 8 to 5. The pH-response sigmoidal curve resembled the imidazole titration curve, suggesting that His residues were involved in the taste-modifying activity. We then constructed an NCL variant in which all His residues were replaced with Ala and found that the variant elicited strong sweetness at neutral pH as well as at acidic pH. Since NCL and the variant elicited weak and strong sweetness at the same neutral pH, respectively, we applied different proportions of NCL-variant mixtures to the cells at this pH. As a result, NCL competitively inhibits the variant-induced receptor activation. All these data suggest that NCL acts as an hT1R2-hT1R3 agonist at acidic pH but functionally changes into its antagonist at neutral pH.—Nakajima, K., Morita, Y., Koizumi, A., Asakura, T., Terada, T., Ito, K., Shimizu-Ibuka, A., Maruyama, J., Kitamoto, K., Misaka, T., Abe, K. Acid-induced sweetness of neoculin is ascribed to its pH-dependent agonistic-antagonistic interaction with human sweet taste receptor.


Key Words: G protein-coupled receptors • histidine • sweet protein


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
SWEETNESS IS A BASIC TASTE that enables organisms to accept sugars and related calorie sources in foods. Humans recognize structurally diverse sweeteners, sugars, glycosides, D-amino acids, peptides, and proteins. All these compounds are received by human sweet taste receptor hT1R2-hT1R3 (1 , 2) , which has different binding sites for different sweeteners (3 , 4) . Recent studies (5) have suggested that the receptor is expressed in the gut as well and functions as a glucose sensor to regulate the secretion of the peptide hormones. Elucidating the activation mechanism of the receptor would suggest a new target of the therapeutics of obesity and diabetes as well as that of sensory research and practice.

Neoculin (NCL) is a sweet protein present in the edible fruits of Curculigo latifolia, a tropical plant (6 , 7) , and tastes 40,000x sweeter than sucrose on a molar basis. It also has the taste-modifying activity to convert sourness to sweetness and is expected as a novel low-calorie sweetener because of its potent sweetness at weakly acidic pH. However, it has long been unclear how NCL induces this unique sensation.

Our previous experiments explained the molecular mechanism of the taste-modifying activity of NCL by pH-dependent NCL-induced activation of hT1R2-hT1R3. First, we used cultured cells transiently expressing hT1R2-hT1R3 and G16Gust25, a promiscuous G protein (G16), the C-terminal region of which was replaced with that of gustducin. The use of a cell-based assay system made it possible to quantitatively observe that NCL activated hT1R2-hT1R3 (8) . Second, human sensory tests showed that the acid-induced sweetness of NCL increased as pH lowered (8) . However, the problems were that the cell-based assay was not able to monitor pH-dependent NCL-induced activation of hT1R2-hT1R3 in a wide pH range and that in the sensory test it was difficult to monitor the pH in the mouth precisely.

In the meantime, we conducted X-ray crystallographic analysis of NCL under neutral pH conditions and showed that it has a shell-shaped, heterodimeric conformation composed of an acidic subunit (NAS) and a basic subunit (NBS) connected by two disulfide bonds (9) . Based on molecular dynamics simulations of NCL at neutral and acidic pHs, a simple receptor-ligand docking model was proposed to explain the pH-dependent taste-modifying activity of NCL. In this proposal, the conformation of NCL is in dynamic equilibrium between the two states of NAS-NBS subunits: a "closed" state at neutral pH and an "open" state at acidic pH. Initially, it was thought that only the open conformation of NCL binds hT1R2-hT1R3 to elicit sweetness at acidic pH.

In the present study, we first quantitatively evaluated the pH-dependent acid-induced sweetness of NCL using an hT1R2-hT1R3-expressing cell-based assay system. The use of this system suggested that His residues were involved in the acid-induced sweetness. We then replaced all the five His residues with Ala to produce an HA variant and found that the variant tasted sweet pH independently even at neutral pH. Since at neutral pH NCL and the HA variant activated hT1R2-hT1R3 weakly and strongly, respectively, we applied different proportions of NCL-HA variant mixtures to the cells at this pH. It revealed that NCL competitively inhibited the HA variant-induced activation of hT1R2-hT1R3. All these data suggest that NCL is exchanged between the two functionally distinct states, agonist at acidic pH and antagonist at neutral pH.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Native NCL preparation
Purification of native NCL from the fruits of C. latifolia (a kind gift from Toru Akita, Nippon Shinyaku Institute for Botanical Research, Kyoto, Japan) was carried out as described previously (7) .

NCL variant production
NCL has five His residues, two in NAS and the other three in NBS (9) . The five His residues were replaced with Ala to produce an NCL variant (HA variant). The HA variant was produced by an Aspergillus oryzae-aided expression system (10) . cDNAs for NAS and NBS variants were constructed by inverse polymerase chaing reaction-based mutagenesis. Each mutant was introduced into an A. oryzae expression plasmid to transform it to the NS-tApE strain, the pepE and tppA protease genes of which were disrupted to improve heterologous protein production (11) . Expression and purification of the HA variant were performed as described previously (10) . The HA variant was produced in the yield of 1.0 mg/L. Recombinant NCL was produced as described previously (10) .

Tastants
NCL, the HA variant, and aspartame (Nacalai, Tokyo, Japan) were each dissolved in assay buffer (10 mM HEPES, 130 mM NaCl, 10 mM glucose, 5 mM KCl, 2 mM CaCl2, and 1.2 mM MgCl2) at pH 7.2. In the cell-based assay under different pH conditions, pH values of the tastant solutions were preadjusted by the addition of citric acid or NaOH. Concentrations of tastants and pH values after application of tastant solutions are indicated in the text and figures. The delay of cell responses and the noises as nonspecific responses were often observed below pH 4.7 (data not shown).

Cell culture and transfection
HEK293T cells were cultured at 37°C in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum (JRH Biosciences, Lenexa, KS, USA) and 100 U/ml penicillin-streptomycin (Sigma-Aldrich Japan, Tokyo, Japan). For transfection, cells were seeded onto 35-mm dishes and transiently transfected with hT1R2, hT1R3, and chimeric G{alpha} protein in the ratio of 4:4:1 using the Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA). Chimeric G{alpha} mutant proteins as promiscuous G16Gust25, G16Gust44, and G15Gi3 are widely used in the functional expression of hT1R2-hT1R3 (1 , 12 , 13) . The former two of the three mutants were constructed by replacing the C terminus of G16 with the C-terminal 25 or 44 residues of gustducin and the other by replacing the C-terminal 5 residues of G15 with those of Gi3. When we functionally expressed hT1R2-hT1R3 together with G16Gust25 or G16Gust44 chimeric mutant, the cells responded to NCL at acidic pH to the same extent as they did at neutral pH (data not shown). We then used G15Gi3 chimera in this study. Transfection efficiency was ~60–80% when estimated by the visualization of cotransfected fluorescence protein.

Calcium imaging
Approximately 105 transfected cells were transferred into 96 CytoWell plates (Nalge Nunc International, Rochester, NY, USA), loaded with 2.5 µM fura-2-AM (Invitrogen), and allowed to stand at 37°C for 30 min. Cells were rinsed and incubated with 50 µl of the assay buffer at room temperature for 20 min. Then the cells were stimulated by the addition of 100 µl each of 1.5x concentrated tastant solutions. Images were recorded for 60 s at 3 s intervals after the addition. Each image field was composed of ~1000 cells. The experiment was completed within 60 min after the cells were allowed to stand at room temperature to minimize nonspecific responses to the acid. The fura-2 fluorescence intensities resulting from excitation at 340 and 380 nm were measured at 510 nm. Changes in the intracellular free calcium ion concentration were represented as changes in the ratio of fluorescence emitted at the two excitation wavelengths (F340/F380). Concentrations of tastants and pH values were kept constant during the recordings. The cells were defined to be positively activated when the increase in the F340/F380 ratio was >0.15. When the buffer solution alone was tested, only 0.3% of the cells was responsive. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2.

Statistical data analysis
The data in Figs. 1B (n=3–4), C (n=5), 2B (n=5), and 4B (n=3–4) were tested for statistical significance by Student’s t test and those in Figs. 2C (n=3) and 3 (n=3) by one-way ANOVA with Tukey’s post hoc test.


Figure 1
View larger version (39K):
[in this window]
[in a new window]

  		Figure 1. Evaluation of the pH-dependent acid-induced sweetness of NCL. A) Representative fluorescent images of cells expressing hT1R2-hT1R3 and G15Gi3 before (top) and after (bottom) addition of 13.3 µM NCL, 6.7 mM aspartame, or buffer at pH 7.2 and 5.5. The images presented are parts of image fields in B. The color scale indicates the F340/F380 ratio. Scale bar = 20 µm. B) Responses of the cells expressing hT1R2-hT1R3 and G15Gi3 to NCL. The number of cells that responded to aspartame did not change under different pH conditions. In contrast, the NCL-responding cell number at pH 7.2 (open bars) was significantly smaller than that at pH 5.5 (solid bars). The numbers of responding cells are each represented as the mean ± SE from 3–4 independent experiments. ***P < 0.001, Student’s t test. Each image field is composed of ~1000 cells. C) Differences in the NCL sweetness score under different pH conditions in human subjects. Panelists tasted NCL solutions at pH 8.0 and 4.0 and rated their sweetness within 30 s. ***P < 0.001, Student’s t test. Each bar represents the mean ± SD (n=5). D) pH-dependent responses to NCL. NCL, 13.3 µM, was applied to cells expressing hT1R2-hT1R3 and G15Gi3 under different pH conditions, pH 4.7–8.0. At low pH, the cell response to NCL was high. Half of the maximal relative response took place at pH 7.1. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2. Each point represents the mean ± SE from 3–4 independent experiments.


Figure 2
View larger version (7K):
[in this window]
[in a new window]

  		Figure 2. The pH-independent sweetness of the HA variant. A) The relative response to the HA variant under different pH conditions, pH 5.2–8.0. The HA variant, 3.3 µM, was added to cells expressing hT1R2-hT1R3 and G15Gi3. Unlike the cell response to NCL, that to the HA variant was pH independent. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2. Each point represents the mean ± SE from 3 independent experiments. B) Human sensory test. The HA variant was equally sweet at pH 4.0 (solid bar) and pH 8.0 (open bar), unlike NCL (Fig. 1 C). Each bar represents the mean ± SD (n=5). NS: not significance in Student’s t test. C) Relative response to rNCL under three different pH conditions: pH 5.5, 7.2, and 7.6. The pH dependency of the relative response to rNCL was similar to that of NCL. The relative response to rNCL approximated that of NCL under the same pH conditions. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2. Each point represents the mean ± SE from 3 independent experiments. ***P < 0.001, one-way ANOVA with Tukey’s post hoc test.

Human sensory test
The human sensory test was conducted by slight modification of the preceding study (8) . The sweetness of NCL or the HA variant was evaluated by five panel members who had fasted for more than 30 min before the sensory test. The panelists evaluated the sweetness within 30 s from the addition of a given tastant. They also tasted 300 µl each of aspartame solutions and remembered the sweetness and then received 100 µl of 40 µM NCL or the HA variant solution (pH 4.0 or 8.0) within 30 s from the addition to rate its sweetness. For the rating, the following 7-point scaling was adopted: 7 for >2.0 mM, 6 for 2.0 mM, 5 for 0.5–2.0 mM, 4 for 0.5 mM, 3 for 0.1–0.5 mM, 2 for 0.1 mM, and 1 for <0.1 mM aspartame. All the panelists gave informed consent. This study was approved by the University Committee on Human Subjects.

Molecular dynamics simulations
The initial structure of molecular dynamics simulations of the HA variant was modeled by replacing all the His residues in crystal structured NCL (PDB ID: 2D04) with Ala. Simulations were performed at neutral pH where the Asp residues were left unprotonated; NCL has no Glu residue. After energy minimization, the system was equilibrated by a 700 ps molecular dynamics simulation at 300 K. During equilibration, the positions of nonhydrogen atoms were restrained to their initial positions with harmonic potentials and the force constants of the restraints were gradually reduced. A 5 ns molecular dynamics simulation was then performed at 300 K without restraints, and snapshot structures were recorded every 1 ps. The simulations were repeated 3x with different initial velocities. As a reference, molecular dynamics simulations of NCL were also carried out at neutral and acidic pHs. Under the acidic conditions, the simulation started from its crystal structure in which all the Asp and His residues were protonated. After the systems were equilibrated, a molecular dynamics simulation was performed for 3 ns at 300 K. All the simulations were performed with the sander module of Amber 8 (14) . A modified version of the Amber parm99 force-field parameter (15 , 16) was used, and the solvation free energy was computed from the generalized Born/surface area model (17 , 18) . The structural ensemble obtained from each molecular dynamics simulation was partitioned into clusters, based on the structural similarity measured with the C{alpha} root mean square deviation (RMSD). First, structures having RMSD values <1.0 Å from the cluster center were grouped into a cluster. The grouping started with a tentative cluster center using one of the snapshot structures and was repeated until the center structure (calculated as the average of the member structures) converged. These clusters were then hierarchically clustered by using an average linking clustering method. The distance between two clusters was defined as the C{alpha} RMSD value between the cluster centers calculated for residues 1–99 of NBS after superimposing the C{alpha} atoms of residues 1–99 of NAS. The dendrogram was truncated at a distance of 5 Å to produce final clusters. For each cluster, the structure closest to the center of the largest subcluster was selected as the representative structure.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
In vitro evaluation of the pH-dependent acid-induced sweetness of NCL
To evaluate the acid-induced sweetness of NCL in vitro, we performed calcium imaging analysis of HEK293T cells transfected with hT1R2-hT1R3 under various pH conditions. We functionally expressed hT1R2-hT1R3 together with G15Gi3. When the cells were treated with 13.3 µM NCL at pH 5.5 and 7.2, the number of NCL-responding cells at pH 5.5 became significantly larger than the number of those at pH 7.2 within 60 s from the addition (Fig. 1 A, B). In contrast, when 6.7 mM aspartame was used as a control (19) , the number of responding cells was almost similar between both the pH conditions (Fig. 1A, B ).

Next, human sensory tests were performed to confirm the validity of our novel cell-based assay system. The panel members (n=5) tasted 100 µl each of 40 µM NCL solutions at pH 4.0 or 8.0, and rated the intensities of sweetness within 30 s from the addition in comparison with the standard aspartame solutions. The average sweetness score at pH 4.0 (6.8) was higher than that at pH 8.0 (3.6) (Fig. 1C ). The result was consistent with the data obtained from the cell assay (Fig. 1B ).

Using this cell-based assay, we evaluated the pH dependency of the taste-modifying activity of NCL quantitatively. The evaluation of the cellular response to a fixed concentration (13.3 µM) of NCL in the pH range of 4.7–8.0 gave the result that the relative response to NCL decreased in a pH-dependent manner as the pH was changed from 4.7 to 8.0 (Fig. 1D ). The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2. The results show that NCL-induced activation of hT1R2-hT1R3 is lessened as the pH value increases. At pH 6.3 or the vicinity, a drop in response was observed, while at pH 7.1 the cells were half as responsive as they were at pH 4.7 (Fig. 1D ). This pH-dependent sigmoidal curve resembled the protonation curve of imidazole, which has a pKa 6.5 (20) , suggesting that the His residues of NCL are involved in its acid-induced sweetness.

HA variant of NCL is a pH-independently sweet protein
To investigate whether the His residues in NCL are important for its acid-induced sweetness, an NCL variant was designed in which all the five His residues, two in NAS and three in NBS, were substituted with Ala. The resulting protein, named HA variant, was produced by an A. oryzae-based expression system (10) . The cell-based assay revealed that 3.3 µM of the HA variant, which induced a nearly saturated cell response, activated hT1R2-hT1R3 in the pH range of 5.2–8.0 (Fig. 2 A). Human sensory tests also confirmed that the HA variant tasted equally sweet with an average score of 7.0 (n=5) at pH 4.0 or 8.0 (Fig. 2B ). In contrast, recombinant NCL (rNCL), which was previously reported as a sweet protein with taste-modifying activity (10) , showed a higher relative response at pH 5.5 than at pH 7.2 and, seemingly, did so at pH 7.2 than at pH 7.6 by the cell-based assay (Fig. 2C ). Also, the relative response to rNCL approximated that of NCL under the same pH conditions (Fig. 2C ). These results indicate that the HA variant elicits strong sweetness in a pH-independent manner, with a loss of its pH-dependent acid-induced sweetness.

Dose-response relationship of NCL and the HA variant
We analyzed the dose-response relationship of NCL under different pH conditions. The cell-based assay revealed that the maximal relative response was ~0.9 with an EC50 of 1.59 µM at pH 5.5. While the maximal relative responses were almost equivalent at pH 6.2 or 7.2, the EC50 values were 10.4 µM at pH 6.2 and 40 µM at pH 7.2 (Fig. 3 ). These dose-response curves suggest that the EC50 values increased from acidic pH to neutral pH without changing the maximal relative response.


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		Figure 3. Dose-response relationship of NCL and the HA variant. Dose-response relationship found for NCL in hT1R2-hT1R3- and G15Gi3-transfected cells at pH 5.5 (red), pH 6.2 (green), and pH 7.2 (black) and that found for HA variant (blue) at pH 7.6. For NCL, EC50 values tended to decrease at lower pH while the maximal relative responses were almost constant under the different pH conditions. For the HA variant, the EC50 value and the maximal relative response at pH 7.6 resembled those of NCL at pH 5.5. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2. Colored asterisks indicate statistical significance (P<0.05) and refer to the closest dataset for which significance was observed. One-way ANOVA with Tukey’s post hoc test. Each point represents the mean ± SE from 3–4 independent experiments.

Since the HA variant elicited sweetness pH independently, we next characterized HA variant-induced activation of receptor at pH 7.6. In this case, its maximal relative response was 0.8 with EC50 of 2.0 µM, which were close to that of NCL at pH 5.5 (Fig. 3) . The result suggests that even at pH 7.6 the HA variant has the activity to activate hT1R2-hT1R3 as NCL does under acidic pH conditions.

NCL as a potential inhibitor of the HA variant-induced hT1R2-hT1R3 activation
Figure 3 shows that NCL and the HA variant activated hT1R2-hT1R3 with different EC50 values at neutral pH. We then prepared different ratios of a mixture of NCL and the HA variant and applied each of them to cells at pH 7.6. In that case, the HA variant was used at a fixed concentration (3.3 µM) and NCL at various concentrations. As a result, the cell response was suppressed as the amount of NCL increased (IC50 = 5.5 µM; Fig. 4 A). Next, the dose-response relationship of the HA variant was evaluated with NCL at 0 or 13.3 µM. In the presence of 13.3 µM NCL, the EC50 value increased from 2.0 to 3.6 µM, and the maximal relative response was not reduced (Fig. 4B ). These results indicate that NCL competitively inhibits the HA variant-induced activation of hT1R2-hT1R3 at neutral pH.


Figure 4
View larger version (13K):
[in this window]
[in a new window]

  		Figure 4. Inhibition of the HA variant-induced hT1R2-hT1R3 activation by NCL at neutral pH. A) Dose-dependent inhibition of the HA variant by NCL at pH 7.6. The cell responses to 3.3 µM the HA variant were suppressed with increasing concentrations of NCL, IC50 of 5.5 µM, at pH 7.6. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2. Each point represents the mean ± SE from 3 independent experiments. B) Dose-dependent response of hT1R2-hT1R3- and G15Gi3-transfected cells to HA variant in the absence (blue) or presence (magenta) of 13.3 µM NCL at pH 7.6. The curve obtained with the HA variant was reproduced from Fig. 3 . In the presence of NCL, the EC50 value of the HA variant increased without any reduction of its maximal relative response. The number of responding cells was normalized relative to the maximum response to aspartame (6.7 mM) at pH 7.2; *P < 0.05, Student’s t test. Each point represents the mean ± SE from 3–4 independent experiments.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Structure of HA variant
In this study, we revealed that the HA variant elicited strong sweetness both at acidic and neutral pH in a pH-independent manner (Fig. 2A, B ). In contrast, NCL activated hT1R2-hT1R3 less than the HA variant did at neutral pH (Fig. 3) . Our previous molecular dynamics simulation suggested that the conformation of NCL is pH-dependently in equilibrium between open (at acidic pH) and closed (at neutral pH) states and only the open conformation of NCL interacts with hT1R2-hT1R3 (9) . We performed molecular dynamics simulation to look at the structural model of the HA variant at neutral pH. As shown in Fig. 5 A, NCL has five His residues, two of which are presented in NAS and the other three in NBS, and all the His residues are exposed to the molecular surface. Figure 5B-D shows the representative structures in the ensembles obtained from the simulations on NCL at neutral and acidic pHs and on HA variant at neutral pH, respectively. The HA variant adopted an open conformation at neutral pH (Fig. 5D ), whereas NCL was in a closed conformation under the same pH conditions (Fig. 5B ). Also, the structure of the HA variant at neutral pH resembled that of NCL at acidic pH (Fig. 5C ). These models may explain why the HA variant elicits strong sweetness even at neutral pH. The His-to-Ala mutation would make it possible to stabilize the open conformation more than the closed one even at neutral pH.


Figure 5
View larger version (36K):
[in this window]
[in a new window]

  		Figure 5. Model of the structures of NCL and the HA variant. A) The crystal structure of NCL (PDB ID: 2D04). NAS and NBS are colored red and blue, respectively. His residues are shown with a green stick model. B–D)The representative structures of the largest clusters in the ensembles obtained from molecular dynamics simulations of NCL at neutral (B) and acidic pHs (C) and of the HA variant at neutral pH (D).

We have recently observed by intrinsic tryptophan fluorescence spectrometry that pH-induced structural change occurs in NCL, while no such change does in the HA variant (unpublished results). In confirmation of these models, we are performing X-ray crystallography on the HA variant at neutral pH. Further investigation will be needed to clarify the mechanism of pH-dependent conformational change of NCL.

Possible mechanism for the acid-induced sweetness of NCL
We previously considered that NCL hardly binds hT1R2-hT1R3 at neutral pH. In this study, however, we found that NCL can bind to hT1R2-hT1R3 at pH 7.6 to act as a competitive antagonist (Fig. 4A, B ). Therefore, the binding site of NCL as antagonist to hT1R2-hT1R3 may be overlapped to the active site of NCL as agonist.

T1R2-T1R3 consists of the three portions, long amino-terminal domain (ATD), cysteine-rich domain (CRD), and transmembrane domain (TMD) (4) . Some studies have suggested that there are multiple binding sites on the receptor for diverse sweeteners. Spectroscopic analysis using bacterially produced ATDs of mouse T1Rs showed that glucose and sucrose were bound to ATDs of mouse T1R2 and T1R3 (21 , 22) . According to the evidence from cell-based assay using human-rodent chimeric T1R, aspartame and neotame are received by ATD of hT1R2, while cyclamate and neohesperidin dihydrochalcone are received by TMD of hT1R3 and brazzein is received by CRD of hT1R3 (3 , 12 , 23 , 24) .

Recently, we found that NCL activated hT1R2-hT1R3 at the hT1R3 ATD, which had not been reported to interact with other sweeteners (25) . It will be important to define how NCL interacts with hT1R2-hT1R3.

Miraculin (MCL), another taste-modifying protein, is contained in the fruits of Richadella dulcifica (26 , 27) . Although NCL and MCL elicit the taste-modifying activity, they do not have amino acid sequence similarity to each other. Interestingly, we found that MCL also antagonized the HA variant-induced activation of hT1R2-hT1R3 at neutral pH (unpublished results). It suggests that NCL and MCL may share a common binding site in hT1R2-hT1R3.

Collating our results with the molecular dynamics simulation, we propose a new simple model for taste modification of NCL regarding the interaction between NCL and hT1R2-hT1R3. NCL is functionally and structurally in equilibrium between an open, active state and a closed, inactive one (Fig. 6 A). At neutral pH, the equilibrium shifts to the closed, inactive state. Under this condition, the open conformation of NCL is just in a minor proportion and is competitively inhibited by the closed conformation of NCL, with a slight degree of hT1R2-hT1R3 activation (Fig. 6B , left). In contrast, at acidic pH, the equilibrium shifts to the open, active state. The majority of NCL adopts the open conformation and binds to hT1R2-hT1R3 to activate hT1R2-hT1R3 strongly (Fig. 6B , right). Thus, as the pH condition is changed from neutrality to acidity, the agonist-to-antagonist ratio increases to elicit the acid-induced sweetness.


Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		Figure 6. Possible model for the acid-induced sweetness of NCL. A) Conformational and activity changes in NCL occur between acidic and neutral pHs. B) NCL molecules tend to take a closed conformation at neutral pH and act as a competitive antagonist to weaken the activation of hT1R2-hT1R3 (left). The majority of NCL molecules taking an open conformation at acidic pH acts as an agonist (right).

In conclusion, we explained the molecular mechanism of acid-induced sweetness of NCL by its pH-dependent agonistic-antagonistic interaction with hT1R2-hT1R3. This bifunctional feature of NCL will help to elucidate the sweet taste reception in paticular and also contribute to a new insight into the activation mechanism of G protein-coupled receptors in general.


   ACKNOWLEDGMENTS
 
This study was performed with a grant from the Research and Development Program for New Bioindustry Initiatives. This work was supported in part by a Japan Society for the Promotion of Science Research Fellowship for Young Scientists 1903545 (to K.N.) and by grants-in-aid for scientific research 19300248 (to T.A.), 18688005 (to T.M.), and 16108004 (to K.A.) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. The authors thank Dr. Toru Akita (Nippon Shinyaku Institute for Botanical Research, Kyoto, Japan) for providing C. latifolia fruits.

Received for publication October 23, 2007. Accepted for publication January 10, 2008.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Li, X., Staszewski, L., Xu, H., Durick, K., Zoller, M., Adler, E. (2002) Human receptors for sweet and umami taste. Proc. Natl. Acad. Sci. U. S. A. 99,4692-4696[Abstract/Free Full Text]
  2. Chandrashekar, J., Hoon, M. A., Ryba, N. J., Zuker, C. S. (2006) The receptors and cells for mammalian taste. Nature 444,288-294[CrossRef][Medline]
  3. Xu, H., Staszewski, L., Tang, H., Adler, E., Zoller, M., Li, X. (2004) Different functional roles of T1R subunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. U. S. A. 101,14258-14263[Abstract/Free Full Text]
  4. Cui, M., Jiang, P., Maillet, E., Max, M., Margolskee, R. F., Osman, R. (2006) The heterodimeric sweet taste receptor has multiple potential ligand binding sites. Curr. Pharm. Des. 12,4591-4600[CrossRef][Medline]
  5. Jang, H. J., Kokrashvili, Z., Theodorakis, M. J., Carlson, O. D., Kim, B. J., Zhou, J., Kim, H. H., Xu, X., Chan, S. L., Juhaszova, M., Bernier, M., Mosinger, B., Margolskee, R. F., Egan, J. M. (2007) Gut-expressed gustducin and taste receptors regulate secretion of glucagon-like peptide-1. Proc. Natl. Acad. Sci. U. S. A. 104,15069-15074[Abstract/Free Full Text]
  6. Yamashita, H., Theerasilp, S., Aiuchi, T., Nakaya, K., Nakamura, Y., Kurihara, Y. (1990) Purification and complete amino acid sequence of a new type of sweet protein taste-modifying activity, curculin. J. Biol. Chem. 265,15770-15775[Abstract/Free Full Text]
  7. Shirasuka, Y., Nakajima, K., Asakura, T., Yamashita, H., Yamamoto, A., Hata, S., Nagata, S., Abo, M., Sorimachi, H., Abe, K. (2004) Neoculin as a new taste-modifying protein occurring in the fruit of Curculigo latifolia. Biosci. Biotechnol. Biochem. 68,1403-1407[CrossRef][Medline]
  8. Nakajima, K., Asakura, T., Oike, H., Morita, Y., Shimizu-Ibuka, A., Misaka, T., Sorimachi, H., Arai, S., Abe, K. (2006) Neoculin, a taste-modifying protein, is recognized by human sweet taste receptor. Neuroreport 17,1241-1244[CrossRef][Medline]
  9. Shimizu-Ibuka, A., Morita, Y., Terada, T., Asakura, T., Nakajima, K., Iwata, S., Misaka, T., Sorimachi, H., Arai, S., Abe, K. (2006) Crystal structure of neoculin: insights into its sweetness and taste-modifying activity. J. Mol. Biol. 359,148-158[CrossRef][Medline]
  10. Nakajima, K., Asakura, T., Maruyama, J., Morita, Y., Oike, H., Shimizu-Ibuka, A., Misaka, T., Sorimachi, H., Arai, S., Kitamoto, K., Abe, K. (2006) Extracellular production of neoculin, a sweet-tasting heterodimeric protein with taste-modifying activity, by Aspergillus oryzae. Appl. Environ. Microbiol. 72,3716-3723[Abstract/Free Full Text]
  11. Jin, F. J., Watanabe, T., Juvvadi, P. R., Maruyama, J., Arioka, M., Kitamoto, K. (2007) Double disruption of the proteinase genes, tppA and pepE, increases the production level of human lysozyme by Aspergillus oryzae. Appl. Microbiol. Biotechnol. 76,1059-1068[CrossRef][Medline]
  12. Jiang, P., Ji, Q., Liu, Z., Snyder, L. A., Benard, L. M., Margolskee, R. F., Max, M. (2004) The cysteine-rich region of T1R3 determines responses to intensely sweet proteins. J. Biol. Chem. 279,45068-45075[Abstract/Free Full Text]
  13. Galindo-Cuspinera, V., Winnig, M., Bufe, B., Meyerhof, W., Breslin, P. A. (2006) A TAS1R receptor-based explanation of sweet "water-taste.". Nature 441,354-357[CrossRef][Medline]
  14. Case, D. A., Darden, T. A., Cheatham, T. E., III, Simmerling, C. L., Wang, J., Duke, R. E, Luo, R., Merz, K. M., Wang, B., Pearlman, D. A., Crowley, M., Brozell, S., Tsui, V., Gohlke, H., Mongan, J., Hornak, V., Cui, G., Beroza, P., Schafmeister, C., Caldwell, J. W., Ross, W. S., Kollman, P.A. (2004) AMBER 8 University of California San Francisco, CA, USA.
  15. Wang, J., Cieplak, P., Kollman, P. A. (2000) How well does a restrained electrostatic potential (RESP) model perform in calculating conformational energies of organic and biological molecules?. J. Comput. Chem. 21,1049-1074[CrossRef]
  16. Simmerling, C., Strockbine, B., Roitberg, A. E. (2002) All-atom structure prediction and folding simulations of a stable protein. J. Am. Chem. Soc. 124,11258-11259[CrossRef][Medline]
  17. Still, W. C., Tempczyk, A., Hawley, R. C., Hendrickson, T. (1990) Semianalytical treatment of solvation for molecular mechanics and dynamics. J. Am. Chem. Soc. 112,6127-6129[CrossRef]
  18. Onufriev, A., Bashford, D., Case, D. A. (2004) Exploring protein native states and large-scale conformational changes with a modified generalized born model. Proteins 55,383-394[CrossRef][Medline]
  19. Schiffman, S. S., Sattely-Miller, E. A., Graham, B. G., Bennett, J. L., Booth, B. J., Desai, N., Bishay, I. (2000) Effect of temperature, pH, and ions on sweet taste. Physiol. Behav. 68,469-481[CrossRef][Medline]
  20. Alberts, B. J., Lewis, A., Raff, J., Roberts, M., Walter, K. P. (2002) Molecular Biology of the Cell 4th Ed. ,62-64 Garland Science, Taylor and Francis Group New York, NY, USA.
  21. Nie, Y., Vigues, S., Hobbs, J. R., Conn, G. L., Munger, S. D. (2005) Distinct contributions of T1R2 and T1R3 taste receptor subunits to the detection of sweet stimuli. Curr. Biol. 15,1948-1952[CrossRef][Medline]
  22. Nie, Y., Hobbs, J. R., Vigues, S., Olson, W. J., Conn, G. L., Munger, S. D. (2006) Expression and purification of functional ligand-binding domains of T1R3 taste receptors. Chem. Senses 31,505-513[Abstract/Free Full Text]
  23. Jiang, P., Cui, M., Zhao, B., Snyder, L. A., Benard, L. M., Osman, R., Max, M., Margolskee, R. F. (2005) Identification of the cyclamate interaction site within the transmembrane domain of the human sweet taste receptor subunit T1R3. J. Biol. Chem. 280,34296-34305[Abstract/Free Full Text]
  24. Winnig, M., Bufe, B., Kratochwil, N. A., Slack, J. P., Meyerhof, W. (2007) The binding site for neohesperidin dihydrochalcone at the human sweet taste receptor. BMC Struct. Biol. 7,66[CrossRef][Medline]
  25. Koizumi, A., Nakajima, K., Asakura, T., Morita, Y., Ito, K., Shmizu-Ibuka, A., Misaka, T., Abe, K. (2007) Taste-modifying sweet protein, neoculin, is received at human T1R3 amino terminal domain. Biochem. Biophys. Res. Commun. 358,585-589[CrossRef][Medline]
  26. Kurihara, Y. (1992) Characteristics of antisweet substances, sweet proteins, and sweetness-inducing proteins. Crit. Rev. Food Sci. Nutr. 32,231-252[Medline]
  27. Ito, K., Asakura, T., Morita, Y., Nakajima, K., Koizumi, A., Shimizu-Ibuka, A., Masuda, K., Ishiguro, M., Terada, T., Maruyama, J., Kitamoto, K., Misaka, T., Abe, K. (2007) Microbial production of sensory-active miraculin. Biochem. Biophys. Res. Commun. 360,407-411[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.07-100289v1
22/7/2323    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nakajima, K.-i.
Right arrow Articles by Abe, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nakajima, K.-i.
Right arrow Articles by Abe, K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS