Food Flavor and Safety - American Chemical Society

---

Chapter 3

Downloaded via UNIV OF CALIFORNIA SANTA BARBARA on July 12, 2018 at 00:19:33 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Do We Recognize Sweetness and Bitterness at the Same Receptor? Kozo Nakamura and Hideo Okai Department of Fermentation Technology, Faculty of Engineering, Hiroshima University, Higashihiroshima, Hiroshima 724, Japan

The relationship between taste and chemical structure has been studied. The experimental data suggest that sweetness and bitterness are recognized at the same receptor. Furthermore, the receptor discriminates between bitter and sweet tastes based upon differences in functional unit combination. A new taste receptor model is proposed and presented. The new model explains many taste phenomena that the older model can not. Current Mechanism for Sweetness and Bitterness Perception. Numerous sweet substances exist having a wide range of chemical structures. In 1967, Shallenberger and coworkers proposed the AH-B theory showing a possible common molecular feature in sweet substances (/). They reported all sweet substances had a proton donor (AH) and a proton acceptor (B) in the molecule with a distance estimated to be an average of 3Â (2). They concluded that sweet taste was produced by the interaction between AH, Β and the corresponding receptor sites. However, the intensity of sweetness could not be explained by the AH-B theory. Many chemists pointed out that a hydrophobic group participate in the intensity of sweetness. Some chemists propagated the theory that a hydrophobic group is one of the sweetness producing units and that sweetness is exhibited by X, AH, and B. This chapter will discuss this newer proposal in light of recent experimental data. Mechanism for Sweetness Exhibition of Aspartame. Aspartame, discovered by Mazur in 1969 (3), is 200 times sweeter than sucrose. Aspartame has a large commercial market as an artificial sweetening agent. It is apparent that the sweetness exhibited by aspartame requires amino (AH, electropositive) and carboxyl (B, electronegative) groups of aspartic acid moiety and the hydrophobic side chain (X) of the phenylalanine moiety (4). The sweetness of aspartame is exhibited by the trifunctional units AH, B, and X. It is thought that when the trifunctional units of aspartame, X, AH, and B, fit the corresponding receptor sites, a sweet taste is produced. An interesting experiment using the model of aspartame (Figure 1) was carried out in 1985. Combinations of two or three different molecules were prepared to examine whether or not sweetness could be produced in the same system as aspartame. The 0097-6156/93/0528-0028$06.00/0 © 1993 American Chemical Society

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

3.

NAKAMURA AND OKAI

Recognizing Sweetness and Bitterness

29

results of the sensory studies show sweetness was produced with several of the test compounds (Table I). It was found that the aspartame molecule is not always necessary for sweetness production, i.e. sweet taste is easily produced by a suitable combination of the components, AH, B, and X (5). The strongest sweetness was observed in the compounds with the combination of AH-X and B. The AH-X component produced bitterness while the Β component produced sourness.

Aspartame AH) I

ΓΒΙ

_ΓΑΊ

ΓΒΊ

TYPEc TYPEd Fig. 1. The Sweetness Production of Aspartame Caused by the Interaction between Three Sweet Units(AH, X , B) and Corresponding Receptor Sites (A , X , B*) (Adopted from ref. 5) 1

1

Mechanism for Bitterness Exhibition. About one thousand bitter peptides were systematically synthesized. The relationship between bitterness and chemical structure of the peptides have been studied (6-72). These studies have now defined clearly the mechanism for bitterness production by peptides. Bitter peptides were found to possess two active sites, a "biding unit" and "stimulating unit." Hydrophobic groups acted as biding units. Bulky basic groups and hydrophobic groups would acts as stimulating units. Bitterness was produced when these active units attached to a corresponding bitter taste determinant. The distance between the two sites was estimated to be 4.1Â (77). A model of the receptor was proposed and is seen in Figure 2. This model is well suited to describe the production of bitterness for all bitter compounds (13).

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

30

FOOD FLAVOR AND SAFETY

Table I. The Results of Sweetness Production by the combination of ΑΗ,Χ and Β combination component** sweetness + TYPE a AH-i Β 5 m M acetic acid 5 mM H-L-Ala-L-Phe-OMe yes 5mM (bitter TV*1.2mM) citric acid yes 5 m M acetic acid 5 m M H-L-Ala-D-Phe-OMe yes 5mM (bitter T V 1 . 4 m M ) citric acid yes 5 m M acetic acid 5 m M H-L-Gly-L-Phe-OMe yes 5mM (bitter TV1.2mM) citric acid yes 5 m M acetic acid 5 m M H-L-Gly-D-Phe-OMe yes 5mM (bitter TV1.2mM) citric acid yes + TYPEb AH X-B 5 mM aq NH3 5 m M succinyl-L-Phe-OMe no 5 mM aq NH3 5 m M succinyl-D-Phe-OMe no 5 mM aq NH3 5 m M glutaryl-L-Phe-OMe no 5 mM aq NH3 5 m M glutaryl-D-Phe-OMe no 5 mM aq NH3 5 m M Bz-NH-(CH2)-COOH no 5 mM aq Et3N 5 m M Bz-NH-(CH2)-COOH yes 5 mM NH2C6H11 5 m M Bz-NH-(CH2)-COOH yes + TYPEc AH-B X 5 mM NH2-(CH2)n-COOH 5 mM Bzl-OH no 20 mM NH2-(CH2)n-COOH 10 mM Bzl-OH no TYPEd AH + X + Β 5 mM aq NH3 5 mM Bzl-OH 5 mM acetic acid no 5mMaqEt3N 5 mM Bzl-OH 5 mM citric acid no 5 mM aq NH2C6H11 5 mM Bzl-OH 5 mM acetic acid yes 5 mM aq NH2C6H11 5 mM Bzl-OH 5 m M citric acid yes * Threshold value. ** Individual taste of AH, X , B, and AH-B components: A H , 5 mM aq NH3, 5 mM aq E G N , and 5 mM cyclohexylamine, is almost tasteless (undeterminable taste). X, 5 mM and lOmM Bzl-OH, is almost tasteless (undeterminable taste ). B, 5 mM acetic acid, 5 mM citric acid is sour. AH-B, all of them ( 5 mM, 20 mM ), is tasteless.

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

3.

NAKAMURA AND OKAI

Recognizing Sweetness and Bitterness

31

Similarity of Sweetness and Bitterness Perception. Figure 3 shows the model for both sweetness and bitterness production. The X component of a sweet compound corresponds to the binding unit of a bitter compound. The A H component of a sweet compound corresponds to the stimulating unit of a bitter compound. The A H component of a sweet compound corresponds to the stimulating unit of a bitter compound. The distances between AH and X , and between the biding units and stimulating units are essentially the same. The lack of the Β component is the only factor that distinguishes between bitterness and sweetness perception. Although the taste phenomenon that is changing bitterness to sweetness by adding sourness appeared strange at first, it is explained readily by this new receptor model.

BITTER PEPTIDE B U : binding unit ( hydrophobic group ) S U : stimulating unit ( hydrophobic or basic group )

Fig. 2. Bitterness Receptor Model (Reproduced with permission from ref. 10. Copyright 1988 Japan Society for Bioscience, Biotechnology, and Agrochemistry)

SWEET BITTER Fig. 3. The Model of Sweetness and Bitterness Production at the Taste Receptor

Sweetness Production by the Combination of Bitter and Sweet Tastes. Sensory tests using typically bitter compounds such as brucine, strychnine, phenylthiourea, caffeine and bitter peptides were performed. Sensory tests using typically bitter compounds such as brucine, strychnine, phenylthiourea, caffeine and bitter peptides were performed. Sensory taste impression were also measured for combinations of acetic acid (sour) and typical bitter compounds (5). The data from these studies indicated that the tastes of these bitter/sour mixtures changed to a sweet taste regardless of their chemical structure of the bitter component (Table Π). From the experimental results, it could be concluded that sweetness and bitterness are recognized in the same taste receptor and the receptor discriminates between bitter and sweet tastes based upon the difference in the functional unit combination. The currently accepted theory is that sweetness and bitterness are recognized by individual specific taste receptors. Datafromthis laboratory propose an alternative mechanism. Data is presented below to demonstrate which theory is most plausible. Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

32

FOOD FLAVOR AND SAFETY

Table II. Tasting Behavior of Combination of Typical Bitter Compounds and Acetic Acid taste**** bitter compound Β bitter -> sweet 0.002 mM brucine (TV *0.0008 mM) + 5 mM acetic acid bitter -» sweet 0.005 mM strychnine (TV 0.003 mM) + 5 mM acetic acid bitter -> sweet 0.1 mM phenylthiourea (TV 0.025mM) + 5 mM acetic acid bitter -> sweet 0.2 mM cyclo(-Leu-Trp-)** (TV 0.06 mM) + 5 mM acetic acid 0.2 mM Arg-Gly-Pro-Pro-Phe-Ile-Val*** bitter -> sweet (TV 0.05 mM) + 5 mM acetic acid SOURCE: Adapted from ref. 5 * Threshold value. ** A bitter peptide by Shiba and Numami, 1974. *** A bitter peptide by Fukui et al., 1973. **** Bitterness + Sourness -> Sweetness. The taste were determined according to the method which has been described in detail in ref. 5.

Taste Behavior in Mixed Sweet and Bitter Solutions. The current accepted theory suggests that a bitter compound and a sweet compound bind independently at specific receptors. This situation will be referred to as "independent" in this report. The data to follow will demonstrate that a bitter compound and a sweet compound bind at the same receptor in a competitive manner. Therefore, this situation will be referred to as "competitive" in this report. Which theory was the functioning mechanism of taste reception should be determinable when one measured the taste intensities of mixed solutions of bitter and sweet tasting compounds. In this experiment the mechanism could be predicted to elicit a considerable difference in taste intensity and response that was varying based on the final concentration of each component. The "independent" receptor mechanism would be expected to yield data in which the intensities of bitter and sweet would be unaffected by mixing the two tastes, no matter what the concentration. On the other hand, with the "competitive" receptor mechanism one would expect both flavors to become altered, i.e., one stronger and the other weaker, as component concentrations varied; the latter would occur because of competition of the substances for the same site. Two kinds of mixed solutions were provided to test the hypothesis. One was a mixed solution of D-phenylalanine and L-phenylalanine. The second was a mixed solution of sucrose and L-phenylalanine. While the chemical properties of D- and L phenylalanine are the same, for the most part, they do possess different tastes. Sucrose has a structure and taste that is significantly different from L-phenylalanine. The sensory data obtainedfroma five member sensory panel is shown in Table III. Table III. The score of sweetness and bitterness SCORE* 0 1 2 3 4 5 L-Phe 02030 40 50 60CONC** D-Phe 02481420(mM) Sucrose 05102030 40* The scores represent the intensity of sweetness or bitterness. Score 1 indicates bitterness of ImM caffeine solution or sweetness of 5mM sucrose solution. The taste-intensities become progressively stronger as the scores increase. **The concentration of the aqueous solution of each compound.

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

3.

NAKAMURA AND OKAI

Recognizing Sweetness and Bitterness

33

Theoretical sensory sweetness and bitterness scores for the two theories are described below for mixed solutions of sweet and sour. This exercise considers the case of a mixture of 20 mM D-phenylalanine (D-PA) and 60 mM L-phenylalanine (LPA). Individually each of these solutions yield a sensory score of 5.0, i.e. the sweetness of 30 mM D-PA is 5 and the bitterness score of 60 mM L-PA is 5. Theoretically, if both compounds bind independently at different taste receptors, the sweetness and bitterness should each be 5. On the other hand, if both compound bind at the same taste receptor in a competitive manner, the sweetness and bitterness scores should decrease. In regards to the latter, the sweet and bitter compound were mixed in a ration of 1:3 (20 mM to 60 mM) which yields a probability of fitting a taste receptor of l-in-4 for sweet and 3-in-4 for bitter. If these probabilities are multiplied by the original concentrations, i.e. 20 mM and 60 mM, the products would be in the expected concentration of 5 mM (1/4 of 20mM) and 45 mM (3/4 of 60mM). The sweetness of a 5 mM solution and the bitterness of a 45 mM solution corresponds to a sensory score of 2 and 3, respectively, and represent the expected sensory score for the competitive hypothesis. Results from sensory evaluation of mixed solution are seen in Table IV. The data list the theoretical response for both the independent and competitive receptor hypothesis as well as the actual sensory score. The actual sensory scores were found to agree fairly well with the competitive model. The minor dissimilarity between the actual and theoretical is due to the inability of individual to taste bitterness in solutions that are extremely sweet, i.e., there is some masking of overall sensory perception which is concentration dependent. The data, therefore, clearly indicate that sweetness and bitterness act in a competitive manner and should be considered to compete for the binding sites at the same receptor.

Table IV. The Results of Sensory Analysis of Mixed Solution combination sweet bitter D-Phe + L-Phe

original score

expected value INDEPENDENTLY COMPETITIVELY

RESULTS*

2 - 3 20mM - 60mM 5 - 5 5 - 5 20 - 40 5 - 3 5 - 3 2 - 2 20 - 20 5 - 1 5 - 1 3 - 0 40 - 20 5 - 1 5 - 1 5 - 0 60 - 20 5 - 1 5 - 1 5 - 0 sucrose + L-Phe 20mM -60mM 3 - 5 3 - 5 0 - 3 20 - 40 3 - 3 3 - 3 1 -2 20 - 20 3 - 1 3 - 1 2 - 0 40 - 20 5 - 1 5 - 1 3 - 0 60 - 20 5 - 1 5 - 1 5 - 0 * D-Phe possess some bitterness in the high concentration solutions. The bitterness of 40mM D-Phe solution equals to score 1. The bitterness of 60mM D-Phe solution equals to score 2. The results is obtained by considering the bitterness.

1 2 3 5 5

-

3 2 1 0* 0*

0 0 2 3 5

-

3 3 1 0 0

Other experiments were conducted to verify the conclusions drawn and described above. For example, benzoyl-e-aminoaproic acid (BACA), possesses a sour taste despite having a hydrophobic group in the molecule, i.e., it has an X-B component. If

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

34

FOOD FLAVOR AND

SAFETY

there is single receptor for sweet and bitter (and perhaps sour), then prebinding or preloading of the bitter receptor site of the receptor with BACA should change the final taste perception of subsequently added chemical since the newly added component cannot bind additional bitter molecules. In this experiment, two bitter tasting solutions caffeine (weak bitterness) and strychnine (strong bitterness) were given after BACA. Data in Table V show that the weak bitterness of caffeine is eliminated and the strong bitterness of strychnine is weakened if they are added after BACA administration thereby supporting the competitive receptor hypothesis.

Table V . Tasting Behavior of Combination of Typical Bitter Compounds and Benzoyl-e-aminocoproic acid bitter compound X-B taste** 0.005 mM strychnine + 5mM Bz-NH-(CH2)5-COOH bitter -> no 0.05mM strychnine(TV*0.003mM) + 5mM Bz-NH-(CH2)5-COOH bitter -> weak + 5mM Bz-NH-(CH2)5-COOH bitter -> no 2 mM caffeine + 5mM Bz-NH-(CH2)5-COOH bitter -> weak 5 mM caffeine(TV1.0 mM) SOURCE: Adapted from ref. 5 * Threshold value. ** Decrease of bitterness by X-B component

Hexamethylenetetramine and cyclohexylamine are both tasteless solutions. Hexamethylenetetramine, a highly water soluble compound, is tasteless even though it is composed of a AH-X component (Figure 1). Its molecular size is too small to exhibit the bitter taste proposed by the theory presented in Figure 1. Cyclohexamine, similarly, is tasteless. When both solutions are mixed together, a bitter flavor is exhibited. Interestingly, the bitter flavor of the mixed components is changed to sweet when acetic acid is added (5), a fact consistent with the competitive theory. A New Receptor Model The data described above led to the prediction of a single taste receptor that recognizes both sweetness and bitterness and to the development of a new molecular model of a taste receptor (Figure 4). According to this model, bitterness is produced when a hydrophobic group labeled "X" and electropositive or a hydrophobic group labeled "AX" attach at the corresponding receptor sites. Sweetness is produced when an electronegative group labeled "B" attaches at the remaining receptor site (Figure 4). This model readily explains all of the experimental data obtained thus far. For example, benzoyl-e-aminocaproic acid (BACA), describe above, produced only a sour taste in spite of having a hydrophobic group in the molecule. According to the model (Figure 4) BACA could not produce a sweet or bitter taste since it lacks an AX component. The new model describes the bifunctional unit of bitter reception as the "hydrophobic group" and the "electropositive of hydrophobic group" whereas the 1985 model of bitter (6-12) described the bifunctional units of bitterness production as the "binding unit" and the "stimulating unit." The new model in Figure 4 describes the perception of sweetness as a AH-B-X response rather than an AH-B response as the proposed by the theory of Shallenberger (7). The new model offers a molecular approach more consistent with the experimental data. Further investigations to develop the theory and widely apply it to various flavor compounds are currently in progress and should prove to be both exciting to the scientist and useful to producers of foods.

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

3.

Recognizing Sweetness and Bitterness

NAKAMURA AND OKAI

SWEET

35

BITTER X ; hydrophobic group Β ; electronegative group A X ; hydrophobic group or electropositive group

Fig. 4. A New Model of Taste Receptor Literature Cited (1) Shallenberger, R.S.; Acree, T. E. Nature (London) 1967, 216, 480. (2) Shallenberger, R.S.; Acree, T. E.; Lee, C. Y. Nature (London) 1969, 221, 555. (3) Mazur, R.; Schlatter. J. M.; Goldkamp. A. H. J. Am. Chem. 1969, 91, 2684. (4) Fujino, M.; Wakimasu, M.; Mano, M.; Tanaka, K.; Nakajima, N.; Aoki, H. Chem. Pharm. Bull. 1976, 24, 2112. (5) Shinoda, I.; Okai, H. J. Agric. Food Chem. 1985, 33, 792-795. (6) Ishibashi, N.; Arita, Y.; Kanehisa, H.; Kouge, K.; Okai, H.; Fukui, S. Agric. Biol. Chem. 1987, 51, 2389. (7) Ishibashi, N.; Sadamori, K.; Yamamoto, O.; Kanehisa, H.; Kouge, K.; Kikuchi, E.; Okai, H.; Fukui, S. Agric.Biol.Chem. 1987, 51, 3309. (8) Ishibashi, Ν.; Ono, I.; Kato, K.; Shigenaga, T.; Shinoda, I.; Okai, H.; Fukui, S. Agric. Biol. Chem. 1988, 52, 91. (9) Ishibashi, N.; Kubo, T.; Chino, M.; Fukui, S.; Shinoda, I.; Kikuchi, E.; Okai, H. Agric. Biol. Chem. 1988, 52, 95. (10) Ishibashi, N.; Kouge, K.; Shinoda, I.; Kanehisa, H.; Okai, H. Agric. Biol. Chem. 1988, 52, 819. (11) Oyama, S.; Ishibashi, N.; Tamura, M.; Nishizaki, H.; Okai, H. Agric. Biol. Chem. 1988, 52, 871. (12) Shinoda, I.; Nosho, Y.; Kouge, K.; Ishibashi, N.; Okai, H.; Tatumi, K.; Kikuchi, E. Agric.Biol.Chem. 1987, 51, 2103. (13) Ishibashi, N. In The dissertation Pepuchido ni okeru nigami hatugen kikou. Hirosima University (1988). Received February 16, 1993

Spanier et al.; Food Flavor and Safety ACS Symposium Series; American Chemical Society: Washington, DC, 1993.