Sweetners State of Knowledge Review

Neurosctence and Baobehavloral Reviews, Vol. 17, pp. 313-345, 1993 0149-7634/93 $6 00 + .00 Printed m the U.S A All rights reserved. Copyr,ght © 1993 Pergamon Press Ltd

Sweeteners: State of Knowledge Review

S. S. S C H I F F M A N j A N D C. A. G A T L I N

Departments o f Psychology and Psychiatry, Duke University, Durham, N C 27706

(Received 15 February 1993)

SCHIFFMAN, S. S. AND C. A. GATLIN. Sweeteners: State of knowledge review. NEUROSCI BIOBEHAV REV 17(3) 313-345, 1993.-Sweeteners are widely used in the food and pharmaceutical industry. The purpose of this paper is to review our current knowledge of sweet taste from chemical, biochemical, electrophysiolog~cal, psychophysical, and psychological points of view. The most common sweeteners likely to be used in food and pharmaceuttcals will be examined in detail. First, the chemtcal structures of sweet compounds including saccharides, diterpene glycosides, polyols, amino acids, dipept~des, and other nonsugars will be discussed. Second, biochemical approaches to understanding sweetener receptors will be reviewed. Third, electrophysiological and behavioral approaches to understanding sweetener receptors will be discussed. Fourth, psychophysical studies in humans will be shown to be consistent w~th biochemical and neurophysiolog~cal data. In addition, the basic mechamsms of sweet taste revealed by psychophysical studies will be given, including the role of multiple receptor s~tes, hydrogen bonding, and sodium transport. Finally, the factors that affect preference for sweet taste including the psychological and physiological variables associated with sweet preference will be explored.

Sweetness Taste Biochemistry Neurophysiology Psychophysics

IN the natural environment, sweet taste is a cue for edibility (185). For this reason, there is an innate tendency in many animal species including humans, to prefer sweet taste. The purpose of this paper is to review our current knowledge of sweet taste from chemical, biochemical, electrophysiological, psychophysical, and psychological points of view. Detailed descriptions of those sweeteners most likely to be used in the food and pharmaceutical industry will also be given.

CHEMICAL STRUCTURE OF SWEETENERS

The chemical structures of molecules that confer a sweet taste are wide and varied, encompassing both sugars and nonsugars (50,52,116,205,211,244). Table 1 displays a repre- sentative sampling of compounds that taste sweet to humans and gives their chemical structures along with other perti- nent information. These compounds include saccharides, diterpene glycosides, polyols, proteins, dipeptides, and other nonsugars.

Our understanding of the chemical properties of these compounds that are responsible for sweet taste is limited. One frequently cited hypothesis is that an essential condition for sweet sensation is a pair of s!multaneous hydrogen bonds separated by approximately 3 A(211,212). This theory depends upon a putative sweet receptor (or receptors) with an AH-B structure, where A and B are electronegative atoms separated by 2.5 to 4.0 .~, and H is a hydrogen atom and part of a polarized system A-H. The assumption is that a complemen- tary AH-B group in the stimulus molecule interacts with the AH-B receptor site, thus forming two simultaneous hydrogen

bonds. For potent sweetness, a third lipophilic site has also been suggested (58,125).

The proposition by Shallenberger and Acree (211) that intermolecular hydrogen bonding is an essential condition for sweet taste is derived from Shallenberger's earlier (210) studies demonstrating that sweetness potency is correlated to the in- verse of the strength of intramolecular hydrogen bonding. Unfortunately, while the AH-B theory has been helpful in accounting post facto for some functions of the sweet molecule, it has resulted in minimal predictive use for development of new compounds with sweet taste. Despite the presence of an AH-B system, a compound will not necessarily taste sweet. Moreover, the universal application of Shallenberger and Acree's theory has been called into question by Jakinovich (117) in studies indicating that an AH-B system may not be required to produce a sweet taste. Although most studies do suggest that hydrogen bonding contributes to the binding of most sweeteners to their receptors, the more specific AH-B hypothesis has probably outlived its usefulness.

Due to the fact that the relationship between chemical structure of molecules and the ability to initiate a sweet sensation is not well understood, it is not surprising that most compounds with sweet taste have been discovered by trial and error. The sweetness of sugars is age-old knowledge, but discoveries of the sweet taste of many of the nonsugars such as saccharin, cyclamate, acesulfame-K, and aspartame have been accidental, some quite recent. Remsen and Fahlberg discovered the sweetness of saccharin in 1879 (187) during investigations of the oxidation of o-toluene-sulfonamides. Similarly, while studying the antipyretic properties of a sulfamic acid

To whom requests for reprints should be addressed.

313

314 SCHIFFMAN AND GATLIN

derivative, Sveda accidently discovered the sweet taste of cyclamate in 1937 (8). Clauss and Jensen's recognition of a sweet taste in a product obtained by reacting butyne with fluorosulfonyl isocyanate (44) resulted in the development of acesulfame-K (formerly called acetosulfam). Tracing the sweetness to 5,6-dimethyl dihydrooxathiazinone dioxide, Clanss and Jensen (44) discovered that it possessed a ring system not pre- viously synthesized. Further studies of structure activity with dihydrooxathiazinone dioxides disclosed that the compound with the best sweet taste was the potassium salt of 6-methyl- 3,4-dihydro- 1,2,3-oxathiazin-4-one-2,2-dioxide (acesuifame- K). In 1965, the dipeptide L-aspartyl-L-phenylaianine methyl ester (aspartame) was accidentally discovered to have sweet- taste properties by J. Schlatter of G. D. Searle while synthesizing gastrin and its tetrapeptide analogs (148,149). Table 1 shows that three of these four sweeteners, acesulfame-K, saccharin, and cyclamate contain a sulfonamide group and a lipophilic part. The dipeptide aspartame, however, is chemically unrelated to the other three.

Other sweeteners have been isolated or synthesized from natural products. For example, stevioside and rebaudioside are derived from the foliage of Stevia rebaudiana Bert., a diminutive shrub which grows in Paraguay. Neohesperidin dihydrochalcone is a product of alkaline hydrogenation of bitter-tasting neohesperidin, a citrus flavanoid. In the 1980's a sesquiterpene named hernandulcin was isolated from the Mexican plant Lippia dulcis Trev (47). Monellin is a protein made up of two dissimilar polypeptide chains found in the red berries of the tropical creeper Dioscoreophyllum cumminsu Diels, native to West Africa. Thaumatin I and II, both sweet- tasting proteins, have been isolated from the fruit of the Afri- can plant Thaumatoccus damelli Benth.

Despite the number of reportedly sweet compounds, detailed investigation of their structure-activity relationships has been limited. This may be due to the considerable expense of synthesizing very complex compounds only to taste them. A remarkable exception is the extensive synthesis of thousands of analogs for L-aspartyl-L-phenylalanine methyl ester (aspartame, APM) after its discovery.

A brief description of structure-activity studies for several sweeteners and their effects on taste follows.

Aspartame

When modifications of aspartame were investigated, a consistent simple pattern was clear: c~-amides of L-aspartic acid (isoasparagine derivatives) are often sweet (150,152). This fun- damental finding suggests that the sweetness quality and sweetness potency are mediated primarily by the steric properties of the amide part of the molecule. Accordingly, L-phenylaianine methyl ester is a complex amine.

Viewed both theoretically and practically, analogs similar in potency to aspartame are not as interesting as analogs with more sweetness potency. Pursuant to the finding that L- phenylalanine could be replaced by D-alanine and other ali- phatic D-amino acids in aspartame (151), a group at Pfizer synthesized highly branched amides of L-aspartyl-D-alanine and L-aspartyl-D-serine (28). The superior compounds ranged in potencies 1000-2000 times greater than sucrose and exhibited highly satisfactory taste profiles. The corresponding simple amides average around 100 times sweeter than sucrose (218). The structure of the most acceptable L-aspartyl-D- alanme analog is called alitame, and its structure is gwen in Table 1.

The Takeda laboratories made the remarkable discovery

that replacing phenylalanine with certain aminomalonic acid diesters results in exceptionally potent sweet compounds (80). A large branched group is required for the highest potencies, similar to the Pfizer compounds. In Shanghai, investigators at the Institute of Organic Chemistry synthesized all four compounds derived from the four possible isomers of the terpene alcohol, fenchol. According to this group, the most potent was 50,000× sucrose (139). Molecular modeling explains (after the fact) how such similar substances manifest sweetness potencies of 1, 10, 100, 1000, or 10,000.

The early studies on aspartame led to the conclusion that a free amino group on aspartic acid was required for sweetness, but studies from the Wyeth laboratories and the chemists at the Universit6 Claude Bernard have refuted this theory (201). A possible explanation may be that aspartame receptors are not the only receptors involved when an N-acylated aspartic acid derivative is perceived as sweet. Consideration of multiple receptors will be taken up later in this review.

One group reported sweetness of trifluoro- and trichloro- acetylaspartic acid p-cyanoanilide as 3000 times more potent than sucrose (134). Another found that the glutamic acid analog is 3000 times more potent than sucrose (122). Numerous sweet glutamic acid derivatives are described as sweet.

A major breakthrough in developing an aspartame receptor model occurred in 1979 at Universite6 Claude Bernard. Jean-Marie Tinti, a graduate student with Professor Claude Nofre, found that a structural hybrid of aspartame and the sweetener known as suosan (177) is approximately 50 times as potent as aspartame and 20 times as potent as suosan (170, 227). Molecular modeling studies based on this finding led to the discovery of the guanidine structural class of sweeteners which presumably act at a common site. One sweetener from this group, sucrononic acid, has a potency of 200,000 times that of a 2% sucrose reference (171). These discoveries led to the development of a model for the aspartame receptor (53) by NutraSweet scientists which has been used to design a new series of aryl urea high-potency sweeteners (164).

Saccharin and Acesulfame-K

There has been extensive research on structural analogs of saccharin (91,105) including acesulfame-K (44), which is really a structural analog of saccharin. Saccharin-like structures have an underlying bitter taste which is somewhat masked by a sweet component, according to one report (91).

Dihydrochalcones

Modification studies of dihydrochalcone analogs in the carbohydrate (68) and the aglycone portion (63) have resulted in several compounds with increased potency but with no dis- cernible improvement in temporal effects. This group of sweeteners are limited in usefulness due to their persistent taste. Because of this, it has not been particularly useful to develop them commercially.

Other Sweeteners and Limitations of Models

Other structure-activity work has been done with sucrose, sulfamate sweeteners, and amino acids. The potency of sucrose can be increased by replacing selected atoms with chlorine (112,115) as exemplified by sucralose in Table 1. Struc- ture-activity studies show delineated size and shape limitations for sulfamate sweeteners (214). A proposed explanation for the difference in taste between amino acid enantiomers is that nonsweet amino acids are prevented from binding by steric hindrance, in this case, a spatial barrier (211).

SWEETENERS: STATE OF KNOWLEDGE REVIEW 315

Finally, even though analogs of the parent compounds have been synthesized and a number of models have been proposed (5, 53,108, 113,122, 136,150, 186,220, 222,226,232, 233), it is questionable whether a single model can account for the stereochemical properties of all or most compounds that confer a sweet taste sensation. This deficiency in predictive value for any geometric model in determining a priori the components of molecules that render a sweet sensation indicates the likelihood of multiple types of receptors, each with unique stereochemical and physicochemical requirements.

BIOCHEMICAL APPROACHES TO UNDERSTANDING SWEET RECEPTORS

The search for taste receptors has not been as successful as the pursuit of neurotransmitter and hormone receptors. Isolation of taste receptors is limited by the low affinity of tastants for receptors as well as the physiology of the gustatory system. Although hormones show Kos in the range of 10-~t M-10 -8 M, the apparent Kos for sweeteners such as sugars are as high as 0.5 M. Moreover, it is difficult to separate mammalian taste buds containing gustatory receptors from the surrounding epithelium. Attempts to understand the nature of sweet receptors continue undeterred by the intrinsic difficulties in methodology. Most biochemical studies thus far have been based on the assumption that sweet-tasting molecules bind to receptor proteins in a reversible manner. The first studies obtained a protein fraction from homogenates of bovine tongue epithelium. In the presence of sugars, saccharin, and sweet-tasting amino acids, these fractions exhibited either refractive index changes or ultraviolet difference spectral changes (54,55,180,181). Preparations from rat and monkey tongues have also exhibited spectral changes in the presence of sweeteners (101-103). Rationale for the hypothesis that this protein is indeed a sweet receptor protein has been offered by Price and DeSimone (181) and Shimazaki et al. (213). Their arguments are:

1. The concentrations of sweeteners that interact with protein fractions are similar to those concentrations found to be effective in behavioral and neural studies.

2. Sweet-tasting amino acids are more interactive with the protein fraction than bitter-tasting ones.

3. In rats, the interaction of sugars with the protein fractions is independent of pH within the range of 5.4-8.6, which is consistent with electrophysiological taste data reported by Noma and Hiji (172).

4. The quantity of sweet-sensitive protein in the homogenate is significantly lower when taste bud turnover is reduced by colchicine or when the taste buds are denervated (101,103).

Nevertheless, uncertainties persist, and the conclusion that the sweet-sensitive protein is a sweetener receptor is not yet fully substantiated. One difficulty is that the protein fraction is not homogeneous and has been shown to contain sugar- metabolizing enzymes as contaminants (182). A second prob- lem is that the protein fraction has been shown to complex with molecules that are not sweet such as dimethyl sulfoxide (169). The question is compounded by the fact that a nongus- tatory protein, hemoglobin, exhibits optical properties similar to those of the protein fraction when complexed with sugar (169), and further obscured by the fact that if the protein were localized strictly in the taste cells that constitute a relatively small portion of the epithelium, the yields of sweet-sensitive protein from the lingual epithelium would be expected to be considerably lower than they are (182). Lastly, Ostretsova et

al. (176) could not demonstrate significant spectral changes in tongue protein from bovine fungiform papillae, the structures in the epithelium actually containing the taste buds. On the other hand, utilizing equilibrium dialysis in the sedimentable fraction instead of the protein fraction reported by Dastoli and Price (55), they were able to demonstrate binding activity.

Rather than using whole lingual epithelium as described above, further investigations have been carried out to study binding of sweeteners both to taste papillae or isolated lingual membranes. Cagan (38) demonstrated that t4C-labeled sugars bound two or three times more to bovine fungiform and circumvallate papillae that contain taste buds than to filiform papillae which lack buds. However, the binding proved weak (KD~ 10- ' -10 -3 M), which is consistent with the neurophysiological data described in the next section. Other binding studies with bovine papillae (140) reported the same conclusion. Cagan and Morris (39) found that binding of the potently sweet protein monellin to human tongue tissues had a higher affinity (K D ~ 10 -5 M). Another study (176) reported ['4C]glucose binding to plasma membrane fragments from bovine circumvallate and fungiform papillae using equilibrium dialysis, and Lum and Henkin (144) concluded that binding of sugars to fractions probably derived from plasma membranes of bovine taste buds.

There have been attempts to understand sweetener receptors using immunological approaches. Hough and Edwardson (110) have raised antibodies against the sweet-tasting plant protein thaumatin in New Zealand white rabbits and found that a variety of other sweet compounds varying widely in molecular size and strength cross-reacted with those antibodies. The compounds included aspartame, calcium cyclamate, sucrose, and monellin. It is interesting to note that there was a high degree of correlation between the immunoreactivity and sweetness potencies of those substances that cross-reacted with the antisera. From this the authors inferred that the structural feature that constitutes the antigenic determinant may be similar to this sweetener taste receptor.

Another approach to understanding taste receptors uses as a model system the channel catfish Ictalurus punctatus. Cat- fish barbels contain large numbers of taste buds, and their binding affinities for ligands, especially amino acids, are usually higher than in mammals. Binding of L-alanine (which tastes sweet to humans) in the sedimentable fraction from catfish barbel has been demonstrated by Krueger and Cagan (131). Goldstein and Cagan (87) reported production of mouse hybridomas that synthesize monoclonal antibodies against L- [3H]alanine-binding activity in a membrane fraction (fraction P2) in channel catfish. Over time, use of monoclonal antibodies to characterize taste receptor molecules may be helpful in yielding specific antagonists.

Although the isolation of sweetener receptors has proven difficult, current consensus is that the sweet taste response is indeed mediated by taste cell surface receptors that utilize the adenylate cyclase system as a second messenger system. The adenylate cyclase system, which is also the cellular signalling system for many hormones, involves the following cascade of events. The agonist (e.g., sweetener molecule) binds to a receptor which transmits a signal via the guanine nucleotide- binding protein (G protein) resulting in activation of adenylate cyclase. Adenylate cyclase then induces hydrolysis of ATP to cAMP which leads to activation of the phosphorylating enzyme known as protein kinase A. The activated kinase then phosphorylates an ion channel m the taste cell membrane lead- ing to depolarization of the taste cell. Evidence that this cascade is involved in sweet taste transduction comes from the


fact that cAMP is elevated in taste cells from rat, cow, and pig after exposure to sucrose or saccharin (217). In addition, saccharin blocks outward potassium currents in hamster taste cells, and this effect is mimicked by perfusion of these cells with cAMP (9).

ELECTROPHYSIOLOGICAL AND BEHAVIORAL APPROACHES IN ANIMALS

The findings just described are consistent with electrophysiological and behavioral data in animals. For example, neurophysiological data indicate that the affinities of sugars for their receptors are weak (16,178). Electrical recordings from taste nerves also show that proteolytic enzymes selectively suppress the neural responses to sweet stimuli implying that the sweet receptor is, in fact, a protein. A report in 1975 (100) demonstrated that Pronase E (pH 7.0) and semialkaline prote- ase (pH 8.0) either eliminated or significantly suppressed the sweet responses of sucrose, glucose, fructose, sorbitol, saccharin, glycine, and DL-alanine but had no effect on salty, bitter, or sour stimuli.

There are other compounds which selectively modify taste by suppressing the responses of taste receptor cells to sweeteners. Examples are: extracts of the leaf of the Indian plant Gymnema sylvestre (3,59,70,95,96,248) and ziziphins, which are saponins found in the leaves of the Chinese jujuba tree Ziziphus jujuba (123,124). A sweet-inducing protein called miraculin has been isolated from berries of the West Africa shrub Synsepalum dulcificum (miracle fruit); it elicits neural responses to acids that resemble responses to sweeteners (29). In monkeys, behavioral tests after miraculin treatments show an increased intake of citric acid which suggests an improvement in taste possibly to sweet (93,94). The precise way in which these substances modify taste is not yet known.

Although extracts of Gymnema sylvestre and Ziziphus jujuba block responses to all sweeteners, suggesting a single sweetener receptor type, most evidence drawn from animal studies points to multiple types of sweet receptors. First, species differ widely in sweetener responses (166,196). For example rats, guinea pigs, and the new world monkey, Sagumus midas tamarin, do not respond to monellin and thaumatin, both potently sweet to humans; on the other hand, the green monkey, Cercopithecus aethiops, exhibits electrophysiological and behavioral responses to these sweeteners (30,84,94). Fur- ther, the gerbil demonstrates greater response to sweeteners than the rat, giving electrophysiological responses to 14 of 21 compounds known to taste sweet to man, although six of these compounds did not yield the expected behavioral responses to a sweetener (118). Extensive comparison of responses to amino acids in rats and man has been described in experiments by Pritchard and Scott (183,184), and both similarities and differences have been reported. Other behavioral studies show that dulcin, which is sweet to humans, is preferred by squirrel monkeys but rejected by rats, whereas saccharin is refused by squirrel monkeys and selected by rats (75b).

A second line of evidence for multiple types of sweetener receptors has been demonstrated by Faurion et al. (71) and Faurion and MacLeod (72). Recording responses from individual taste fibers of the chorda tympani nerve, they observed that among neurons the response spectra to a range of sweet stimuli was not consistent; that is, although a single neuron reacted strongly to one sweetener, less to a second, and only weakly to another, this pattern did not necessarily obtain in a different neuron.

A third basis for the probability of multiple receptor sites for sweetness, is the discovery that alloxan selectively de-

presses neural response to sugars but does not suppress neural activity to artificial sweeteners such as sodium saccharin (250). Although other methods for inhibiting neural responses to sweeteners including receptor photolabeling studies (64) have been found to block of all sweet responses, the alloxan selec- tivity suggests multiple sweetener receptor sites.

All of this evidence suggests that there are multiple types of sweetener receptors, possibly with multiple sensitivities tuned to respond to more than one sweetener. For example, if aspartame binds to several types of sweet receptors, a single geometric shape for an "aspartame receptor" may describe the interactions too restrictively. Extending from this rationale is the likely inference that each receptor type would have different physicochemical and stereochemical requirements.

PSYCHOPHYSICAL STUDIES IN HUMANS

Psychophysical studies in humans are consistent with biochemical, neurophysiological, and animal data. These studies further imply that the sweet taste itself has multidimensional characteristics transduced by multiple sweet receptor mechanisms.

Multiple Receptor Sites Mediate Sweetness

Human psychophysical data like animal studies imply that more than one receptor mechanism mediates sweetness. The first line of evidence derives from many experiments that point to nonhomogeneous variability among sweeteners in individual subjects for thresholds, intensity ratings, and the effect of Pronase E. In 1980, Faurion et al. (74) obtained threshold values of seven sweeteners from 98 subjects and found that the thresholds for sweeteners varied independently from subject to subject. This study disclosed that it was not possible to predict the threshold of one compound for a given subject using the knowledge of his/her threshold for another compound. The inference from these threshold data is that there are numerous receptor sites, varying in number from one individual to another. The same interindividual variability was found for intensity ratings at suprathreshold concentrations (73,74). The findings of Faurion and colleagues conform to those reported by Schiffman et al. (206). The relative reduction of sweetness by Pronase E in humans has been shown to have a similar effect in that the individual profiles of inhibition for different sweeteners do not covary (71,72). If there were only one receptor site type, there would be no possible explanation for these data.

A second line of evidence indicating a multiplicity of sweet receptor types has been derived from experimental data utilizing the method of cross-adaptation (197). Cross-adaptation experiments are based on the assumption that two tastants may share common receptor sites if prolonged exposure (adaptation) to one results in a decreased response to another 054). Conversely, if a diminished sensation to one tastant does not occur after adaptation to another, a possible implica- tion is that a second subtype of cells with different sweetener receptors is involved. In one protocol (197), subjects first tasted sweet test solution A and estimated its intensity. They then held a sweet-adapting solution B in their mouths until the sweet taste vanished (30-60 s), swirling it around to ensure complete adaptation. The subjects next retasted test solution A and reestimated its sweetness. Using this model, the degree to which the post adaptation intensity was greater than, equal to, or less than the preadaptation intensity could be delineated. The degree of cross adaptation varied depending on the sweeteners that were tested. For example, sodium saccharin


and acesulfame-K cross-adapt with one another (i.e., sodium saccharin is significantly reduced in intensity by adaptation to acesulfame-K and vice versa). D-tryptophan and aspartame also mutually cross-adapt unlike sodium saccharin and aspartame which do not cross-adapt. The results suggest that molecules with related pharmacophores cross-adapt most strongly. Acesulfame-K and Na saccharin, strongly mutually cross- adaptive, have equivalent RSO2NHCOR- pharmacophores (197), Aspartame and o-tryptophan may similarly be related through their COOH/NH2 moieties which may be isosterically arranged in space. Adaptation to xylitol and glucose reduces the intensity of all the sweeteners suggesting that these two sweeteners activate all subsets of sweet-sensitive receptor cells.

A third line of evidence for multiple sweet receptors comes from an experiment in which subjects were asked to discern similarity in quality of 17 sweeteners varying widely in chemical structure (205). Multidimensional scaling (MDS), a com- puter-based mathematical technique, was applied to the mea- surements of similarity to arrange the sweeteners in a spatial map. Sweeteners assessed as similar in taste were placed close to one another in the map, and dissimilar ones further from one another. A three-dimensional map was derived revealing that sweetness is a multidimensional concept with differing characteristic sweet qualities as well as different side tastes and temporal properties. The experimental subjects continued to report that the nature of sweet sensation among these sweeteners varied, and that "sweet" itself is not a unity or undiffer- entiated property. In addition, subjects also commented that the location of sweetness on the tongue also varied across sweeteners. This finding was also reported by van der Wel and Arvidson (235) who found that sweet-tasting proteins such as thaumatin produce more intense sweetness at the edges of the tongue while sucrose is more intense at the tip of the tongue. In addition, the sweetness of thaumatin and monellin develops more slowly and lasts longer than that for sucrose. These findings suggest that multiple receptors must exist to produce these differences in sensory characteristics.

A fourth line of evidence comes from the variability in dose-response curves for different sweeteners (65). Dose-response curves for a wide range of sweeteners were constructed from suprathreshold intensity judgments made by a trained taste panel. The panel assigned intensity values to a large number of concentrations of a given sweetener using six concentrations of sucrose (2070, 5°7o, 7.5%, 9070, 12°/0, and 1607o) as standards. The dose-response curves for sucrose, fructose, and sugar alcohols continued to increase in intensity beyond the equivalent sweetness of 1607o sucrose. The intensity of aspartame and alitame never increase beyond the sweetness of 15-16°70 sucrose even at maximum solubility. Thaumatin never became sweeter than a 907o sucrose equivalent.

A fifth line of evidence for multiple sweet receptors comes from "magnitude estimation" experiments that characterize dose-response relationships for suprathreshold sweeteners in both old and young subjects (202). Loss in perceived sweetness with age was not uniform across sweeteners; the greatest loss in growth of perceived intensity with concentration was for large molecules which have the greatest number of possible AH-B systems, such as thaumatin, rebaudioside, and neohesperidin dihydrochalcone. This finding suggests that receptors that are capable of concerted intermolecular hydrogen bonding may be more affected by aging than receptors that can bind only one AH-B type or pharmacophore. Another possibility is that there is selective loss of some subtypes of sweet- sensitive receptor ceils.

A sixth line of evidence comes from studies by Frankmann

and Green (78) who studied the effect of cooling the tongue on perceived intensity of taste. They found that cooling the tongue had differential effects on sweeteners. While glucose and fructose had temperature sensitivities similar to sucrose, saccharin did not.

Finally, studies of the effect of methyl xanthines on taste further emphasize the multiplicity of sweet taste receptor mechanisms. Schiffman et al. (200) found that caffeine en- hances the taste of some sweeteners including acesulfame-K, D-tryptophan, neohesperidin dihydrochalcone, sodium saccharin, stevioside, and thaumatin. However, it had no effect on other sweeteners including aspartame, calcium cyclamate, fructose, and sucrose. This work also suggests multiple transduction pathways in different cell subtypes.

Correlations of Psychophysical Data With Hydrogen Bonding

Because the presence of an AH-B group does not assure that a stimulus molecule will confer a sweet taste, it has been difficult to substantiate the AH-B hypothesis. Nevertheless, acetylation of amino groups, eliminating the possibility of forming two simultaneous hydrogen bonds, distinctly diminishes the sweetness of amino acids (204) as well as the proteins monellin and thaumatin (234). Moreover, correlations between the number of possible types of AH-B sites (or pharmacophores) and sweet taste cross-adaptation are illustrated above.

Schiffman et al. (202) determined taste detection and recognition thresholds for 11 sweeteners whose chemical structures differed widely. A strong correlation between the number of possible types of systems for hydrogen bonding (i.e., AH-B systems) and mean detection thresholds was found. Sweeteners with many possible types of AH-B systems (e.g., monellin) had the lowest thresholds whereas sweeteners with only one type have the highest. A similar correlation between number of AH-B groups and detection thresholds for elderly subjects was found, although the thresholds for older individuals were 2.72 times higher on average. While the recognition thresholds for young subjects were 3.54 times higher on average than the detection thresholds, they were found to follow the same pattern. Although the number of AH-B site types and thresholds were correlated, this does not prove that the AH-B system is the means by which sweetness is conferred.

Three-Dimensional Nature of Sweet Receptor Sites

Studies of taste characteristics of amino acids and dipeptides give some insight into the stereochemicai nature of sweet receptors (198,200,202). Both D and L forms of amino acids with short side chains taste sweet. However, as the side chain increases, it appears that steric hindrance occurs for the L forms preventing binding to sweet receptors.

The L- and D-enantiomers differ considerably in increase of perceived intensity with concentration. When the logs of the concentrations (C) of the amino acids were plotted against the logs of the magnitude estimates of the intensity of their sensations (S), a regression line could be fit to the data. In most cases, the slopes for the regression lines for D- and L- amino acids are different suggesting that enantiomers may not occupy the same receptor sites.

The three-dimensional nature of receptors is also implied by the fact that the taste of a dipeptide cannot be predicted from the constituent amino acids (200) and that the sequence of amino acids is critical in determining taste quality, For seven pairs of dipeptides containing the same residues but in reversed sequence, none had identical tastes. For example,


while glycyl-L-alanine is notably salty and bitter, L-alanylgly- cine has little taste. Furthermore, dipeptides with two identical constituent amino acids do not have the same taste as the amino acid itself. While glycine alone is sweet, glycylglycine tastes bitter.

Inhlbltton and Induction of Sweet Taste by Modtfiers

Studies show that compounds that inhibit neurophysiological and behavioral responses to sweeteners in animals also inhibit sweetness in humans. Pronase E inhibits the taste of numerous sweeteners in humans to degrees varying with the individual subjects (71,72). Ziziphins (90,123,157) and Gym- nema sylvestre (13,156) also suppress sweet tastes. Lately, a monosaccharide, methyl-4,6-dichloro-4,6-dideoxy-D-galactopyranoside, has been demonstrated to specifically inhibit sweet taste in both hamsters (115) and humans (207). Miracle fruit (Synsepalum dulcificum) or the salts of chlorogenic acid and cynarin found in artichoke (12) can elicit sweet taste.

Recently, two new sweet taste inhibitors have been described in human studies. Lindley (137) found that substituted phenoxyalkanoic acids are potent and immediate inhibitors of sweetness. Unlike most sweetness inhibitors, pretreatment of the tongue was not necessary to achieve the effect, and the inhibition was rapidly reversible. Another sweetness inhibitor derived by Muller et al. (164) from the synthetic sweetener suosan has similar properties.

The Role of a Sodium Transport Pathway

That an amiloride-sensitive transport mechanism is involved in the perception of sweetness in humans has been suggested by Schiffman et al. (203). When amiloride (N- amidino-3,5-diamino-6-chloropyrazine carboxamide), a potassium-sparing diuretic and a potent inhibitor of sodium transport in a wide variety of cellular and epithelial transport systems, was applied to the human tongue, it reduced the perceived taste intensity of Na+and Li + as well as sweeteners. Application of 5 x 10 -4 M amiloride lowered the perceived intensity of all the sweeteners tested; the range of percentage inhibition was from 80.80/0 for stevioside to 44.2°7o for fructose. Amiloride also diminished the tastes of amino acid compounds with a sweet or salty component. Amiloride had no effect on the bitter- or sour-tasting compounds. Tennissen (223) recently confirmed that amiloride blocked sweet taste in human subjects.

The speculation that an amiloride-sensltive sodium channel plays a role in sweet taste transduction is supported by studies of lingual epithelium isolated from dogs. DeSimone et al. (56) demonstrated that sodium ions are actively transported when dog tongue epithelium is mounted between symmetrical Krebs- Henseleit buffer solutions. When hyperosmotic NaCI was applied to the dorsal lingual surface, increased transepithelial currents and potentials resulted. Application of glucose and other sugars to the dorsal surface also gave a hyperosmotic response. The inward current induced by sugars was significantly blocked by amiloride.

These human psychophysical data and transepithelial potential differences from isolated dog tongue epithelium suggest, therefore, that a transduction step for sweet-tasting non- electrolytes may involve a transcellular ion flow via an amiloride-sensitive pathway. How this transcellular flow is related to the adenylate cyclase system (217) or outward potassium currents (9), which are both known to play a role in sweet taste transduction, is not presently known.

FACTORS THAT AFFECT PREFERENCE FOR SWEET TASTE

In humans and animals, preference for sweet taste is innate. Distinct individual differences in adults are present, however, attributable to early feeding experiences and genetic differences in the ability to taste PROP (6-n-propylthiouracil). Sweet taste preferences may also be influenced by physiological state (208).

Although there is a preference for sweet taste, a number of studies have demonstrated that appetite and food intake are not stimulated by sweet sensations. Studies that use high- potency sweeteners, such as aspartame, in place of sugar in foods have shown that subjects do not necessarily compensate for the caloric deficits created by the substitution of a high- potency sweetener for sugar. Thus use of artificial sweeteners may result in greater success at weight loss and long-term weight maintenance (208).

The purpose of this section is to explore the psychological and physiological variables associated with sweet preference. The contribution of sweetness to satiety will be examined as well.

Inborn Sweet Taste Preference in Humans and Animals

Humans are born with the capacity to distinguish between the taste of sugars and other qualities of taste (138). Neonates have been observed to ingest sugar water more vigorously (174) and longer (51) than they did plain water. Preference for sweet taste is also inferred from gustofacial responses (216), although these responses are open to interpretation (49).

Rat studies have also revealed an innate preference for the taste of sugars. Newborn rats display positive ingestive responses to oral infusions of sucrose solutions (1,89) and complex carbohydrates (236). Twelve-day-old rats also respond positively to Na saccharin (114,219).

Individual Differences for Preference for Intense Sweet Stimuli in Adult Humans

In adults, there is considerable individual variability in hedonic reactions to sweet stimuli (240). "Sweet likers" are peo- ple who evaluate highly sweet sucrose solutions positively. "Sweet dislikers" are those who respond to intense sucrose concentrations negatively. Sweet liker/disliker status is constant during fasted and fed states, although for sweet dislikers, the unpleasantness of strongly sweet solutions lessens during fasting (142). The sweet liker/disliker distinction is unchanged when fructose or glucose are used as sweeteners (141).

Early feeding experiences (14) or genetic differences in 6-n- propylthiouracil (PROP) taster status may account for individual differences in sweet preference. Sweet likers tend to be nontasters of PROP, while sweet dislikers tend to be PROP tasters (143). PROP is a compound similar to phenylthiocarbamide (PTC) which has a bimodal threshold distribution (243). Those who taste PROP detect its bitterness at low concentrations; nontasters detect the bitter taste only at high concentrations. Sweet likers experience a fairly complex taste sensation with sucrose solutions; sweet dislikers report a purely sweet gustatory experience (143).

Htgh-Potency Sweeteners: Effect on Appetite and Food Intake m Humans

Two kinds of investigations have been employed for evaluating the effect of high-potency sweeteners on appetite and food intake. The first, the "additive" approach, tests the effect


that perception of a sweet taste has on subsequent responses; responses to unsweetened foods are compared to responses to nutritionally identical sweet foods. Aspartame, which gives a sweet taste in calorically insignificant quantities, has been most frequently used in additive studies to provide sweetness without changing the nutritional value of the substance. For example, noncaloric soft drinks have been compared with mineral water (21), and a milkshake sweetened with aspartame has been compared with one not sweetened (27).

The second, the "substitutive" approach, substitutes a high- potency sweetener for a caloric sweetener in a food, producing a food of equal sweetness but significantly fewer calories. As the sensory properties of the meal are the same, the substitutive study reveals the effect of nonsensory variables, such as caloric density, on satiety.

Using these techniques, investigations usually have failed to find any difference in appetite ratings following the sweet and unsweet preloads (21a, 27,42,194). A few studies have found that a sweet load was, in fact, more satiating, as inferred from a greater decrease in subsequent hunger ratings when compared to an unsweetened, nutritionally identical meal (146,239).

There are three reports of enhancement of appetite following ingestion o f aspartame-sweetened liquid (2 lb, 22,191 ). Ap- petite ratings were obtained from subjects during the hour following ingestion of plain water, and the results were compared with appetite ratings after subjects drank aspartame- sweetened water. Ratings of hunger and of desire to eat were higher after ingestion of the aspartame-sweetened water. These two studies (22,191), however, used a concentration of aspartame (81 mg/100 ml) considered to be quite sweet, equal to 9.5% sucrose (65). In a subsequent report from the laboratory (192), a sweeter solution of aspartame, 117.5 mg/100 ml, produced no significant effect on appetite scores compared to plain water. In evaluating these studies, it is important to note that, although flavored or carbonated sweet beverages are experienced as pleasant, plain water is usually preferred over purely sweet water (142). The results of these studies are ap- parently confounded or at least inconclusive, because sweet aqueous solutions may have been unpleasant or less pleasant than plain water. Black et al. (21b) found an increase in appetite after aspartame-sweetened mineral water but only when 280 ml were consumed; no increase in appetite was found when 560 ml were consumed.

Even if these results are taken to indicate appetite stimulation by sweetness, it must be borne in mind that the higher appetite scores which occurred after consumption of aspartame-sweetened water were not followed by increased food intake. On the contrary, studies by Rogers et al. (191,192) have demonstrated a marginally lowered food intake after consumption of solutions of aspartame, consistent with Mattes' (147) observation that correlation between hunger ratings and food intake is poor. Many other investigations have demonstrated no appreciable difference in food intake after consumption of aspartame-sweetened or unsweetened beverages (20,21a,21b,27,42,194,189). From these, it appears that addition of sweetness to beverages does not affect subsequent food intake.

The findings of an investigation utilizing saccharin to sweeten yogurt (190) are at variance with the above aspartame studies. Following ingestion of yogurt sweetened with saccharin, subjects felt less satiated 30-60 min later than they did after ingesting equicaloric unsweetened yogurt. Moreover, their subsequent food intake increased after eating saccharin-

sweetened yogurt compared to the plain yogurt. Rogers and Biundell (190) account for the increase as due to the sweetness of saccharin, but this does not accord with their further findings in the study that glucose-sweetened yogurt and equicaloric, unsweetened yogurt had comparable effects on appetite scores and subsequent food intake.

Another study evaluating the effects of aspartame-sweetened yogurt preloads on subsequent intake (162), which did not include an unsweetened yogurt as a control, reported a nega- tive relationship between yogurt preload sweetness intensity and total 24-hour caloric intake.

Thus, the majority of studies do not support the contention that sweetness per se stimulates appetite and/or food intake. On the contrary, the few researchers who have shown an increase in subjective appetite scores following aspartame- sweetened preloads have found contradictory results in their own work. Further, consumption of aspartame-sweetened preloads have never been shown to result in increased food intake.

Using the "substitutive" protocol, foods of comparable sweetness but different in caloric value are evaluated for their effect on subsequent appetite and food intake. Gelatin desserts, puddings, and fruit-flavored drinks sweetened with aspartame have been evaluated with respect to their sugar- sweetened equivalents. Findings usually showed that ingestion of preloads with equal sweetness but differing caloric content resulted in similar effects on appetite ratings (2,42,193-195). It can be inferred then, at least initially, that bulk and volume of a substance have a greater effect on appetite than its caloric content. Also expectations about caloric content of a food influence appetite and satiety (23,224). These findings are not consonant with those of Blundell and Hill (22) or Rogers et al. (191) who found that appetite ratings rose significantly after ingestion of aspartame-sweetened water, but fell with glucose-sweetened water. No significant difference in subsequent food intake has been found after consumption of a low calorie preload compared to the high calorie preload (2,42, 191,194). Findings from an investigation by Rogers and Blun- dell (190) also showed that food intakes following preloads of yogurt containing 131 Cals sweetened with saccharin did not differ significantly from those after yogurt containing 295 Cals and sweetened with glucose.

These studies on short-term caloric compensation after preloads of equal sweetness but differing caloric content imply that caloric compensation is imperfect in the short term. It must be emphasized that these studies use an entirely different protocol from long-term studies of caloric compensation (43,76,77) and should not be equated.

A recent report by Drewnowski et al. (61) combined the additive and substitutive study designs. These studies utilized four kinds of preloads with a base of soft white cheese: (A) unsweetened, 300 Cais; (B) aspartame-sweetened, 300 Cals; (C) sugar-sweetened, 700 Cals; and (D) aspartame-sweetened, with maltodextrin, 700 Cals. Comparison of (A) vs. (B) re- flects the additive effect, and comparison of (B) vs. (C) re- flects the substitutive effect. Moreover, comparison of (C) vs. (D) enables assessment of the unique effect of aspartame on appetite and food intake under conditions where sweetness and calories are controlled. The studies revealed no effect of sweetness on subsequent intake: Food intake following (A) was similar to that following (B). Nor was there caloric compensation: Total daily caloric intake following (B) was approximately 400 Cals less than that after treatment (C), from which it can be inferred that the subjects did not perceive the


400 calorie difference in preload (B) and increase subsequent intake. Moreover, total caloric intake after treatments (C) and (D) were similar, further supporting the conclusion that aspartame does not stimulate food intake.

Effect o f Physiological State on Hedomc Response to Sweet Taste

The pleasantness of sweet taste has a relationship to nutritional state. During fasting, sucrose solutions are rated as more pleasant than after nutritional satiety (35,69). Investiga- tions using rodent models also reveal that sweet taste is more appealing to starved animals (17,36,37). "Alliesthesia," or shifts in the pleasantness of sweet taste as a function of nutritional state (34), may be a component of food intake regulation. When internal energy stores have been depleted and caloric intake is important, perception of sweetness is highly pleasant and increases the probability that a meal will be eaten. If the gut has adequate energy supplies and is not, therefore, in need of calories, the attractiveness of sweetness lessens, thereby decreasing the probability of feeding.

Additional evidence for an integral relationship between gustation and nutritional state are found in electrophysiological studies. When internal metabolic conditions mimic reple- tion, neural activity elicited by sugar on the tongue decreases. Exogenous administration of glucose (82) and insulin (83) lowered neural activity evoked by the sweet taste of glucose, but had negligible effect on responses to hydrochloric acid's sour taste or quinine hydrochloride's bitterness. Similarly, stomach distention without nutrient ingestion lowered neural responses to sweet, but not to bitter taste (85).

Other investigations lend support to the hypothesis that perceptual changes in the pleasantness of sweetness reflect function. During the luteal phase of the menstrual cycle, sweetness is experienced as more pleasant and intake of sweet foods increases (26). Because metabolic rate rises during the luteal phase (241), there is an increased palatability and intake of sweet foods then would be compensatory for increased energy expenditure. It has also been demonstrated that children ranging from 9 to 15 yr of age preferred more intense sweet taste than did adults (57). Because the adolescent generally has high activity levels and therefore high caloric needs, elevated preference for sweetness may be due to an adaptive response to physiological need.

Preference for Sweet Taste in Restrained Eaters and Obese Persons

"Restrained eaters," a term for those who diet chronically and preoccupy themselves with food and body weight (98), characteristically ingest fewer calories per day than unrestrained eaters (133). However, such caloric restraint is followed by episodic consumption of large quantities of calories, and these calories are often obtained by hinging on sweet foods rather than nonsweet foods (229,238). It is possible that

sweet foods compensate for a prior dearth of orosensory stimulation.

Orosensory stimulation contributes to satiety (4,15,119, 221,246). In other words, a meal unaccompanied by orosensory stimulation is not highly satiating. However, dieters will characteristically select foods viewed as "dietetically correct," such as cottage cheese, salad, plain poultry, and fruit, but these foods are bland and unsatisfying from a sensory stand- point. Habitual dieters limited to these foods experience a deficiency in orosensory stimulation as they limit their caloric intake. Following this rationale it can be inferred that the reason sweet foods are so attractive to restrained eaters is their potent sensory impact and palatability, which ameliorates the dieters' self-imposed deficit in orosensory satiety.

Addition of aspartame to a low-calorie diet can improve adherence to a dietary regimen (120). Obese subjects were given a low caloric diet for 12 wk: Half were advised to include aspartame-sweetened products in the diet, while the rest were instructed to avoid food containing aspartame. The central finding was that the female subjects supplementing their diets with aspartame-sweetened foods lost more weight than those adhering to a diet nutritionally identical but with no aspartame-sweetened foods. The relationship of aspartame usage to greater weight loss indicates better compliance with the low calorie dietary guidelines. Free access, then, to low calorie or calorie-free sweet foods and beverages seems to result in greater adherence to a diet restricted in calories. On follow-up after one year of long-term weight maintenance, women who used aspartame-sweetened products maintained their weight loss significantly better than the women in the control group (121).

Addition of low caloric sweetness to a weight reduction diet may have an additional benefit. Increased metabolic rate is associated with sweetness (135). For human subjects, tasting a sweet food but not swallowing it resulted in increased energy expenditure that was not caused by oromotor activity.

CONCLUSIONS

Sweeteners vary widely in the chemical structures and include saccharides, diterpene glycosides, polyols, amino acids, dipeptides, and other nonsugars. An overview of biochemical, electrophysiological, behavioral, and psychophysical studies suggest that there are multiple sweetener receptor types. The transduction mechanism for sweetness involves sodium transport as well as activation of the adenylate cyclase system. Although there is an innate preference for sweet taste, intake of high-potency sweeteners does not lead to increased food intake.

ACKNOWLEDGEMENTS

The authors would like to thank Drs. H. H. Butchko, G. E. Du- Bois, B. E. Homler, F. N. Kotsonis, and W. Stargel for their comments on the final manuscript.

REFERENCES

1. Ackroff, K.; Vigorito, M.; Sclafam, A. Fat appetite in rats: The response of infant and adult rats to nutritive and non nutritive oil emulsions. Appetite 15:171-188; 1990.

2. Anderson, G. H.; Saravls, S.; Schacher, R.; Zlotkin, S.; Leiter, L. A. Aspartame: Effect on lunchtime food intake, appetite and hedonic response m children. Appetite 13:93-103; 1989.

3. Andersson, B.; Landgren, S.; Olsson, L.; Zotterman, Y. The

sweet taste fibres of the dog. Acta Physiol. Scand. 21:105-119; 1950.

4. Antin, J.; Gibbs, J.; Smith, G. P. Intestinal satiety reqmres pregastric food stimulation. Physiol. Behav. 18:421-425; 1977.

5. Ariyoshi, Y. The structure-taste relationships of aspartyl dipeptide esters. Agnc. Biol. Chem. 40:983-992; 1976.

6. Arnold, D. L.; Moodie, C. A.; Grice, H. C.; Charbonneau, S.

S W E E T E N E R S : S T A T E O F K N O W L E D G E R E V I E W 321

M.; Stavric, B.; Colhns, B. T.; McGuire, P. F.; Zawidzka, Z. Z.; Munro, I. C. Long-term toxicity of ortho-toluenesulfona- mide and sodium saccharin the rat. Toxic. Appl. Pharmacol. 52:113-152; 1980.

7. Askar, A. A rewew on sweeteners. Ernahrung/Nutrition. vol. 12. 1988:706-712.

8. Audrieth, L. F.; Sveda, M. Preparation and properties of some N-substituted sulfamic acids. J. Org. Chem. 9:89-101; 1944.

9. Avenet, P.; Kinnamon, S.; Roper, S. In situ recording from hamster taste cells: Responses to salt, sweet and sour. Chem. Senses 16:498; 1991.

10. Baer, A. Sugar alcohols in the diabetic diet. In: Kretchmer, N.; Hollenbeck, C., eds. Sugars and sweeteners. Boca Raton, FL: CRC Press; 1991:131-150.

11. Bar, A. Xyhtol. In: Nabors, L. O.; Gelardi, R. C., eds. Alter- native sweeteners. 2nd ed. New York: Marcel Dekker; 1991: 349-379.

12. Bartoshuk, L. M. Separate worlds of taste. Psychol. Today 14: 48-63; 1980.

13. Bartoshuk, L. M.; Dateo, G. P.; Vandenbelt, D. J.; Buttnck, R. L.; Long, L., Jr. Effects of Gymnema sylvestre and Synsepalum dulclfacum on taste in man. In: Pfaffmann, C., ed. Olfaction and taste III. New York: Rockefeller Univ. Press; 1969:436-44.

14. Beauchamp, G. K.; Moran, M. Dietary experience and sweet taste preference in human infants. Appetite 3:139-152; 1982.

15. Bedard, M.; Weingarten, H. P. Postabsorptive glucose decreases excitatory effects of taste on ingestion. Am. J. Physiol. 256:R1142-R1147; 1989.

16. Beidler, L. M. Taste receptor stimulation. Prog. Biophys. Bio- phys. Chem. 12:109-151; 1962.

17. Berndge, K. C. Modulation of taste affect by hunger, caloric satiety, and sensory-specific satiety in the rat. Appetite 16:103- 120; 1991.

18. Bertorelh, A. M.; Czarnowski-Hill, J. V. Review of present and future use of nonnutritive sweeteners. Diabetes Educator 16: 415-420; 1990.

19. Billaux, M. S.; Fluorie, B.; Jacquemin, C.; Messing, B. Sugar alcohols. In: Marie, S.; Piggott, J. R. Handbook of sweeteners. Glasgow: Blackie; 1991:72-103.

20. Birch, L. L.; McPhee, L.; Sullivan, S. Children's food retake following drinks sweetened with sucrose or aspartame: Time course effects. Physiol. Behav. 45:387-395; 1989.

21a. Black, R. M.; Tanaka, P. A.; Leiter, L. A.; Anderson, G. H. Soft drinks with aspartame: Effect on subjective hunger, food selection, and food intake of young adult males. Physiol. Behav. 49:803-810; 1991.

2lb. Black, R. M.; Leiter, L. A.; Anderson, G. H. Consuming aspartame with and without taste: Differential effects on appetite and food intake of young adult males. Physiol. Behav. 53:459-466; 1993.

22. Blundell, J. E.; Hill, A. J. Paradoxical effects of an intense sweetener (aspartame) on appetite. Lancet I:1092-1093; 1986.

23. Booth, D. A.; Lee, M.; McAleavey, C. Acquired sensory control of satiation in man. Br. J. Psychol. 67:137-147; 1976.

24. Bopp, B. A ; Price, P. Cyclamates. In: Nabors, L. O.; Gelardi, R. C., ¢ds. Alternative sweeteners, 2nd ed. New York: Marcel Dekker; 1991:71-95.

25. Bopp, B. A.; Sonders, R. C.; Kesterson, J. W. Toxicological aspects of cyclamate and cyclohexylamine. CRC Crit. Rev. Tox- icol. 16:213-306; 1986.

26. Bowen D. J.; Grunberg, N. E. Variations in food preference and consumption across the menstrual cycle. Physiol. Behav. 47:287-291; 1990.

27. Brala, P. M.; Hagen, R. L. Effects of sweetness perception and caloric value of a preload on short term intake. Physiol. Behav. 30:1-9; 1983.

28. Brennan, T. M.; Hendnck, M. E. Branched amides of L- aspartyl-D-amino acid dipeptides. U. S. Patents 4,399,163 and 4,411,925. 1983.

29. Brouwer, J. N.; Glaser, D.; Hard, A. F.; Segerstad, C.; Helle- kant, G.; Ninomiya, Y; van der Wel, H. The sweetness-inducting

effect of miraculin; behavioral and neurophysiological experiments in the rhesus monkey Macaca mulatta. J. Physiol. (Lon- don) 337:221-240; 1983.

30. Brouwer, J. N.; Hellekant, G.; Kasahara, Y.; van der Wel, H.; Zotterman, Y. Electrophysiological study of the gustatory effects of the sweet proteins monellin and thaumatin in monkey, guinea pig, and rat. Acta Physiol. Scand. 89:550-557; 1973.

31. Budavari, S., ed. Merck index, l l t h ed. Rahway, N J: Merck & Co.; 1989.

32. Butchko, H. H.; Kotsonis, F. N. Acceptable dally intake vs. actual intake: The aspartame example. J. Am. College Nutr. vol. 10. 3:258-266; 1991.

33. Butchko, H. H.; Kotsonis, F. N. Aspartame: Review of recent research. Comments Toxicology 3:253-278; 1989.

34. Cabanac, M. Physiological role of pleasure. Science 173:1103- 1107; 1971.

35. Cabanac, M.; Duclaux, R. Specificity of internal signals in producing satiety for taste stimuli. Nature 227:966-967; 1970.

36. Cabanac, M.; Lafrance, L. Ingestive/aversive response of rats to sweet stimuli. Influences of glucose, oil, and casein hydrolyzate gastric loads. Physiol. Behav. 51:139-143; 1992.

37. Cabanac, M.; Lafrance, L. Postingestive alliesthesia: The rat tells the same story. Physiol. Behav. 47:539-543; 1990.

38. Cagan, R. H. Biochemical studies of taste sensation. I. Binding of ~4C-labeled sugars to bovine taste papillae. Biochim. Biophys. Acta 252:199-206; 1972.

39. Cagan, R. H.; Morris, R. W. Biochemical studies of taste sensation: Binding to taste tissue of 3H-labeled monellin, a sweet- tasting protein. Proc. Natl. Acad. Sci. USA 76:1692-1696; 1979.

40. Calorie Control Council. Acesulfame-K backgrounder. Suite 500-E. 5775 Peachtree-Dunwoody Road, Atlanta, Ga. 30342.

41. Calorie Control Council. Acesulfame-K. Sweetener fact sheet. Suite 500-E. 5775 Peachtree-Dunwoody Road, Atlanta, Ga. 30342.

42. Canty, D. J.; Chan, M. M. Effects of consumption of caloric vs. noncaloric sweet drinks on radices of hunger and food consumption in normal adults. Am. J. Clin. Nutr. 53:1159-I164; 1991.

43. Caputo, F. A.; Mattes, R. D. Human dietary responses to perceived mampulation of fat content in a midday meal. Am. J. Clin. Nutr. 17:237-240; 1993.

44. Clauss, K.; Jensen, H. Oxathiazinone dioxides: A new group of sweetening agents. Angew. Chem. 12:869-881; 1973.

45. Collings, A. J. Metabolism of cyclamate and its conversion to cyclohexylamine, In: Kretchmer, N.; Hollenbeck, C. B., eds. Sugars and sweeteners. Boca Raton, FL: CRC Press; 1991: 217-227.

46. Commission. Proposal for a Council Directive on sweeteners for use m foodstuffs. Official J. Eur. Commun. 90/C 242:4-15; 1990.

47. Compadre, C. M.; Pezzuto, J. M.; Kinghorn, A. D.; Kamath, S. K. Hernandulcin: An intensely sweet compound discovered by review of ancient literature. Science 227:417-419; 1985.

48. Council on Sclentific Affairs. Aspartame. Review of safety issues. JAMA. 254:400-402; 1985.

49. Cowart, B. J.; Beauchamp, G. K. Early development of taste perception. In: McBride, R.; MacFle, H., eds. Psychological basis of sensory evaluation. Barking, England: Elsevier; 1990.

50. Crammer, B.; Ikan, R. Propertles and syntheses of sweetneing agents. Chem. Soc. Rev. 6:431--465; 1977.

51. Crook, C. K. Taste perception in the newborn infant. Infant Behav. Dev. 1:52-69; 1978.

52. Crosby, G. A. New sweeteners. CRC Crit. Rev. Food Sci. Nutr. 7:297-323; 1976.

53. Culberson, J. C.; Walters, D. E. Three-dimensional model for the sweet taste receptor. Development and use. In: Waiters, D. E.; Orthoefer, F. T.; DeBois, G. E., eds. Sweeteners: Discovery, molecular design, and chemoreception. ACS Symposium Series 450. Washington, DC: American Chemical Society: 1991: 214-223.

54. Dastoli, F. R.; Lopiekes, D. V.; Price, S. A sweet-sensitive pro-

322 S C H I F F M A N A N D G A T L I N

tein from bovine taste buds: Purification and partial characterization. Biochemistry 7:1160-1164; 1968.

55. Dastoll, F. R.; Price, S. Sweet-sensitive protein from bowne taste buds: Isolation and assay. Science 154:905-907; 1966.

56. DeS~mone, J. A.; Heck, G. L.; Mlerson, S.; DeSimone, S. K. The active ion transport properties of canine lingual epithelia in vitro. J. Gen. Physiol. 83:633-656; 1984.

57 Desor, J.; Greene, L.; Mailer, O. Preferences for sweet and salty m 9- to 15-year old humans. Science 190:686-687; 1975.

58. Deutsch, E. W.; Hansch, C. Dependence of relative sweetness on hydrophobic bonding. Nature (London) 211:75; 1966.

59. Diamant, H.; Oakley, B.; Strom, L.; Wells, C.; Zotterman, Y. A comparison of neural and psycholphysical responses to taste stimuli in man. Acta Physiol. Scand. 64:67-74; 1965.

60. Ddls, W. L., Jr. Sugar alcohols as bulk sweeteners. Annu. Rev. Nutr. 9:161-86; 1989.

61. Drewnowski, A.; Louis-Sylvestre, J.; Massein, C.; Fricker, J.; Chapelot, D.; Apfelbaum, M. Effects of sucrose and aspartame on hunger, energy intake, and taste responsiveness m normal- weight men and women. Abstract presented at the Annual Meet- ing, Society for the Study of Ingestive Behavior, Princeton, N J; June 1992.

62. DuBois, G. E. Sweeteners nonnutnttve. Encyclopedia of food science and technology. New York: John Wiley & Sons; 1991: 2470-2487.

63. DuBois, G. E.; Crosby, G. A.; Stephenson, R. A. Dihydrochal- cone sweeteners: A study of the atypical temporal phenomena. J. Med. Chem. 24:408--428; 1981.

64. DuBo~s, G. E.; Waiters, D. E.; Kellogg, M. S. The rational design of ultra-high potency sweeteners. In: Birch, G; Mathlou- thl, M., eds. Sweet taste chemoreception. Barking, England: Elsevier; (m press).

65. DuBols, G. E.; Waiters, D. E.; Schiffman, S. S.; Warwick, Z. S.; Booth, G. E.; Pecore, S. D.; Gibes, K.; Cart, B. T.; Brands, L. M. Concentration-response relationships of sweeteners: A systematic study. In: Waiters, D. E.; Orthoefer, F. T.; DuBois, G. E., eds. Sweeteners: Discovery, molecular design, and chemoreception. ACS Symposium Series 450. Washington, DC: American Chemical Society: 1991:261-276.

66. Dwivedi, B. K. Polyalcohols: Sorbitol, mannitol, maltitol and hydrogenated starch hydrolysates. In: Nabors, L. O.; Gelardi, R. C., eds. Alternative sweeteners, 1st ed. New York: Marcel Dekker; 1986.

67. Dwivedi, B. K. Sorbltol and mannitol. In: Nabors, L. O.; Gel- arch, R. C., eds. Alternative sweeteners. 2rid ed. New York: Marcel Dekker; 1991:333-348.

68. Esaki, S.; Tanaka, R.; Kamiya, S. Structure-taste rrelationships of flavanone and dihydrochalcone glycosides contaumng (1-2) linked disaccharides. Agric. Biol. Chem. 47:2319-2328; 1983.

69. Fantino, M.; Hosotte, J.; Apfelbanm, M. An opioid antagonist, naltrexone, reduces preference for sucrose m humans. Am. J. Physiol. 251:R91-R96; 1986.

70. Faull, J. R.; Halpern, B. Reduction of sucrose preference m hamster by gymnemic acid. Physiol. Behav. 7:903-907; 1971.

71. Faurion, A.; Bonaventure, L.; Bertrand, B.; MacLeod, P. Mul- tiple approach of the sweet taste sensory continuum: Psycho- physical and electrophysiological data. In: Van der Starre, H., ed. Olfaction and taste VII. London: IRL; 1980:86.

72. Faunon, A.; MacLeod, P. Sweet taste receptor mechanisms. In: B~rch, G. G.; Parker, K. J., eds. Sweet taste receptor mechanisms. London: Applied Science Publishers; 1982:247-273.

73. Faunon, A.; Saito, S.; MacLeod, P. Experimental evidence for several receptor sites involved in sweet taste chemoreception. In: LeMagnen, J.; MacLeod, P., eds. Olfaction and taste VI. London: IRL; 1977:60.

74. Faunon, A.; Saito, S.; MacLeod, P. Sweet taste involves several distract receptor mechanisms. Chem. Senses 5:107-121; 1980b.

75a. Federal Register. Food addiuves permitted for direct addition to food for human consumption; Aspartame. 21 CFR Part 172 [Docket No. 87F-0344]. Federal Register 58(73):21098-21099; 1993.

75b. Fisher, G. L.; Pfaffmann, C.; Brown, E. Dulcin and saccharin

taste m squirrel monkeys, rats, and men. Science 150:506-507; 1965.

76. Foltin, R. W.; Fischman, M. W.; Moran, T. H.; Rolls, B. J.; Kelly, T. H. Caloric compensation for lunches varying in fat and carbohydrate content by humans in a residential laboratory. Am. J. Clin. Nutr. 52:969-980; 1990.

77. Foltin, R. W ; Rolls, B. J.; Moran, T. H.; Kelly, T. H.; Mc- Nelis, A. L.; Fischman, M. W. Caloric, but not macronutrient, compensation by humans for required-eating occasions with meals and snack varying in fat and carbohydrate. Am. J. Clin. Nutr. 55:331-342;1992.

78. Frankmann, S. P.; Green, B. G. Effect of cooling the tongue on the sweetness of natural and artificial sweeteners. Chem. Senses 12:656; 1987.

79. Frostell, E. Natural and added sweet carbohydrates in foods and diets. In: Koivlstoinen, P.; Hyvonen, L. Carbohydrate sweeteners in foods and Nutrition. London: Academic Press; 1980: 1-14.

80. Fujino, M.; Wakimasu, M.; Mano, M.; Tanaka, K.; Nakajima, N.; Aoki, H. Structure-taste relationships of L-aspartyl- aminomalonic acid diesters. Chem. Pharm. Bull. 24:2112-2117; 1976.

81. Furia, T. E. Handbook of Food Additives, 2nd ed. Cleveland, Ohio: CRC Press; 1972.

82. Giza, B. K.; Scott, T. R. Blood glucose selectively affects taste- evoked activity in rat nucleus tractus solitarius. Physiol. Behav. 31:643-650; 1983.

83. Giza, B. K.; Scott, T. R. Intravenous insulin infusions decrease gustatory-evoked responses to sugars. Am. J. Physiol. 252: R994-R1002;1987.

84. Glaser, D.; Hellekant, G.; Brouwer, J. N.; van der Wel, H. The taste responses in primates to the proteins thaumatin and monellin and their phylogenetic implications. Folia Primatol. 29:56-63; 1978.

85. Glenn, J. F.; Erlckson, R. P. Gastric modulation of gustatory afferent activity. Physiol. Behav. 16:561-568; 1976.

86. Glinsman, W. H.; Dennis, D. A. Regulation of nonnutntive sweeteners and other sugar substitutes. In: Kretchmer, N.; Hol- lenbeck, C. B., eds. Sugars and sweeteners. Boca Raton: CRC Press; 1991:257-285.

87. Goldstein, N. I.; Cagan, R. H. Biochemical studies of taste sensation: Monoclonal antibody against L-alanine binding activity of catfish taste epithelium. Proc. Natl. Acad. Sci. USA. 79: 7595-7597: 1982.

88. Grenby, T. H. Intense sweeteners for the food industry: An overview, trends in food science and technology. 2:2-6; 1991.

89. Hall, W. G.; Bryan T. E. The ontogeny of feeding in rats: IV. Taste development as measured by intake and behavioral responses to oral infusions of sucrose and quinine. J. Comp. Phys- iol. Psychol. 95:240-251; 1981.

90. Halpern, B. P.; Smith, V. V. Selective modification of sweetness by ziziphins. In: Yoshlda, M., ed. Proceedings of the 16th symposium on taste and smell. Osaka: Japanese Society for Taste and Smell; 1982:81-84.

91. Hamor, G. H. Correlation of chemical structure and taste m the saccharin series. Science 134:1416-1417; 1961.

92. Harkins, R. W.; Miller, G. A. McNeil Specialty Products Co. personal communication. 1992.

93. Hellekant, G. Preference for sweet taste and the taste of sweeteners in ammals. J. Atom. Physiol. Anim. Nutr. Suppl. 11:43- 52; 1980.

94. Hellekant, G.; Glaser, D.; Brouwer, J. N.; van der Wel, H. Gustatory effects of miraculin, monellin and thaumatin in the Sagumus midas tamarm monkey studied with electrophys~ologi- ca] and behavioral techniques. Acta Physiol. Scand. 97:241-250; 1976.

95. Hellekant, G.; Gopal, V. On the effects of gymnemic acid in hamsters and rats. Acta Physiol. Scand. 98:136-142; 1976.

96. Hellekant, G.; Roberts, T. W. Study of the effect of gymnemic acid on taste in hamster. Chem. Senses 8:195-202; 1983.

97. Hendrick, M. E. Alltame. In: Nabors, L. O.; Gelardi, R. C.,

S W E E T E N E R S : S T A T E O F K N O W L E D G E R E V I E W 323

eds. Alternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991:29-38.

98. Herman, C. P.; Mack, D. Restrained and unrestrained eating. J. Personality 43:647-660; 1975.

99. Higginbotham, J. D. Talin protein (thaumatin). In: Nabors, L. O.; Gelardi, R. C., eds. Alternative sweeteners. 1st ed. New York: Marcel Dekker; 1986:i03-134.

100. Hiji, Y. Selective elimination of taste responses to sugars by proteolytic enzymes. Nature (London) 256:427-429; 1975.

101. Hiji, Y.; Kobayashi, N.; Sato, M. A "sweet-sensitive protein" from the tongue of the rat. Kumamoto Med. J. 21:137-139; 1968.

102. Hiji, Y.; Kobayashi, N.; Sato, M. Binding capacities of sugars with the "sweet sensitive protein" from the rat tongue. Kuma- moto Med. J. 22:104-107; 1969.

103. Hiji, Y.; Kobayashi, N.; Sato, M. Sweet-sensitwe protein from the rat tongue: Its interaction with various sugars. Comp. Bio- chem. Physiol. B 39B:367-375; 1971.

104. Hill, E. M.; Flaltz, C. M ; Frost, G. R. Sweetener content of common pediatric oral liquid medications. Am. J. Hosp. Pharm. vol. 45. 135-142; 1988.

105. Holleman, A. F. On some derivatives of saccharin. Recl. Tray. Chim. Pays-Bas 42:839-845; 1923.

106. Homier, B. E.; Deis, R. C.; Shazer, W. H. Aspartame. In: Nabors, L. O.; Gelardl, R. C., eds. Alternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991:39-69.

107. Hood, L. L.; Campbell, L. A. Developing reduced calorie bak- cry products with sucralose. Cereal Foods World, vol. 35. No. 12. 1990:1171-1182.

108. Hoppe, K. System analysis of the sweetness receptor. Stud. Bio- phys. 97:97-102; 1983.

109. Horowitz, R. M.; Gentili, B. Dihydrochalcone sweeteners from citrus flavanones. In: Nabors, L. O.; Gelardi, R. C., eds. Al- ternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991: 97-115.

110. Hough, C. A. M; Edwardson, J. A. Antibodies to thaumatin as a model of the sweet taste receptor. Nature (London) 271:381- 383; 1978.

111. Hough, L.; Khan, R. Intensification of sweetness. Trends Bio- chem. ScL 3:61-63; 1978.

112. Hough, L.; Phadnis, S. P. Enhancement in the sweetness of sucrose. Nature (London) 263:800; 1976.

113. lwamura, H. Structure-sweetness relationship of L-aspartyl dipeptide analogues: A receptor site topology. J. Med. Chem. 24: 572-583; 1981.

114. Jacobs, H. L. Observations on the ontogeny of saccharine preference in the neonate rat. Psychon. Sci. 1:105-106; 1964.

115. Jakinovich, W. Jr. Methyl 4,6-dichloro-dideoxy-ct-v-galactopyranoside: An inhibitor of sweet taste responses in gerbils. Science 219:408-410; 1983.

116. Jakinovich, W., Jr. Comparative study of sweet taste specificity. In: Cagan, R. H.; Kare, M. R., eds. Biochemistry of taste and olfaction. New York: Academic Press; 1981a:117-138.

117. Jakinovich, W., Jr. Stimulation of the gerbil's gustatory receptors by saccharin. J. NeuroscL 2:49-56; 1982.

118. Jakinovich, W., Jr. Stimulation of the gerbil's gustatory receptors by artificial sweeteners. Brain Res. 219:69-81; 1981b.

119. Jordan, H. A. Voluntary intragastnc feeding: Oral and gastric contributions to food intake and hunger in man. J. Comp. Phys- iol. Psychol. 68:498-506; 1969.

120. Kanders, B. S.; Lavin, P. T.; Kowalchuk, M. B.; Greenberg, I.; Blackburn, G. L. An evaluation of the effect of aspartame on weight loss. Appetite (Suppl. 11):73-84; 1988.

121. Kanders, B.; Blackburn, G.; Lavin, P.; Whatley, J.; Keller, S.; Conaway, B. Aspartame facilitates long-term weight maintenance in a population of reduced obese women undergoing mul- tidisciphnary treatment for obesity. Meeting of the North Amer- ican Association for the Study of Obesity/Society for the Study of Ingestive Behavior. Sacramento, NY; 1991.

122. Kawai, M.; Nyfeler, R.; Berman, J. M.; Goodman, M. Peptlde- sweeteners. 5. Side-chain homologues relating zwltterionic and trifluoroacetylated amino acids anilide and dipeptide sweeteners. J. Med. Chem. 25:397-402;1982.

123. Kennedy, L. M.; Halpern, B. P. A blphas~c model for action of the gymnemic acids and ziziphms on taste receptor cell membranes Chem. Senses 5:149-158; 1980a.

124. Kennedy, L. M.; Halpern, B. P. Fly chemoreceptors: A model system for the taste modifier zizaphm. Physiol. Behav. 24:135- 143; 1980b.

125. Kier, L. B. A molecular theory of sweet taste. J. Pharm. Sci. 61:1394-1397; 1972.

126. Kim, S. H.; DuBois, G. E. Natural high potency sweeteners. In: Mane, S.; Plggott, J. R. Handbook of sweeteners. Glasgow: Blackie; 1991:116-185.

127. Kinghorn, A. D.; Compadre, C. M. Less common high-potency sweeteners. In: Nabors, L. O.; Gelardl, R. C., eds. Alter- natwe sweeteners. 2nd ed. New York: Marcel Dekker; 1991: 197-218.

128. Kinghorn, A. D.; Soejarto, D. D. Stevioside. In: Nabors, L. O.; Gelardi, R. C., eds. Alternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991:157-171.

129. Kojima, S.; Ichibagase, H. Studies on synthetic sweetening agents. VIII. Cyclohexylamine, a metabolite of sodium cyclamate. Chem. Pharm. Bull. 14:971-974; 1966.

130. Kokoski, C. Sweeteners: Reproductive toxicity and teratology. In: Williams, G. M., ed. Sweeteners: Health effects. Princeton, NJ: Princeton Scientific Publishing Co.; 1988:131.

131. Krueger, J. M.; Cagan, R. H. Biochemical studies of taste sensation. J Biol. Chem. 251:88-97; 1976.

132. Kumar, A.; Weatherly, M. R.; Beaman, D. C. Sweeteners, fla- vorings: Dyes in antibiotic preparations. Pediatrics 87(3):352-

. 360; 1991. 133. Laessle, R. G.; Tuschl, R. J.; Kotthaus, B. C.; Pirke, K M.

Behavioral and biological correlates of dietary restraint in normal life. Appetite 12:83-94; 1989.

134. Lapldus, M.; Sweeney, M. L-4"-Cyano-3-(2,2,2-tnfluoro- acetamido) succmanilic acid and related synthetic sweetening agents. J. Med. Chem. 16:163-166; 1973.

135. LeBlanc, J.; Cabanac, M. Cephahc postprandial thermogenesis in human subjects. Physiol. Behav. 46:479-482; 1989.

136. Lehmann, P. A. The correlation of sweetness and bitterness of enantiomeric aminoacids. Life Sc~. 22:163 !- 1636; 1978.

137. Llndley, M. G. Phenoxyalkanoic acid sweetness inhibitors. In: Waiters, D. E.; Orthoefer, F. T.; DuBois, G. E., eds. Sweeten- ers: Discovery, molecular design, and chemoreceptlon. Wash- ington DC: American Chemical Society; 1991 ;251-260.

138. Lipsitt, L. P.; Behl, G. Taste-medmted differences m the sucking behavior of human newborns. In: Capaldi, E. D.; Powley, T. L., eds. Taste, experience, and feeding. Washington DC: Ameri- can Psychological Assocmtion; 1990;75-93.

139. Liu, Y.; Cao, T.; Xie, H.; Jian, G. Studies on the sweetness of dipeptides. Yi Yue Gong Ye (Pharmaceutical Industry) 10:11- 14; 1980.

140. Lo, C. H. The plasma membranes of bovine circumvallate papillae: Isolation and partial characterization B~ochlm. Biophys. Acta 291:650-661; 1973.

141. Looy, H.; Callaghan, S.; Welngarten, H. P. Hedonic response of sucrose likers and dislikers to other gustatory stimuli. Physiol. Behav. 52:219-225; 1992.

142. Looy, H.; Wemgarten, H. P. Effects of metabolic state on sweet taste reactivity in humans depend on underlying response profile. Chem. Senses 16:123-130; 1991.

143. Looy, H.; Weingarten, H. P. Facial expressions and genetic sensitavity to 6-n-propylthiouracil predict hedonic response to sweet. Physiol. Behav. 52:75-78; 1992.

144. Lum, C. K. L.; Henkin, R. I.Sugar binding to purified fractions from bovine taste buds and epithelial tissue. Relationslup to bloactivity. Biochim. Biophys. Acta 421:380-394; 1976.

145. Mandrile, E. L.; De Pfiner, G. M. B.; Cortella, A. Sweeteners of vegetable origin. I. Thaumatin, monellin, mlraculin, glycyrrhizin, stevioside and hernandulcin. Acta Farm. Bonaerense. Vol.7. 117-129;1988. Translated from Spamsh by Quest Tech- nology, Inc. Arlington Heights, IL.

146 Mattes, R. Effects of aspartame and sucrose in hunger and energy intake in humans. Physiol. Behav. 47:1037-1044; 1990.

324 S C H I F F M A N A N D G A T L I N

147. Mattes, R. Hunger ratings are not a valid proxy measure of reported food intake in humans. Appetite 15:103-113; 1990a.

148. Mazur, R. H. Aspartame-A sweet surprise. J. Toxicol. Environ. Health 2:243-249; 1976.

149. Mazur, R. H. Discovery of aspartame. In: Stegmk, L. D.; Fder, L. J., eds. Aspartame: Physiology and biochemistry. New York: Marcel Dekker; 1984:3-9.

150. Mazur, R. H.; Goldkamp, A. H.; James, P. A.; Schlatter, J. M. Structure-taste relationships of aspartic acid amides. J. Med. Chem. 13:1217-1221; 1970.

151. Mazur, R. H.; Reuter, J. A.; Swiatek, K. A.; Schlatter, J. M. Synthetic sweeteners. 3. Aspartyl dipeptide esters from L- and D-alkylglycines. J. Med. Chem. 16:1284-1287; 1973.

152. Mazur, R. H.; Schiater, J. M.; Goldkamp, A. H. Structure-taste relationships of some dipept~des. J. Am. Chem. Soc. 91:2684- 2691; 1969.

153. Mazur, R. (personal communication) 1992. 154. McBurney, D. Gustatory cross adaptation between sweet-tasting

compounds. Percept. Psychophys. 11:225-227; 1972. 155. McNeil Specialty Products Co. Food additive petition submitted

to U.S. Food and Drug Administration. Feb. 1987. 156. Meiselman, H. L.; Halpern, B. P. Effects of Gymnema sylvestre

on complex tastes elicited by amino acids and sucrose. Physiol. Behav. 5:1379-1384; 1970.

157. Meiselman, H. L.; Halpern, B. P.; Dateo, G. P. Reduction of sweetness judgments by extracts from the leaves of Ztztphus jujuba. Physiol. Behav. 17:313-317; 1976.

158. Miller, G. A. New Sweeteners. In: Kroger, M.; Freed, A. Changing food technology 2. Lancaster, PA: Technomic; 1989: 107-126.

159. Miller, G. A. Sucralose. In: Nabors, L. O.; Gelardi, R. C., eds. Alternative Sweeteners. 2nd ed. New York: Marcel Dekker; 1991:173-195.

160. Miller, S. A.; Frattali, V. P. Saccharin. In: Kretchmer, N.; Hol- lenbeck, C.B., eds. Sugars and sweeteners. Boca Raton, FL: CRC Press; 1991:245-255.

161. M~tchell, M. L.; Pearson, R. L. Saccharin. In: Nabors, L. O.; Gelardi, R. C., eds. Alternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991 : 127-156.

162. Monneuse, M. O.; Bellisle, F.: Louis-Sylvestre, J. L. Responses to an intense sweetener in humans: Immediate preference and delayed effects on intake. Physiol. Behav. 49:325-330; 1991.

163. Moskowltz, A. H. Maltitol and hydrogenated starch hydroly- sate. In: Nabors, L. O.; Gelardi, R. C., eds. Alternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991:259-282.

164. Muller, G. W.; Madigan, D. L.; Culberson, J. C.; Waiters, D. E.; Carter, J. S.; Klade, C. A.; DuBols, G. E.; Kellogg, M. S. High-potency sweeteners derived from beta-amino acids. In: Waiters, D. E.; Orthoefer, F. T.; DeBois, G. E., eds. Sweeten- ers: Discovery, molecular design, and chemoreception. ACS Symposium Series 450. Washington, DC: American Chemical Society; 1991:113-125.

165. Nabors, L. O.; Gelardh R. C. Intense sweeteners. In: Mario, S.; P~ggott, J. R. Handbook of sweeteners. Glasgow: Blackie; 1991:104-115.

166. Nalm, M.; Rogatka, H.; Yamamoto, T.; Zehavi, U. Taste responses to neohespendin dihyrochalcone m rats and baboon monkeys. Physiol. Behav. 28:979-986; 1982.

167. National Academy of Sciences, National Research Council, Committee on the Evaluation of Cyclamate for Carcinogenicity. Evaluation of cyclamate for carcinogenicity. Washington, DC: National Academy Press; 1985.

168. Newbrun, E. Dental effects of sugars and sweeteners. In: Kretchmer, N.; Hollenbeck, C. B., eds. Sugars and sweeteners. Boca Raton, FL: CRC Press; 1991:175-193.

169. Nofre, C.; Sabadie, J. Apropos de la proteine lingual dlte "sen- sible aux sucre." C. R. Hebd. Seances Acad Sci. Ser. D 274: 2913-2915; 1972.

170. Nofre, C.; Tinti, J.-M. US Patent 4,645,678. 1987. 171. Nofre, D.; Tinti, J -M.; Ouar Chatzopoulos, F. US Patent

4,921,939. 1990. 172. Noma, A.; HIjI, Y. Effects of chemical modifiers on taste re-

sponses in the rat chorda tympani. Jpn. J. Physiol. 22:393--402; 1972.

173. Nordmann, H. J. Will the new EC guideline on sweeteners lead to new nutritional awareness? FO Licht, Jahrbuch; 1991:9-10.

174. Nowlis, G. H.; Kessen, W. Human newborns differentiate differing concentrations of sucrose and glucose. Science 191:865- 866; 1976.

175. Oser, B. L. The saga of man-made sweeteners. In: Williams, G. M. Sweeteners: Health effects. Princeton, N J: Princeton Scien- tific Publishing Co.; 1988:25-37.

176. Ostretsova, I. B.; Safarian, E. K.; Etingof, R. N. On the presence and localization of glucose-binding proteins in the tongue. Dokl. Akad. Nauk SSSR 223:1484-1487; 1975.

177. Petersen, S.; Muller, E. Uber eine neue Gruppe yon Substoffen. Chem. Ber. 81:31-38; 1948.

178. Pfaffmann, C. Taste, its sensory and motivating properties. Am. Sci. 52:187-206; 1964.

179. Price, J. M.; Biava, C. G.; Oser, B. L.; Vogin, E. E.; Steinfield, J.; Ley, H. L. Bladder tumors in rats fed cyclohexylamine or high doses of a mixture of cyclamate and saccharin. Science 167: 1131-I 132; 1970.

180. Price, S. "Sweet-sensitive protein" from bovine tongues: Stereo- specific interactions with amino acids. In: Schneider, D., ed. Olfaction and taste IV. Struttgart: Wiss. Verlagsges.; 1972: 214-220.

181. Price, S.; DeSimone, J. A. Models of taste receptor cell stimulation. Chem. Senses Flavour 2:427-456; 1977.

182. Price, S.; Hogan, R. M. Glucose dehydrogenase activity of a "sweet-sensitive protein" from bovine tongues. In: Pfaffmann, C., ed. Olfaction and taste III. New York: Rockefeller Univ. Press; 1969:397-403.

183. Pritchard, T. C.; Scott, T. R. Amino acids as taste stimuli. I. Neural and behavioral attributes. Brain Res. 253:81-92; 1982a.

184. Pntchard, T. C.; Scott, T. R. Amino acids as taste stimuli. II. Quality coding. Brain Res. 253:93-104: 1982b.

185. Ramirez, I. Why do sugars taste good? Neuroscl. Biobehav. Rev. 14:125-34; 1990.

186. Rekker, R. F. The hydrophobic fragment constant, vol. 1. New York: Elsevier; 1977.

187. Remsen, I.; Fahiberg, C. On the oxidation of substitution products of aromatic hydrocarbons. Am. Chem. J. 1:426-438; 1879.

188. Reynolds, J. E. F., ed. Martindale the extra pharmacopoeia. 29th ed. London: The Pharmaceutical Press; 1989.

189. Rodin, J. Comparative effects of fructose, aspartame, glucose, and water preloads on calorie and macronutrient retake. Am. J. Clin. Nutr. 51:428-435; 1990.

190. Rogers, P. J.; Blundell, J. E. Separating the actions of sweetness and calories: effects of saccharin and carbohydrates on hunger and food intake m human subjects. Physiol. Behav. 45:1093- 1099; 1989.

191. Rogers, P. J.; Carlyle, J.; Hill, A. J.; Blundell, J. E. Uncoupling sweet taste and calories: Comparison of the effects of glucose and three intense sweeteners on hunger and food intake. Physiol. Behav. 43:547-552; 1988.

192. Rogers, P. J.; Pleming, H. C.; Blundell, J. E. Aspartame ingested without tasting inhibits hunger and food intake. Physiol. Behav. 47:1239-1243; 1990.

193. Rolls, B. J.; Hetherington, M.; Laster, L. J. Comparison of the effects of aspartame and sucrose on appetite and food intake. Appetite (Suppl. 11):62-67; 1988.

194. Rolls, B. J.; Kim, S.; Federoff, I. C. Effects of drinks sweetened with sucrose or aspartame on hunger, thirst and food intake in men. Physiol. Behav. 48:19-26; 1990.

195. Roils. B. J.; Laster, L. J.; Summerfelt, A. Hunger and food retake following consumption of low-calorie foods. Appetite 13: 115-127; 1989.

196. Sato, M.; Hiji, Y.; Ito, H. Taste discrimination in monkey. In: LeMagnen, J.; Macleod, P. eds. Olfaction and taste VI. Lon- don: IRL; 1977:233-240.

197. Schiffman, S. S.; Cahn, H.; Lindley, M. G. Multiple receptor sites mediate sweetness: Evidence from cross adaptation. Phar- macol. Biochem. Behav. 15:377-388; 1981.

S W E E T E N E R S : S T A T E OF K N O W L E D G E R E V I E W 325

198. Schiffman, S. S.; Clark, T. B.; Gagnon, J. Influence of chirality of amino acids on the growth of perceived taste intensity with concentration. Physiol. Behav. 28:457-465; 1982.

199. Schiffman, S. S.; Diaz, C.; & Beeker, T. G. Caffeine intensifies taste of certain sweeteners: Role of adenosine receptor. Pharma- col. Biochem. Behav. 24:429-432; 1986.

200. Schiffman, S. S.; Engelhard, H. H. Taste of dipeptides. Physiol. Behav. 17:523-535; 1976.

201. Schiffman, S. S.; Hopfinger, A. J.; Mazur, R. H. The search for receptors that mediate sweetness. In: Conn, M., ed. The receptors, vol. IV. New York: Academic Press; 1986:1129-1137.

202. Schiffman, S. S.; Lindley, M. G.; Clark, T. B.; Makino, C. Molecular mechanism of sweet taste: Relationship of hydrogen bonding to taste sensitivity for both young and elderly. Neuro- biol. Aging 2:173-185; 1981c.

203. Schiffman, S. S.; Lockhead, E.; & Maes, F. W. Amiloride reduces the taste intensity of Na + and Li+; salts and sweeteners. Proc. Natl. Acad. ScL 80: 6136-6140; 1983.

204. Schiffman, S. S.; Moroch, K.; Dunbar, J. Taste of acetylated amino acids. Chem. Senses Flavor 1:387-401; 1975.

205. Schiffman, S. S.; Reilly, D. A.; Clark, T. B. Qualitative differences among sweeteners. Physiol. Behav. 23:1-9; 1979.

206. Schiffman, S. S.; Sennewald, K.; Gagnon, J. Comparison of taste qualities and thresholds of D- and L-amino acids. Physiol. Behav. 27: 51-59; 1981.

207. Schiffman, S. S.; Sugarman, D.; Jakinowch, W. Jr.; Paikin, A.; Crofton, V. Inhibition of sweet taste in humans by methyl 4,6-dichloro-4,6-dideoxy-a-D-galactopyranoside. Chem. Senses 12:71-76; 1987.

208. Schiffman, S. S.; Warwick, Z. S.; Mackey, M. Sweetness and appetite in normal, overweight, and elderly persons. In: Fern- strom, J. D.; Miller, G. D., eds. Nutrition and central nervous system function. Washington: ILSI; 97-110; 1993.

209. Sela, M. N.; Steinberg, D. Glycyrrhizin: the basic facts plus medical and dental benefits. In: Grenby, T. H., ed. Progress in sweeteners. London: Elsevier Applied Science; 1989:71-96.

210. Shallenberger, R. S. Hydrogen bonding and the varying sweetness of the sugars. J. Food Sci. 28:584-589; 1963.

211. Shallenberger, R. S.; Acree, T. E. Chemical structure of compounds and their sweet and bitter taste. In: Beldler, L. M., ed. Handbook of sensory physiology, vol. IV, part 2. Berhn and New York: Springer-Verlag; 1971:221-277.

212. Shallenberger, R. S.; Acree, T. E; Lee, C. Y. Sweet taste of D and L-sugars and amino-acids and the steric nature of their chemoreceptor site. Nature (London) 221:555-556; 1969.

213. Shimazaki, K.; Sato, M.; Takegami, T. Binding of [3SS] saccharin to a protein fraction of rat tongue epithelia. Biochim. Bio- phys. Acta 677:331-338; 1981.

214. Spillane, W. J.; McGlinchey, G. Structure-activity studies on sulfamate sweeteners. II. Semiquantitative structure-taste relationship for sulfamate (RNHSO3) sweeteners: The role of R. J. Pharm. Sci. 70:933-935; 1981.

215. Stamp, J. A. Sorting out the alternative sweeteners. Cereal Foods World. 35:395--400; 1990.

216. Steiner, J. E. The human gustofacial response. In: Bosma, J. F., ed. Fourth symposmm on oral sensation and perception. Development in the fetus and infant. Bethesda, MD.: National Institutes of Health; [DHEW pubhcation no. (NIH) 73-546.]; 1973:254-278.

217. Striem, B. J.; Pace, U.; Zehavi, U.; Naim, M.; Lancet, D.Sweet tastants stimulate adenylate cyclase coupled to GTP-binding protein in rat tongue membranes. Biochem. J. 260:121-126; 1989.

218. Sukehiro, M.; Minematsu, H.; Noda, K. Structure-taste rela- tions of aspartyl peptide sweeteners. I. Synthesis and properties of L-aspartyl-D-aianine amides. Seikatsu Kagaku (Science of Hu- man Life) 11:9-16;1977.

219. Swithers-Mulvey, S. E.; Hall, W. G. Control of mgesUon by oral habituation in rat pups. Behav. Neurosci. 4:710-717; 1992.

220. Tahara, A.; Nakata, T.; Ohtsuka, Y. New type of compound with strong sweetness. Nature (London) 233:619-620; 1971.

221. Taylor, R. Hunger in the infant. Am. J. Dis. Child 14:233-257; 1917.

222. Temussi, P. A.; Lelj, F.; Tancredl, T. Three-dimensional map- ping of the seet taste receptor site. J. Med. Chem. 21:1154--1158; 1978.

223. Tennissen, A. M. Amiloride reduces intensity responses of human fungiform papillae. Physiol. Behav. 51:1061-1068; 1992.

224. Tepper, B. J.; Mattes, R. D.; Farkas, B. K. Learned flavor cues influence food intake in humans. J. Sensory Stud. 6:89-100; 1991.

225. The NutraSweet Company. A technical overview of Nutra- Sweet ®. The NutraSweet Company. Deerfield, Illinois.

226. Tinti, J.-M.; Durozard, D.; Nofre, C. Sweet taste receptor: Evi- dence of separate speeific sites for C O 0 - and NO2/CN groups in sweeteners. Naturwissenschaften 67:193-194; 1980.

227. Tinti, J.-M.; Nofre, C. Why does a sweetener taste sweet? A new model. In: Waiters, D. E.; Orthoefer, F. T.; and DeBois, G. E., eds. Sweeteners: Discovery, molecular design, and chemoreception. ACS Symposium Series 450. Washington, DC: American Chemical Society; 1991:206-213.

228. Tisdel, M. O.; Nees, P. O.; Harris, D. L.; Derse, P. H. Long- term feeding of saccharin in rats. In: Inglett, G. E., ed. Sympo- sium: Sweeteners. Westport, CT: Avi Publishing; 1974:145-158.

229. Tomarken, A. J.; Kirschenbanm, D. S. Effects of plans for future meals on counterregulatory eating by restrained and unrestrained eaters. J. Ab. Psychol. 93:458-472; 1984.

230. Tunaley, A. Perceptual characteristics of sweeteners, In: Grenby, T. H., ed. Progress m sweeteners. London: Elsevier; 1989:291-310.

231. United States Government, Federal Food, Drug, and Cosmetic Act, as Amended, and Related Laws. HHS Publication no (FDA) 8901051, Stock Number 1990 0-249-576-QL3. U.S. Gov- ernment Printing Office, Washington, D.C.; 1990.

232. van der Heijden, A.; Brussel, L. B. P.; Peer, H. G. Quantative structure-activity relationships (QSAR) in sweet aspartyl dlpep- tide methyl esters. Chem. Senses Flavour 4:141-152; 1979.

233. van der Heijden, A.; van der Wel, H.; Peer, H. G. Structure- activity relationships in sweeteners. I. Nitroanalines, sulfamates, oximes, isocoumarins and dipeptides. Chem. Senses 10:57-72; 1985.

234. van der Wel, H. Structural modification of sweet proteins and its influence on sensory properties. Chem. Ind. 1:19-22; 1983.

235. van der Wel, H.; Arvidson, K. Qualitative psychophysical studies on the gustatory effects of the sweet tasting proteins thaumatin and monellin. Chem. Senses 3:291-297; 1978.

236. Vigorito, M.; Sclafani, A. Ontogeny of Polycose and sucrose appetite in neonatal rats. Dev. Psychobiol. 21:457-465; 1988.

237. yon Rymon Lipinski, G. W. Acesulfame-K. In: Nabors, L. O.; Gelardi, R. C., eds. Alternative sweeteners. 2nd ed. New York: Marcel Dekker; 1991:11-28.

238. Warwick, Z. S.; Costanzo, P. R.; Schiffman, S. S. Differential response to sweetness, saltiness, and taste variety by restrained vs. unrestrained eaters. Proc. Ann. Meet. Eastern Psychol. Assoc. 34; 1990.

239. Warwick, Z. S.; Schiffman, S. S. Effect of meal sensory properties on postprandial hunger and taste reactivity in human subjects. Fourteenth Annual Meeting of the Assoc. Chemoreceptlon Sci. Sarasota; 1992.

240. Warwick, Z. S.; Schiffman, S. S. Sensory evaluations of fat- sucrose and fat-salt mixtures: Relationship to age and weight status. Physiol. Behav. 48:633-636; 1990.

241. Webb, P. Energy expenditure and the menstrual cycle. Am. J. Clin. Nutr. 44:614-619; 1986.

242. Wells, A. G. The use of intense sweeteners in soft drinks. In: Grenby, T. H., ed. Progress in sweeteners; London: Elsevier Applied Science; 1989:169-214.

243. WhisselI-Bueehy, D. Genetic basis of the phenylthiocarbamide polymorphism. Chem. Senses 15:27-37; 1990.

244. Wicker, R. J. Some thoughts on sweetening agents old and new. Chem. Ind. (London) 41:1708-1716; 1966.

245. Williams, G. M. Sweeteners: Health effects. Princeton, NJ: Princeton Scientific Publishing Co., Inc.; 1988.


246. Wolf, S.; Wolff, H. O. Human gastric function. London: Ox- ford University Press; 1947.

247. Wursch, P.; Anantharaman, G. Aspects of the energy value assessment of the polyols. In: Grenby, T. H., ed. Progress in sweeteners; London: Elsevier Applied Science; 1989:241-266.

248. Yackzan, K. S. Biological effects of gymnema sylvestre fractions. II. Electrophysiology-effect of gymnemic acid on taste receptor response. Ala. J. Med. Sci. 6:555-563; 1969.

249. Ylikahri, R. H.; Pelkonen, R. Carbohydrate sweeteners in metabolism and diseases. In: Kolvistoinen, P.; Hyvonen, L., eds. Carbohydrate sweeteners in Foods and Nutrition. London: Aca- demic Press; 1980:1-35.

250. Zawalich, W. S. Comparison of taste and pancreatic beta cell receptor systems. In: Schneider, D., ed. Olfaction and taste IV. Stuttgart: Wlss. Verlagsges.; 1972:280-286.

TABLE l

ACESULFAME-K

Physical and Chemwal

Chemical category: N-sulfonyl amide dihydrooxathiazinone dioxide (methyl derivative)

6-methyl- 1,2,3-oxathiazine-4-(~H)-one-2,2-dioxide potassium salt (237)

Molecular formula: C4H4KNO4S Molecular weight: 201.2 (188) Molecular structure:

O

Solubility: Acesuifame-K is readily soluble in water, N,N-di-

methylformamide, and dimethyl sulfoxide (31). The solubility of acesulfame-K in water at 20°C is approximately 270 g/l; at 100°C the solubility increases sharply to more than I000 g/l. The solubility of acesulfame-K in alcohol is 1 g/1 (237).

Stability (pH, heat): The sweetness does not appear to be affected by baking or

pasteurization temperatures (40). The dry form has a very good shelf life under dry conditions at room temperature, independent of exposure to light; aqueous solutions with pH above 2.5-3 show relatively good stability (7,242). However, the hydrolytic stability of acesulfame-K is less than that of saccharin (62). For example, in a cola system, 15070 is lost in one year at room temperature. Decomposition occurs in extreme conditions and the hydrolytic products are mainly ace- tone, CO2, ammonium salts, sulfate, and amidosulfonate (237).

Degradation route, in vivo: Acesulfame-K is not metabolized, but is excreted by the

kidneys unchanged; therefore, has no caloric value (41,237). Known incompatabilities: Acesulfame-K shows variable compatibilities with flavor-

ings used in commercial food products (242). Bulking agents such as polydextrose or polyalcohols are required for baking applications because acesulfame-K does not have the bulking properties of sugars (165).

Sweetness

Potency: Pw2 = 204, Pw5 = 140, Pw~0 = 34 where Pw2, Pws, and Pw~0 are the potencies in water equivalent to 2°70 sucrose, 5% sucrose, and 10% sucrose, respectively (65).

Typical concentration (range):

15'

t~

I0" e~

5"

Acesulfame-K

° - i t 500 1000 150O Concentration (ppm)

Response indicates sucrose sweetness equivalent in percent (65).

Temporal properties: The sweetness potency relative to sucrose decreases with

increasing concentration. The sweet taste of acesulfame K is quickly perceptable and diminishes slowly, perhaps more so than that of sucrose (242). The sweetness may be slightly enhanced in acidic solutions relative to neutral solutions (237). Acesulfame-K is synergistic with many other high-potency sweeteners, including aspartame and cyclamate. Blends of acesulfame-K and aspartame (1 : 1 by weight) or acesulfame-K and sodium cyclamate (1 :5 by weight) have been recommended (237). The sweetness quality of acesulfame-K is compatible with nutritive sweeteners including sorbitol, xylitol, isomalt, and maltitol (237).

Acesulfame-K has a strong bitter taste to many persons, especially at concentrations greater than 1000 ppm (205). It also has a metallic component (62).

Limits of Use

Dose restrictions: The Joint Expert Committee on Food Additives (JECFA)

of the FAO/WHO established an acceptable dally intake (ADI) level for acesulfame-K of 9 mg/kg body weight in 1983. The FDA established an ADI of 15 mg/kg body weight in 1988 (40,237). The ADI is the amount of a food additive that can be taken dally in the diet over a lifetime without risk.

The European Economic Community (EEC) has developed


a directive which specifies the acceptable maximum level of sweeteners that may be used as food additives to give specific foods a sweet taste. For acesulfame-K, label designation E950, the food categories and levels are nonalcoholic drinks 350 mg/ l, desserts 350 mg/kg, breakfast cereals 350 mg/kg, various fruit products 200-1000 mg/kg, confectioneries 500-2000 mg/ kg, bakery products 1000 mg/kg, specified alcoholic drinks 350 mg/l, sauces 350 mg/kg, and special dietary items 350 mg/kg (46).

Physiological and toxicological effects: The oral toxicity of acesulfame-K is very low with an LDso

of 6.9-8.0 g/kg body weight. In a multigenerational study in rats fed acesulfame-K at levels of 0, 0.3, 1.0, and 3.0070, no adverse effects were reported in reproductive performance or on teratologic examination (130). Extensive toxicological studies have found no potential carcinogenicity or mutagenicity (175,237).

Current Use

Source(s): Synthetic Dihydrooxathiazinone dioxides can be produced from ke-

tones, 15-diketones, derivatives of 15-oxocarbonic acids and al- kynes which are reacted with halogen sulfonyl isocyanates. Compounds resulting from this reaction are transformed into N-halogen sulfonyl acetoacetic acid amide. In the presence of potassium hydroxide this compound cyclizes to the dihydrooxathiazinone ring system (237). Salts of the ring system are formed because dihydrooxathiazinone dioxides are very acidic (165).

Commercial products and concentration(s): Acesulfame-K is suitable for low calorie and diabetic bever-

ages, jams and marmalade, confectionery items, sugarless chewing gums, reduced-calorie baked goods, fruit-flavored dairy products, oral hygiene products, pharmaceuticals, tobacco products and animal feedstuffs (237).

Frequency/advantages: Internationally, acesulfame-K is available in 100 products

(165). It is heat stable (165) and synergistic with aspartame and cyclamates (237).

Availability (manufacturers): Diamin (Vitalia, UK) (188) Sunette (Hoechst Celanese Corp, UK, US) (40,165,188) Hermesetas Gold (Jenks Brokerage, UK) Sweetex Plus (Crookes Healthcare, UK) Regulatory status: Acesulfame-K is approved for use in the US, UK, France,

Italy, Netherlands and Switzerland (46). In the US, acesulfame-K is approved for use in chewing gum, dry mixes for beverages, instant coffee, instant tea, gelatins, puddings, nondairy creamers, and as a tabletop sweetener (165). A directive for its use in countries in the EEC has been issued (46).

History

The sweetness of a dihydrooxathiazinone dioxide was accidentally discovered in 1967 in Germany by K. Clauss and H. Jensen at Hoechst AG while they were carrying out reactions with butyne and fluorosulfonyl isocyanate (44,165). Analogs of this chemical class were later synthesized and 6-methyl- 1,2,3-oxathiazine-4-(3H)-one-2,2-dioxide was found to have the most favorable taste qualities (237). In 1988, the FDA approved acesulfame-K as a food additive in the United States, specifically as a sugar substitute (in small packages and tablets), in chewing gum, and in dry bases for beverages,

instant coffee, instant tea, gelatins, puddings, pudding desserts and dairy product analogs (see 86 for review). US approval also includes confectionary products (hard and soft candies). Subsequent petitions in the US submitted by Hoechst Celanese have been for use in nonalcoholic beverages and beverage bases and in baked hoods and baking mixes. Acesul- fame-K has been approved in more than 50 countries, including the United Kingdom, Switzerland, France, Italy, Belgium, Australia, New Zealand, and South Africa (41,175). In countries other than the US, acesulfame-K is used in soft drinks, candies, toothpastes, mouthwashes, and pharmaceutical preparations (168).

ALITAME

Physical and Chemical

Chemical category: dipeptide L~-aspartyl-N-(2,2,4,4-tetramethyl-3-thietanyl)-D-alanin-

amide (31) composed of L-aspartic acid, D-alanine, and the amide of the amine group 2,2,4,4-tetramethylthietanylamine (215).

Molecular formula: CI4H25N3OaS Molecular weight: 331.43 Molecular structure:

CH~ NH H,C~\

CO,H CH~

Solubility: Alitame is very soluble in water (97,242). At 20°C, its solu-

bility ranges from 48.7% w/v at pH 2.0-24.907o w/v at pH 8.0.

Stability (pH, heat): In the neutral pH range (5-8), alitame is stable at room

temperature for more than 1 yr. Stability is less, however, as the pH is reduced to pH 2-4 (215). The half-life of alitame solutions at pH 7.0 and 100°C is 13.4 h.

Degradation route, in vivo: During metabolism the aspartic acid portion is hydrolyzed

off; the majority of the remainder is excreted as a mixture of metabolites. Only a small percentage is excreted unaltered.

Known incompatabilities: Alitame is incompatible with acidic food products includ-

ing cola under storage (Schiffman, personal observation), high levels of reducing sugars, hydrogen peroxide, and sodium bisulfite (97,242). A foul odor forms on storage of any acidic product.

Sweetness

Potency: Pw2 = 4500, Pws = 3430, Pw~o = 1640, where Pw2, Pws, and P,~o are the potencies in water at a sweetness equivalent to 2070 sucrose, 5°'/0 sucrose, and 10070 sucrose, respectively (65).



15

lo o t~

~ 5

A l l t a m e

0 I00 200 Concentr'aUon (ppm}


Temporal properties: Alitame has a clean sweet taste which develops rapidly and

lingers slightly. It is synergistic with acesulfame-K and cyclamates (97,215)

Limtts of Use

Dose restrictions: Alitame solutions exhibit taste saturation in which the

sweetness does not increase beyond 15070 sucrose equivalent (65).

Physiological and toxicological effects: The no effect level (NOEL) for alitame in animals is 100

mg/kg which is > 300 times the estimated mean chronic human exposure level. Animals treated with levels > 100 mg/kg for 5 days to 2 yr had increased liver weights secondary to the induction of hepatic microsomai metabolizing enzymes. In man, no enzyme induction was observed over a 14-day period at doses of 15 m g / k g / d a y (approximately 44 times the mean chronic intake estimate). No carcinogenic, embryotoxic, teratogenic, or mutagenic activity has been found (97). Alitame contributes approximately 1.4 kcal/g or 0.0207o of calories of replaced sucrose (165,215).

Current Use

Source(s): Alitame is synthetically formed from the amino acids L-

aspartic acid and D-aianine with a novel C-terminal amide moiety formed from 2,2,4,4-tetramethylthietanylamine.

Commercial products and concentration(s): Not available. Frequency/advantages: Does not degrade with heat. Availability (manufacturers): Manufactured by Pfizer. Regulatory status: The Food Additive Petition submitted to the US Food and

Drug Administration by Pfizer in 1986 was for the following application categories: baked goods and baking mixes, pre- sweetened ready-to-eat cereals, milk products, frozen desserts and mixes, fruit and water ices and mixes, fruit drinks including ales and mixes, confections and frosting, jams as well as jellies, preserves, and sweet spreads, sweet sauces including toppings and mixes, gelatins as well as puddings, custards, filling, and mixes, nonalcoholic beverages and mixes, dairy product analogs, sugar substitutes, sweetened coffee and tea beverages, candy, and chewing gum. Permission for use in a number of foreign countries has also been requested (97).

History

After the accidental discovery of the potent sweet taste of the dipeptide aspartame in 1965, Pfizer Central Research started an intensive research program in the 1970's to system- aticaily search for other-high potency sweeteners. They synthesized a large number of dipeptides of diverse structural types and ultimately patented alitame and structurally related dipeptide sweeteners. It is not yet approved anywhere in the world (97).

A S P A R T A M E


Chemical category: dipeptide N-L-c~-aspartyl-L-phenylaianine- 1-methyl ester 3-amino-N-(ct-methoxycarbonylphenethyl) succinamic acid Molecular formula: C~4HIeN20 ~ (188) Molecular weight: 294.3 (188) Molecular structure:

O

( o ) - " • CH2"CH" N H ' C ' C H " NH3 |

CH2 k 0 0 C H 3 ,,

Solubility: At pH 5.2, the isoelectric point, aspartame has a solubility

of 1°70 in water and 0.37070 in ethanol at 25°C (188,225) It is more soluble in acidic solutions and hot water (188). It is not soluble in fats and oils (106).

Stability (pH, heat): Aspartame is most stable in solid form and should be

stored in an airtight container. Under certain moisture, temperature and pH conditions, aspartame hydrolyzes to form aspartylphenylaianine (AP) or is converted to its diketopipera- zinc (DKP). DKP can convert to AP which can further convert to individual amino acid components. Hydrolysis to free amino acids tends to occur at pH 3.4 and lower; above pH 5.0 cyclization to diketopiperazine tends to occur. Hydrolysis and cyclization result in loss of sweetness because neither AP nor DKP are sweet (188, 225). In addition to DKP, there are two other minor degradation products: L-aspartyl-L-phenylalanine and fl-L-aspartyl-L-phenylalanine methyl ester; neither of these exhibits off tastes (I 26).

The stability in aqueous solutions is dependent on concentration, temperature, pH, buffer type and water activity (215). Aspartame is most stable at pH 4.3 (7,188).

Aspartame can be used in High Temperature Short Time (HTST) systems and ultra-high temperature (UHT) pasteurization and aseptic systems where the product can be cooled quickly, or when the aspartame is added aseptically after pasteurization. While early studies showed that aspartame could not be used in situations that required prolonged heating times such as in baking (7,215,225), an FDA approval for a heat-stable form of aspartame for baking was approved in 1993 (75a).

Degradation route, in vivo: Metabolic studies indicate that aspartame is hydrolyzed

rapidly and completely in the gastrointestinal tract to its con-


r ,

stituent amino acids, L-phenylalanine and L-aspartic acid, and to methanol; it is not absorbed as an intact molecule (32,225). Children metabolize aspartame as effectively as adults. Ad- ministration of 34-200 mg aspartame/kg body weight (equivalent to 4-24 liters of aspartame-sweetened beverage) produced plasma concentrations of aspartic acid and phenylalanine well below levels associated with toxicity. For example, a serving of skim milk has approximately 13 times more aspartic acid and 6 times more phenylalanine than an equal volume of beverage sweetened with aspartame. Approximately 4 to 5 times more methanol is obtained from a serving of tomato juice than an equivalent volume of aspartame-sweetened beverage. Acute doses of aspartame up to 100 mg/kg body weight are also metabolized effectively by persons heterozygous for phenylketonuria, a genetic disease characterized by a deficiency in the ability to metabolize phenylalanine (32).

It provides 4 kcal/g; however, because its potency is as much as 250 times that of sucrose, only very small amounts are needed to sweeten foods. This leads to a dramatic reduction in total calories consumed in aspartame-containing products compared to an equally sweet sucrose-containing product (215).

Known incompatabilities: Aspartame has a normally reactive amino group and car-

boxyl group as well as an amide bond and a methyl ester. Usually, foods do not contain large quantities of chemicals that might spontaneously react with a carboxyl or amino function. An exception is cinnamon flavor which contains cinna- maldehyde, an aldehyde which reacts readily with an unhin- dered amino group. Consequently, it is difficult to formulate cinnamon-flavored chewing gum with aspartame (153). This may be a general limitation with flavors significantly dependent on aldehydes (e.g. acetaldehyde, benzaldehyde, and va- nillin).

Sweetness

Potency: Pw2 = 250, Pws = 196, Pw~0 = 107, where Pw2, Pws, and Pw~o are the potencies in water at a sweetness equivalent to 2% sucrose, 5% sucrose, and 10% sucrose, respectively (65).

Aspartame sweetness potency in food products will vary depending on the sweetener concentration and its application to a particular formulation.

Aspartame is used in beverage and food products at typical concentrations of 600 ppm in nonalcoholic beverages and 1000 ppm in desserts (173). Aspartame combined with other sweeteners provides an acceptable taste profile. Aspartame-xylitol mixture (2.4:97.6) and blends of 50 :50 with saccharin or acesulfame-K have been used (7).


15'

I0 '

5'

0 0

Aspar tame

1000 2000 3000 Concentration [ppm)


Temporal properties: For most persons aspartame has a clean, sweet taste with-

out a bitter, chemical or metallic aftertaste (106,242). How- ever, some persons report that it has an aftertaste characterized by lingering sweetness or bitter sweetness, predominantly after repeated tasting (7). The sweetness develops more gradu- ally and persists slightly longer than sucrose (106,242). The AT (appearance time) for sweetness of aspartame is 5 s compared to 4 s for sucrose; the ET (extinction time) is 19 s for aspartame compared with 14 s for sucrose (126).

Limits of Use

Dose restrictions: The Joint F A O / W H O Expert Committee on Food Addi-

tives approved aspartame with an ADI of up to 40 mg/kg body weight/day in 1981 (7,188). The ADI of the diketopiperazine, a decompostion product of aspartame, is 7.5 mg/kg (188). The US FDA established, in 1984, an ADI of 50 mg/kg body weight/day with a request for a postmarket surveillance system to monitor aspartame consumption (32,215). The results of postmarketing surveillance have demonstrated that aspartame consumption by the general population (90th percentile) is only about 2.5-3 mg/kg/day, which is well below the ADI (32).

Products containing aspartame must be labeled to alert persons with phenylketonuria (PKU) of their need to restrict intake of phenylalanine from all dietary sources (88). PKU is a rare metabolic disease in which there is an absence of the enzyme phenylalanine 4-monooxygenase which is involved in the metabolism of phenylalanine. The phenylalanine content from aspartame in aspartame-contalning products is far less than would be found in meat, milk, and other protein foods.

The EEC has proposed a directive which specifies the acceptable maximum level of sweeteners that may be used as food additives to give specific foods a sweet taste. For aspartame, label designation E 951, these levels are 600 mg/I for nonalcoholic drinks, 1000 mg/kg for desserts; I000 mg/kg for breakfast cereals; 800 mg/kg for edible ices; 300 and 1000 mg/kg for fruits, depending on the product; I000-5500 mg/ kg for confectionery items, depending on the product; 1700 mg/kg for bakery products, 600 mg/l for alcoholic drinks; 300 mg/kg for fish products; 300 mg/kg for vegetables; 300 and 350 mg/kg for sauces; and 600-2000 mg/kg for special dietary items (46).

Physiological and toxicological effects: Toxicological studies have found no embryotoxic, terato-

genic or carcinogenic effects up to 2000-4000 mg aspartame/ kg body weight. The methanol obtained from aspartame metabolism is oxidized in the body to formic acid and ultimately to CO2 and H20; however, formic acid does not accumulate from aspartame consumption, and the levels are far below those where toxicity is observed. Intake of 200-500 mg methanol/kg body weight is required to produce toxic levels of formic acid; this is equivalent to a single dose of 240-600 liters of aspartame-sweetened beverage. Consumption of aspartame has no effect on neurologic function; it does not cause head- aches or seizures nor does it alter mood, cognition or behavior (33).

Because aspartame, as a high-potency sweetener, does not affect the blood glucose concentration in the diabetic person, it is useful as a sugar substitute to reduce caloric intake in the


Type II diabetic individual for whom weight loss is a goal. Aspartame provides the insulin-dependent diabetic with the flexibility of an unscheduled snack and the ingestion of other- wise substituted carbohydrates (175).

Aspartame is noncariogenic, and there is some evidence that it may be anti-cariogenic by inhibition of bacterial- growth and/or reduction in plaque formation by Streptococ- cus mutans, (175); however this has not been firmly established.

Current Use

Source(s): Aspartame is prepared by coupling the two constituent

amino acids, aspartic acid and phenylalanine, by two general procedures, one enzymatic and the other chemical coupling. In the enzymatic method, carbobenzoxy-L-aspartic acid (Z- Asp) is coupled with L- or DL-phenylalanine methyl ester in the presence of an enzyme under very special conditions. Be- cause the enzyme is specific for the formation of protein-like peptide bonds with L-amino acids, only the ~-carboxyl of Z- Asp reacts and only with L-Phe-OMe. This gives cleanly Z-L- Asp-L-Phe-OMe which can be catalytically hydrogenated to aspartame. The classical chemical method activates a protected L-aspartic acid by conversion to the anhydride and reaction of the latter with L-Phe or L-Phe-OMe. The Asp protecting group is either formyl or carbobenzoxy. Final purification, after deprotection, usually depends on the low water solubility of aspartame hydrochloride which is then neutralized to yield aspartame. The chemical synthesis method is less expensive than the enzymatic process.

Commercial products and concentration(s): Equal and NutraSweet (The NutraSweet Company, USA) Canderel and Flix (Searle, UK) (188), (Ajinomoto,

Japan) (Holland sweetener, Holland), (Miwon, Korea)

Frequency/advantages: Aspartame is available worldwide in over 5,000 products.

The largest single use is in carbonated soft drinks. The Nu- traSweet brand is most readily available in the United States and is used in soft drinks in an approximate concentration of 525 mg/l or ppm. This would give a product with sweetness equivalent to 9°70 sucrose if the potency of aspartame at this concentration is 180 times that of sucrose. The sweetness intensity by a particular concentraion depends greatly on the specific formulation.

Aspartame is also commonly used as a tabletop sweetener with the quantity equivalent to two teaspoons of sucrose pack- aged in a packet. This is typically 35 mg aspartame plus various extenders such as glucose or maltodextrin.

Availability (manufacturers): The NutraSweet Company, USA Searle, UK Ajinomoto, Japan Holland Sweetener, Holland Miwon, Korea Regulatory status: Aspartame is approved for use in over 90 countries, includ-

ing all major countries. The directive proposed by the EEC will uniformly regulate its use in the twelve European Commu- nity (EC) countries, and presumably those that would join the EC in the future.

History

In 1965, Jim Schlatter, working with Dr. Robert Mazur on the synthesis of the C-terminal tetrapeptide of gastrin, discovered the sweet taste of aspartylphenylalanine methyl ester (aspartame). From 1965 to 1970 over 200 analogs of aspartame were synthesized but none was more satisfactory than aspartame itself (33,106). In 1974, the FDA approved aspartame for use in dry products including cold breakfast cereals, chewing gum, dry beverage mixes, instant tea and coffee, gelatins, puddings, fillings, dairy product analog toppings, and tabletop sweeteners. However, the FDA stayed the regulation and appointed a public board of inquiry to resolve safety issues (155). In 1981, the marketing stay was lifted, and aspartame was approved for dry uses, according to the 1974 ruling. In 1983, aspartame was approved for use in carbonated beverages. Since 1983, additional uses have been approved in the US. However, it is estimated that at least 80070 of the US market is carbonated soft drinks with a significant part of the remainder being tabletop (packet) use (48). In addition to the US, aspartame has been approved for use in foods, beverages, and as a tabletop sweetener in over 90 countries (106).

CALCIUM CYCLAMATE


Chemical category: sulfamate Calcium cyclohexylsuifamate Calcium cyclohexanesulfamate (31,188) Molecular formula: Cj2H 2 N206S2 Ca (24,188) Molecular weight: 396.5 (24,188), 432.6 as dihydrate (188) Molecular structure:

[ 2

Solubility: 1 g/4 ml in water 1 g/60 ml in ethanol 1 g / l .5ml in propylene glycol (24,81) Stability (pH, heat): Solutions are stable to heat, light, and air throughout the

pH range 2-10 (24,188,242). Degradation route, in vivo: The breakdown of calcium cyclamate does not provide any

calories. Sodium and calcium cyclamate and cyclamic acid are incompletely absorbed from the gastrointestinal tract; some cyclamate is excreted in the urine. About 25% of a human population may convert cyclamates to cyclohexylamine which is then excreted in the urine. The conversion results from the action of microflora on the nonabsorbed cyclamate in the intestinal tract rather than metabolism by human cells (24).

The amount of cyclohexylamine converted varies greatly between individuals from < O. 1 070 to a maximum of 60070 (25,188).


Known incompatabilities: Calcium cyclamate is incompatible with nitrites in acid solution

and exhibits limited comparability with potassium salts, carbon- ates, citrates, phosphates, sulfates, and tartrates (188). Incompa- tabilities with caramel and pectin have been reported (242).

Sweetness

Potency: Pw2 = 26, Pw5 = 32, Pw~ 5 = 18 where Pw2 is the potency in water at a sweetness equivalent to 2% sucrose, Pw5 equivalent to 5°7o sucrose and Pw~ 5 equivalent to 7.5°70 sucrose (65).

Typical concentration (range): (see sodium cyclamate) Temporal properties: When the concentration of calcium cyclamate approaches

0.507o in the formulation, a bitter taste becomes noticeable (188). Sweetness of cyclamate builds more slowly to a maximal level and persists longer than that of sucrose (24). The taste of cyclamate is very compatible with fruit flavors (24). The off-tastes for calcium cyclamate begin at a slightly lower concentration than for sodium cyclamate (24).

Limits of Use

Technical properties: Advantages of calcium cyclamate in product formulation

are that it is nonhygroscopic and does not support mold or bacterial growth. It also does not provide the bulk, body, or texture that are characteristic of sugar (65).

Dose restrictions: The following table shows the results of laboratory tests to

determine the LDso of calcium cyclamate when dosed by various routes in various animals.

animal route LDs0 (25)

mouse oral 7.2 g/kg mouse intravenous 0.57 g/kg rat subcutaneous 0.1 g/kg hamster oral 4.6 g/kg rabbit intravenous 0.12 g/kg

Cyciohexylamine, a metabolic by-product of cyclamates, has an oral LDs0 of 156-614 mg/kg in rats; this metabolite is hence more toxic than cyclamates, and this toxcicity limits the use of the sweetener (24,25).

The EEC has adopted a directive which specifies the acceptable maximum level of sweeteners that may be used as food additives in ready-for-use commercial products to give specific foods a sweet taste. For cyclamic acid and its sodium and calcium salts, designated E 952 in this directive, the food categories and levels are nonalcoholic drinks 400 mg/l , desserts 250 mg/kg, breakfast cereals 250 mg/kg, edible ices 400 mg/kg, fruit 1000 mg/kg, fruit preparations 250 mg/kg, confectionery items 1000 mg/kg, sugar free chewing gum 3000 mg/kg, bakery products 1700 mg/kg, alcoholic drinks 350 mg/l , and special diet formulas/supplements 400 and 1000 mg/kg (46). The EEC designates a temporary ADI = 11.

Physiological and toxicological effects: Calcium cyclamate is not cariogenic, but it also is not anti-

cariogenic as are some other high-potency sweeteners (24,25). Softening of the stools and diarrhea are common side ef-

fects of cyclamate ingestion; this laxative action is due to a change in the osmotic activity of the unabsorbed fraction of cyclamate salts in the gastrointestinal tract (24,25).

Cyclohexylamine in the diet is known to produce testicular

atrophy, reduced body weight gain, and hyperactivity in rats (25,45). It may also raise blood pressure in susceptible individuals (24). A report issued by a National Academy of Sciences National Research Council Committee concluded that cyclamate by itself is not carcinogenic; however, cyclamate may be a co-carcinogen because cyclamate-saccharin mixtures increase the risk of bladder cancer in animal models (24,86,167). Also see section on sodium cyclamate.

Current Use

Source(s): Synthetic Commercial products and concentration(s): Sucaryl Calcium (Abbott, Canada), Cyclan Frequency/advantages: A mixture of cyclamate: saccharin in a ratio of 10 : 1 was

popular during the 1960s. Mixtures of cyclamate and saccharin are synergistic and produce sweetness levels that are 20°7o higher than would be expected based on the levels of the individual components (24). Patents have recently been filed that indicate cyclamate has potential in combination with other sweeteners including aspartame and acesulfame-K (24).

Calcium cyclamate can be used to mask the bitterness and unpalatable tastes of drugs, especially liquid formulations and chewable tablets (24). Cyclamates have a low solid content at a desirable sweetness level which makes a suspension more fluid and reduces caking. Cyclamate disintegrates rapidly in tablet form.

Availability (manufacturers): Abbott Laboratories, USA Numerous manufacturers in Korea Regulatory status: Calcium cyclamate is available as a table-top sweetener or

for use in foods and beverages in over 40 countries, including Canada (24,25). It has been approved by the EEC Scientific Committee for Food and by the JECFA of FAO/WHO. It has been allocated a temporary ADI of 11 mg/kg body weight. It is presently permitted for use in Australia, New Zealand, Switzerland, Spain, and Germany. It may not be used in any drugs, other than those with approved new drug applications, or in any foods (including beverages) that are intended for use in the United States (242).

History

Sodium cyclamate was synthesized in 1937 by Sveda, who accidentally discovered that it has a sweet taste (8,25). Cycla- mates were first introduced in tablet form as a table-top sweetener for use by diabetics by Abbott Laboratories in the United States in 1951 (24,25,45). After enactment of the Food Addi- tive Amendment in 1958, sodium and calcium cyclamate were classified as GRAS (Generally Regarded As Safe) by the US FDA. Mixtures of cyclamate and saccharin (10: 1) subsequently became popular in the US during the 1960s (25). In the UK, cyclamate was allowed in soft drinks under the 1964 Soft Drink Regulations (242).

In 1966 it was reported that cyclamates could be metabolized by intestinal bacteria to cyclohexylamine (129). In 1968, an interim report from the National Academy of Sciences concluded from new animal studies that unrestricted use of cyclamate was not warranted (86). Approval for its use was withdrawn in the US when it was removed from the GRAS list in 1969 (215). In 1970, a long-term study in rats was published that reported an increased incidence of bladder tumors


compared to controls in animals fed mixtures of sodium cyclamate and sodium saccharin (10 : 1) and cyclohexylamine (179). This finding led to the ban of cyclamates as a food additive in many countries, including the United States. In 1982, Abbott Laboratories repetitioned for approval for its use (215).

FRUCTOSE


Chemical category: carbohydrate Molecular formula: C6HI206 (188) Molecular weight: 180.2 (188) Molecular structure:

CH2OH

:H2OH H H

OH H

Solubility: 1 g/0.3 ml water, 1 g/15 ml alcohol Stability (pH, heat): May be sterilized by autoclaving; store in airtight container

at temperature not exceeding 25 °C. Degradation route, in vivo: Absorbed from the gastrointestinal tract but more slowly

than dextrose; metabolized more rapidly than dextrose, mainly in the liver where it is phosphorylated and part is converted to dextrose. Insulin is considered not to be necessary for its metabolism. Fructose produces little effect upon the blood-sugar concentration, except in diabetic patients, who may metabolize it to dextrose to a greater extent than do nondiabetic subjects; quantitative advantages are not great.

Known incompatabilities: chlortetracycline was incompatible with solutions of levulose 10% in saline solution.

Sweetness

Potency: Pw2 = 1.3, Pw5 = 1.3, Pw,o = 1.3, where Pw2, Pws, and Pw,0 are the potencies in water at a sweetness equivalent to 2%0 sucrose, 5% sucrose and 10% sucrose, respectively (65).

Relative: The relative sweetness is dependent on temperature which af-

fects the mutarotational behavior of fructose. Sweetness is greater at 20°C than 80°C. Pure crystals have maxiumum sweetness because they have not undergone mutarotation in solution.


15"

i 10'

0 0

F r u c t o s e

5 10 Concentration (%1


Current Use

Source(s): Fructose is found naturally in many fruits and honey. How-

ever commercial fructose is synthesized from liquid dextrose.

GLYCYRRHIZIN OR GLYCYRRHIZIC ACID


Chemical category: triterpenoid glycoside 20/3-carboxy-I l-oxo-30-norolean-12-en-3fl-yl-2-O-fl-D-gluco-

pyanuronosyl-c~-D-glucopyranosiduronic acid Molecular formula: C42I-t620,6 (31) Molecular weight: 822.92 (31) Molecular structure:

aje. .CNI

¢ ~

I gl I

t

glll I ~I

U U

Solubility: Extracted from the plant as the K + and Ca ++ salts of

glycyrrhizic acid it is water soluble (145); conversion to ammoniated glycyrrhizin, the fully ammoniated salt, results in a more water-soluble compound (127).

Stability (pH, heat): Glycyrrhizin is reasonably stable at elevated temperatures

(127). Ammoniated glycyrrhizin precipitates at pH levels below 4.5.

Degradation route, in vivo: Glycyrrhizin is hydrolyzed by human intestinal flora to

glycyrrhetic acid which is absorbed from the small intestine in that form (209). Glycyrrhetic acid binds to plasma protein, enters the enterohepatic circulation, and is almost completely metabolized (88). Both glycyrrhizin and glycyrrhetic acid have mineralocorticoid and glucocorticoid properties because they possess chemical structures that resemble steroids (209). Thus, glycyrrhizin can affect electrolyte metabolism and result in retention of sodium.

Sweetness

Potency: Pw2 -- 252, Pw5 = 109, Pw~ = 14, where Pw2, Pw5, and Pw7 are the potencies in water at a sweetness equivalent to 2% sucrose, 5%0 sucrose and 7% sucrose, respectively (65).

Typical concentration (range): Response indicates sucrose sweetness equivalent in percent

(65).


Temporal properties:

15 M o n o a m m o n l u m Glycyrrhlztnate

0

0 - 0 I000 2000

ConcentraUon Ippm] 3000

Glycyrrhizin has a slow onset of sweetness and a long aftertaste (127). The AT (appearance time) is 16 s compared with AT = 4 s for sucrose. The ET (extinction time) is 69 s compared to 14 s for sucrose. Ammoniated glycyrrhizin has similar hedonic properties to glycyrrhizin. It has a very slow onset of sweetness which subsequently lingers very strongly (127).

Limits of Use

Dose restrictions: The Ministry of Health in Japan has issued the caution

that glycyrrhizin use should be less than 200 mg/day when used in drug formulations (127). The widespread use of glycyrrhizin and ammoniated glycyrrhizin by humans has been related to cases of pseudoaldosteronism with symptoms of hypertension, edema, sodium retention, and mild potassium diuresis (127).

Physiological and toxicological effects: Glycyrrhizin and liquorice possess pharmacological prop-

erties including anti-inflammatory and anti-cariogenic effects. Glycyrrhizin inhibits dental plaque formation by cariogenic bacteria and protects enamel from demineralization due to its antiglycolytic coating and buffering capacities (209). Glycyr- rhizin and liquorice have also been used to treat oral ulcers, herpetic lesions, bronchitis, and coughs (127,209). High intake has been related to hypokalemia, hyperprolactinemia (209), and pseudoaldosteronism manifested as hypertension, edema, sodium retention, and potassium diuresis (127).

Current Use

Source(s): Natural, extracted from the plant Glycyrrhiza glabra. Commercial products and concentration(s): In its natural form, glycyrrhizin is present in the plant in

yields of 6-1407o w/w as a mixture of various metallic salts (127,145). Commercial extraction and purification follows established procedures (127). In 1987 glycyrrhizin extract had almost a 3007o share of the high-potency sweetener market in Japan, being widely used for flavoring and sweetening food, beverages, medicines, cosmetics and tobacco (127). Liquorice is the British Adopted Name.

Frequency/advantages: Ammoniated glycyrrhizin is used as a flavorant, flavor

modifier, and foaming agent in confectionery and dessert items and in carbonated beverages with pH above 4.5. It is used at a concentration of 30-300 ppm to enhance the flavor of cocoa and chocolate-flavored products.

Availability (manufacturers): Glycyrrhiza Fluidextract, USNF

Pure Glycyrrhiza Extract, USNF Liquorice Liquid Extract, B.P. Deglycyrrhizinised Liquorice Extract, B.P. MagnaSweet (Wisconsin) Ammoniated glycyrrhizin Regulatory status: Ammoniated glycyrrhizin is listed as GRAS (generally

regarded as safe) by the United States Food and Drug Ad- ministration (FDA) when used as a natural flavoring agent (127). It is not approved for use as a sugar substitute at this time (215).

History

Glycyrrhizin (glycyrrhizic acid) is extracted from the rhi- zomes and roots of the plant Glycyrrhlza glabra which means "sweet root" in Greek. Glycyrrhtza glabra, better known as liquorice, is a wild perennial herb that grows in deep sandy, fertilized soil near water. Different varieties of Glycyrrhiza glabra are grown in Southern Europe, North Africa, and West and Central Asia. Both Glycyrrhiza glabra and glycyrrhizin itself have been used for centuries as a crude drug (209).

MALTITOL


Chemical category: carbohydrate O-a-D-glucopyranosyl-(1-4)-sorbitol Molecular formula: C~2H~O. Molecular weight: 346.34 Molecular structure:

CH2OH CH2OH H Jl~m=~O H H .L~ ~ O H

H OH H OH

Solubility: "Very soluble" (66) Stability (pH, heat): Like the other polyalcohols, due to the absence of free

carbonyl groups, maltitol does not undergo Maillard reactions to produce caramel-colored species in the presence of proteins (66,163). Maltitol is more heat stable and more resistant to microbial degradation than maltose. Maltitol is moderately hygroscopic (66).

Degradation route, in vivo: Mahitol is hydrolyzed by disaccharidases in the gastrointes-

tinal tract to yield free glucose and sorbitol. The glucose, which digestibility studies show accounts for about 50°70 of an ingested dose, (10,247) is slowly formed and rapidly absorbed, as seen by a relatively fiat blood glucose curve and slow insulin response (10,247). The sorbitol portion is incompletely absorbed; the absorbed portion being oxidized to fructose, which is further metabolized as fructose-l-phosphate independent of insulin (10).


Sweetness

Potency: Pw2 = 0.73, P.s = 0.72, P.75 = 0.71, where Pw2, P~s, and P.7 5 are the potencies in water at a sweetness equivalent to 2070 sucrose, 5070 sucrose and 7.507o sucrose, respectively (65).


Maltltol

O ~ T I ~ T !

0 10 20 Concentration (%1


Current Use

Source(s): Maltitol is prepared commercially by hydrogenation of

corn syrup that has a high maltose content. This maltose is the product of enzymatic hydrolysis of starch. Purification and concentration of the hydrogenated syrup yields crystalline maltitol in the proportion of 90-99% with the remainder being small amounts of sorbitol and hydrogenated trisaccharides (10). It is produced in substantial quantities only in Japan at the present time (66).

Commercial products and concentration(s): Hayashibara Biochemical Laboratories pioneered the pro-

cess of hydrogenation of maltose to produce maltitol and li- censes the patent to Town Chemical Industry, Co., Ltd., an affiliate of Mitsubishi, Tokyo, Japan and to ANIC and their subsidiary Melida, Sp.A, Milan, Italy. Town produces and markets crystalline maltitol and maltitol syrup under various brand names and is limited to specific geographic regions, which include Japan and the United States. The license re- stricts ANIC to marketing in Europe (163).

Availability (manufacturers): Malbit TM by ANIC/Melida Amalty TM and Mabit TM by Towa Chemical Industry Co, Ltd.

Htstory

Maltitol was developed in Japan and has been used there since about 1964 (163).

M A N N I T O L

Phystcal and Chemical

Chemical category: carbohydrate Molecular formula: C6H~406 (188) Molecular weight: 182.2 (188)

Molecular structure:

CH20H t

HO-C-H HO-~.,-H

H-(~-OH H-C-OH

Solubility: 220 g/l in water (67) Stability (pH, heat): Moderately hygroscopic (188); more heat stable than corre-

sponding mono- and disaccharides (67); stable to sterilization by autoclaving or filtration

Degradation route, in vivo: Mannitol and sorbitol have common metabolic pathways;

both are slowly absorbed by passive diffusion through the intestinal lumen. There are two different pathways for mannitol utilization: i) oxidation to D-fructose by the enzyme mannitol dehydrogenase in the liver, and ii) indirect metabolism by intestinal flora via fermentative degradation (60,67). The D-fructose is metabolized by the fructose-l-phosphate pathway to pyruvate or to glucose and glycogen, depending on the metabolic state of the individual (67).

Because mannitol is a poor substrate for enzymatic degradation, a significant portion of the ingested dose is excreted in the urine (60). Studies suggest that approximately 50°?0 of the ingested dose is utilized by the body, due to its incomplete metabolism and incomplete reabsorption by the kidney. Mannitol has a caloric value of 2 cal/g, the majority of this being secondary to the indirect metabolism by microbial fermentation (60).

Sweetness

Potency: Pw2 = 0.55, Pw5 = 0.64, P~0 = 0.67, where P,2, P~s, and P , J0 are the potency in water at a sweetness equivalent to 2°70 sucrose, 5070 sucrose, and 10% sucrose, respectively.


c- o

01 o

M a n n i t o l 1 2 "

1 0 "

8 "

6 .

4 "

2 '

0 2 4 S 8 10 12 14 16

Concentration (%)

Response indicates sucrose sweetness equivalent in percent.


Ltmits of Use

Dose restrictions: Mannitol is permitted as a food additive pending further

study. Any food product containing mannitol that could con- ceivably result in the ingestion of more than 20 g in one day is required to be labeled with the statement "Excess consumption may have a laxative effect" (60). In the United States it has interim food additive status and use levels may not exceed the following: Pressed mints 98o70, hard candies 5o70, cough drops 5%, chewing gum 31°/0, soft candies 40°70, confections and frostings 8%, nonstandardized jams and jellies 15%, and others 2.50/0 (67).

Physiological and toxicological effects: Mannitol is only slowly fermented by oral microorganisms,

thus decreasing the acidification of plaque, making it significantly less cariogenic than sucrose (60). Further study is necessary to understand the mechanisms and significance of the effect of enhanced mineral absorption by the polyols (60).

Current Use

Source(s): (natural, synthetic, both) Mannitol is widely distributed in plants, including the exu-

date from the ash tree, seaweeds, and mushrooms. It is commercially prepared by the hydrogenation of sugar, which yields a mixture of sorbitol and mannitol from which the mannitol readily crystallizes due to its lower solubility and can thus be separated (67).

Commercial products and concentration(s): Mannitol is used as a dusting powder and anticaking agent

because of its nonhygroscopic nature. Manuitol is administered intravenously as a 15-25% solution as an osmotic diuretic in acute renal failure, to reduce intracranial pressure, to reduce intraocular pressure, and to promote excretion of toxic substances (188).

Frequency/advantages: Mannitol is widely used in food manufacture in the US.

In a recent survey of excipients in pharmaceutic antibiotic preparations, mannitol was used in 7.7% of them, either alone or in combination with sucrose and/or saccharin (132). It is used more commonly in chewable tablets than in liquid syrups or suspensions. Information on the quantity of specific excipients is rarely provided by the drug manufacturer due to proprietary exclusion, but surveys of sweetener use in pharmaceuticals for pediatric use found mannitol used in the following ratios in chewable antibiotic tablets: sucrose : mannitol : saccharin 215 : 169 : 1 and 6.25 : 73 : 1 and mannitol : saccharin 171 : 1 (104).

Regulatory status: The regulations governing the use of mannitol vary in dif-

ferent countries, some listing it as sweetener and others only as a food additive. The regulations as described by countries include: South Africa and United States, permitted sweetener; United Kingdom and Federal Republic of Germany, permitted except where standard prohibits use but not listed as sweetener; Belgium, Denmark, Australia, Greece, Spain, Sweden and Switzerland, listed as sweetener and permitted in certain food stuffs only; France, Canada, and Japan, listed as food additive only; and Italy, Netherlands, Austria, Finland, Nor- way, not described (67).

NEOHESPERIDIN DIHYDROCHALCONE


Chemical category: dihydrochalcone glucoside 3,5-Dihydroxy-4- [3-(3-hydroxy-4-methoxyphenyl) propio-

nyl]phenyl 2-O- ( 6-deoxy-t~-L-rhamnopyranosyl)-/3-D-glucopy- ranoside (188)

Molecular formula: C28H36OI5 (188) Molecular weight: 612.6 (188) Molecular structure:

ONON

Solubility: 8.1 x 10 -4 M in distilled water at room temperature; 0.50

g/l increasing to approximately 653 g/1 at 80°C; sodium and calcium salts of neohesperidin dihydrochalcone have a solubility greater than 1 g/mi (109).

Stability (pH, heat): Neohesperidin dihydrochalcone is resistant to hydrolysis at

pH values above 2. In aqueous solutions stored for years at room temperature and exposed to light, there is little change in sweetness although some yellowing may occur (109).

Degradation route, in vivo: The metabolism of neohesperidin dihydrochalcone is similar

to that of other flavonoids and is brought about predominantly by the intestinal microflora. During metabolism there is cleavage between the substituted phloroglucinl A-ring and the adjacent carbonyl group to yield a series of C3-C6 acids derived from the B ring. The metabolic fate of the A ring is not well understood. Thin-layer chromatograms suggest that the metabolites from neohesperidin and neohesperidin dihydrochalcone are similar (109).

Sweetness

Potency: Pw2 = 1470, Pw5 = 906, Pw75 = 434, where Pw2, Pws, and Pw75 are the potencies in water at a.sweetness equivalent to 2o70 sucrose, 5o70 sucrose, and 7.5% sucrose, respectively (65).

Relative: Neohesperidin dihydrochalcone is approximately 1800 times

more potent than sucrose by weight at or near threshold. As the concentration increases to the intensity of 8.5°70 sucrose, it is only 200-340 times more potent than sucrose (109).


15- Neohcsperldin Dihydrochalcone

~ 10'

0 250 500 750 1000 Concentration (ppm)



Temporal properties: Neohesperidin dihydrochalcone (NHDC) has a slow onset

of sweetness with a lingering aftertaste which has been described as having a menthol or licorice-like quality. Mixture of neohesperidin dihydrochalcone with/3-gluconolactone may overcome some of the initial lag in sweetness perception (109).

Limits of Use

Dose restrictions: The EEC has developed a directive that specifies the ac-

ceptable maximum level of sweeteners that may be used as food additives in ready-for-use products to give specific foods a sweet taste. For neohesperidin dihydrochalcone, designated E 959 in the directive, the foods and levels include nonalcoholic drinks 30 and 50 mg/l , desserts 50 mg/kg, edible ices 50 mg/kg, fruit 50 mg/kg, confectioneries 100 and 400 mg/kg, and alcoholic drinks 10 and 20 mg/1 (46). NHDC has also been recommended for use in pharmaceuticals, especially to mask the bitter taste of the drugs. The Scientific Committee for Food of the European Committee gave NHDC an ADI of 5 mg/kg body weight in 1987 (109). The ADI is the amount of a food additive that can be taken dally in the diet over a lifetime without risk.

Physiological and toxicological effects: NHDC has an estimated caloric value of not more than 2

cal/g. Rats kept on a 10% diet of NHDC for 11 mo exhibited

growth inhibition and increased testes weight relative to body weight. A 5070 diet also showed reduced growth rate but no carcinogenicity in rats. In dogs fed a diet of approximately 6070 NHDC, liver and thyroid weights were elevated and thyroid hypertrophy and hyperplasia were found. Testicular atrophy and degeneration also occurred in several of these animals (109). Neohesperidin dihydrochalcone is not mutagenic according to the Ames test.

Current Use

Source(s): Flavanones, in the presence of alkali, undergo a ring-

opening reaction that produces chalcones. Catalytic reduction of chalcones yields dihydrochalcones (109).

Regulatory status: Neohesperidin dihydrochalcone has not been approved by

the FDA for use in the US because additional toxicological tests are needed. Neohesperidin dihydrochalcone is approved for use in chewing gum and some beverages in Belgium. Its use throughout the European Community is currently pending approval of the EC directive on sweeteners for use in foodstuffs (109).

History

The sweetening properties of dihydrochalcones were discovered by Horowitz and Gentili over 30 years ago (109). Both naringin and neohesperidin dihydrochalcone were synthesized from bitter citrus flavonones and found to have sweet tastes. Neohesperidin has been extracted from Seville oranges and converted to neohesperidin dihydrochalcone (109).

SODIUM CYCLAMATE


Chemical category: sulfamate sodium cyclohexylsulfamate (188)

Molecular formula: C6I--II2NO3S Na (188) Molecular weight: 201.2 (188) Molecular structure:

Na+

Solubility: 1 g/5 ml of water 1 g/250 ml of alcohol 1 g/25 ml of propylene glycol (188) Stability (pH, heat): Solutions are stable to heat, light and air throughout the

pH range of 2-10 (24,242). Degradation route, in vivo: The breakdown of sodium cyclamate does not provide any

calories. Sodium and calcium cyclamate and cyclamic acid are incompletely absorbed from the gastrointestinal tract; some cyclamate is excreted in the urine. About 25070 of a human population may convert cyclamates to cyclohexylamine which is then excreted in the urine. The conversion results from the action of microflora on the nonabsorbed cyclamate in the intestinal tract rather than metabolism by human cells (24). The amount of cyclohexylamine converted varies greatly between individuals from <0.1°70 to a maximum of 60070 (25,188).

Known incompatabilities: Sodium cyclamate is incompatible with nitrites in acid solu-

tion and has limited compatibility with potassium salts (188).

Sweetness

Potency: Pw2 = 26, Pw5 = 32, Pw~o = 18 where Pw2 is the potency in water at a sweetness equivalent to 2070 sucrose, P,,5 equivalent to 507o sucrose and Pwm equivalent to 10% sucrose (65).


15'

~ I0 ¸

0 0

Sodium Cyclamate

t t t 2500 5000 7500 I0000

Concentration {ppm|


Temporal properties: A bitter taste becomes noticeable as the concentration of

cyclamate in the formulation approaches 0.507o (188). Sweet- ness of cyclamate builds more slowly to a maximal level and


persists longer than does that of sucrose (24). The taste of cyclamate is very compatible with fruit flavors (24). The off- tastes begin at a slightly lower concentration for calcium cyclamate than for sodium cyclamate (24).

Limits o f Use

Dose restrictions: The Expert Committee of Food Additives of F A O / W H O

established an ADI of 11 mg/kg body weight in 1985 (7). The following table shows the results of laboratory tests to

determine the LDs0 of sodium cyclamate when dosed by different routes to different animals.

animal route LDs0 (25)

mouse oral 10-17 g/kg mouse intraperitoneal 7.1-12 g/kg mouse intravenous 4-4.8 g/kg rat oral 12-17.5 g/kg rat intraperitoneal 6 g/kg rat intravenous 3.5 g/kg hamster oral 10-12 gm/kg

Cyclohexylamine, a metabolic product of cyclamates, has an oral LDs0 of 156-614 mg/kg in rats; this metabolite is hence more toxic than cyclamates and this toxicity limits the use of the sweetener (24).

The EEC has developed a directive which specifies the acceptable maximum level of sweeteners that may be used as food additives in ready-for-use commercial products to give specific foods a sweet taste. For cyclamic acid and its sodium and calcium salts, designated E 952 in this directive, the food categories and levels are nonalcoholic drinks 400 mg/l , desserts 250 mg/kg, breakfast cereals 250 mg/kg, edible ices 400 mg/kg, fruit 1000 mg/kg, fruit preparations 250 mg/kg, confectionery items 1000 mg/kg, sugar-free chewing gum 3000 mg/kg, bakery products 1700 mg/kg, alcoholic drinks 350 mg/1, and special diet formulas/supplements 400 and 1000 mg/kg (24). The EEC has designated a temporary ADI of 11 for cyclamate.

Physiological and toxicological effects: Photosensitive dermatitis has been reported. Softening of

the stools and diarrhea are common side effects of cyclamate ingestion; this laxative action is due to a change in the osmotic activity of the unabsorbed fraction of cyclamate salts in the gastrointestinal tract (24,25). Cyclohexylamine in the diet is known to produce testicular atrophy, reduced body weight gain, and hyperactivity in rats (24,45). It may also raise blood pressure in susceptible individuals (24). A report issued by a National Academy of Sciences National Research Council Committee in 1985 concluded that cyclamate by itself is not carcinogenic; however, cyclamate may be a co-carcinogen because cyclamate-saccharin mixtures increase the risk of bladder cancer (24,86,167). Also see the section on calcium cyclamate.

Current Use

Source(s): Sodium cyclamate is synthetic. Commercial products and concentration(s): Sucaryl Sodium, Sucrosa, Assugrin Frequency/advantages: Sodium cyclamate can be used to mask the bitterness and

unpalatable tastes of drugs, especially liquid formulations and

chewable tablets (24). Cyclamates have a low solid content at a desirable sweetness level which makes a suspension more fluid and reduces caking. Cyclamate disintegrates rapidly in tablet form (24).

The actual relative sweetness varies with pH and product type. A mixture of cyclamate: saccharin in a ratio of 10:1 was popular during the 1960s. Mixtures of cyclamate and saccharin are synergistic and produce sweetness levels that are 10-20°70 higher than would be expected based on the levels of the individual components (24). Patents have recently been filed that indicate cyclamate has potential in combination with other sweeteners, including aspartame and acesulfame-K (24).

Availability (manufacturers): Abbott Laboratories Numerous manufacturers in Korea Regulatory status: Sodium cyclamate is available as a table-top sweetener or

for use in foods and beverages in over 40 countries, including Canada (24,25). It has been approved by the EEC Scientific Committee for Food and by the Joint Expert Committee on Food Additives of FAO/WHO. It has been allocated a temporary ADI of 11 mg/kg body weight. It is presently permitted for use in Australia, New Zealand, Switzerland, Spain, and Germany. It may not be used in any drugs, other than those with approved new drug applications, or in any foods (including beverages) that are intended for use in the United States (242).

History

Sodium cyclamate was synthesized in 1937 by Sveda who accidentally discovered that it has a sweet taste (8,25). Cycla- mates were first introduced in tablet form as a table-top sweetener for use by diabetics by Abbott Laboratories in the United States in 1951 (24,25,45). After enactment of the Food Addi- tive Amendment in 1958, sodium and calcium cyclamate were classified as GRAS (Generally Regarded As Safe) by the US FDA. Mixtures of cyclamate and saccharin (10: 1) subsequently became popular in the US during the 1960s (25). In the UK, cyclamate was allowed in soft drinks under the 1964 Soft Drink Regulations (242).

In 1966, it was reported that cyclamates could be metabolized by intestinal bacteria to cyclohexylamine (129). In 1968, an interim report from the National Academy of Sciences concluded from new animal studies that unrestricted use of cyclamate was not warranted (86). Approval for its use was withdrawn in the US when it was removed from the GRAS list in 1969 (215). In 1970, a long-term study in rats was pub- fished that reported an increased incidence of bladder tumors compared to controls in animals fed mixtures of sodium cyclamate and sodium saccharin (10 : 1) and cyclohexylamine (179). This finding led to the ban of cyclamates as a food additive in many countries, including the US. In 1982, Abbott Labora- tories repetitioned for approval for its use (215).

SODIUM SACCHARIN


Chemical category: N-sulfonyl amide 1,2-benzisothiazol-3(2H)-one- 1,1-dioxide, Na salt (31); 2,3-dihydro-3-oxobenzisosulfonazole, Na salt (31); dihydrate of sodium salt of 1,2 benzisothiazolin-3-one-l, 1-

dioxide (188) Molecular formula: C~H4NNaO3S; C~H4NNaO3S • 2H20 Molecular weight: 241.2 (188)



O

e eNa

2

Solubility: 1 g / l .2 ml water (31) 1 g/50 ml alcohol (31) Stability (pH, heat): Solutions of sodium saccharin are stable to pasteurization

but will undergo temperature-dependent, slow hydrolysis at low pH (242). The hydrolysis products are 2-sulfobenzoic acid and 2-sul famoylbenzoic acid (161 ).

Degradation route, in vivo: Saccharin is readily absorbed from the gastrointestinal

tract with almost all being excreted unchanged in the urine within 24-48 h (188). Because it is not metabolized, it provides no calories.

Known incompatabilities: Saccharin is compatible with other soft-drink components

(242).

Sweetness

Potency: Pw2 = 510, Pw5 = 444, PwTs = 247 where Pw2, Pws, and Pw75 are the potencies in water at a sweetness equivalent to 2070 sucrose, 5% sucrose, and 7.5070 sucrose, respec- ttvely (65).


15'

~ 10'

5'

0;" 0

Sodium Saccharin

. . t ! t 250 500 750 1000

Concentration (ppm)


Temporal properties: A bitter, metallic, and astringent aftertaste is detectable by

about 25070 of the population in products with a saccharin concentration over 0.01070. This aftertaste increases at higher concentrations. This characteristic taste is inherent in the saccharin molecule and not attributable to impurities. Various additives have been used in an attempt to mask this aftertaste, including sucrose, lactose, and tartrates. Saccharin has been combined most successfully with other high-potency sweeteners and with sugar substitutes. Examples are saccharin with cyclamate 1 : 10, saccharin with fructose 0.75 : 100, and saccharin with aspartame 1 : 1 (7,242). The temporal profile of saccharin is similar to sucrose. The AT (appearance time) is 4 s and the ET (extinction time) is 14 s (126).

Limits of Use

Dose restrictions: In the US the FDA has recommended that dally intake

should not exceed 1 g (188). The 28th Report of the Joint F A O / W H O Expert Committee on Food Additives gave an estimated temporary acceptable ADI for saccharin, including its calcium, potassium and sodium salts, of up to 2.5 mg/kg body weight. In 1969, the ADI had been 15 mg/kg.

The EEC has proposed a directive which specifies the acceptable maximum level of sweeteners that may be used as food additives to give specific foods a sweet taste. For saccharin and its No, K, and Ca salts, label designation E 954, these foods and levels are nonalcoholic drinks 80 mg/1, desserts 100 mg/kg, breakfast cereals 100mg/kg, edible ices 100 mg/kg, fruit 100-200 mg/kg, confectionery items 200-1200 mg/kg, bakery products 170 mg/kg, fish 160 mg/kg, vegetables 160 mg/kg, alcoholic drinks 80 mg/l , sauces 100-320 mg/kg, and special diet items 80-500mg/kg (46).

Physiological and toxicological effects: Sodium saccharin has low acute toxicity: LDs0 (oral rat)

= 14,200 mg/kg. High dose rat studies in a sensitive strain have suggested adverse effects including bladder cancer. How- ever, a review panel which studied the safety issues related to saccharin in 1983 concluded that the biochemical and physiological changes, including bladder tumors, which occur in rat at high doses do not occur in humans under normal patterns of use. A large bioassay conducted by the International Re- search Development Corporation (IRDC) in Mattawan, MI found that in rat, the lowest statistically significant incidence of bladder tumors was at 3070 of the rats diet (equivalent to 750 cans of soft drink per day over a lifetime). The tumors in rat appear to be species- and organ-specific (161). Epidemio- logic studies in diabetics support these findings. No increased risk of cancer (including bladder cancers) has been found in diabetics who consumed saccharin for extended periods (161). Saccharin is noncariogenic (242).

Current Use

Source(s): synthetic Toluene is treated with chlorosulfonic acid which yields

the ortho- and para-toluenesulfonyl chlorides. Upon further treatment with ammonia, the corresponding toluenesulfon- amides are formed. The ortho form is separated from the para form and is oxidized to ortho-sulfamoylbenzoic acid which cyclizes to saccharin when heated (161).

Commercial products and concentration(s): Saccharin Sodium Tablets (U.S.P.) Saccharin Sodium Oral Solution (U.S.P.) Saccharin Solution (B.P.C. 1954) Saccharin Tablets (B.P.C. 1973) According to the EC guidelines of 1991, saccharin is per-

mitted for use in nonalcoholic beverages, desserts, breakfast cereals, ice cream, fruits, confectioneries, bakery products, and alcoholic beverages. Approved maximum limits are 50 ppm in nonalcoholic beverages comparable to a sweetness provided by 2°70 sucrose and 100 ppm in desserts to compare to sweetness provided by 307o sucrose (173).

Frequency/advantages: Sodium saccharin is most commonly used in soft drinks,

dessert mixes, yogurt, and as a table-top sweetener. Availability (manufacturers): Various brand names and manufacturers in France Cumberland Packing Corp. (US) PCM Spec (Ohio)


Regulatory status: see History.

History Saccharin was accidentally synthesized by the American

chemist Constantine Fahlberg, and its potential as a sweetening agent was rapidly utilized commercially (see Ref. 160 for a review of history).

Saccharin, as well as the sodium and calcium salts of saccharin, were first listed as GRAS in 1959; ammonium saccharin was added in 1961. In 1972, the FDA removed saccharin and its salts from the GRAS list based on questions of safety raised by the Wisconsin Alumni Research Foundation (WARF) study (228). However, continued food use was regulated under an interim food additive regulation (see Ref. 86 for review.)

In 1977, the FDA proposed revocation of the interim food additive regulation and recommended a ban on saccharin use except as a nonprescription, single-ingredient drug, because there was evidence that is was a weak carcinogen producing bladder tumors in rats (6). The proposed ban was based on the Delaney Anticancer Clause which specifies that a carcinogenic food additive cannot be found as safe under the 1958 Food Additives Amendment to the 1938 Federal Food, Drug, and Cosmetic Act passed by Congress (see Ref. 25 for review). Congress intervened and passed the Saccharin Study and La- beling Act (SSLA) (231), which prohibited the FDA from ban- ning saccharin for use in food, drugs, or cosmetics for 18 too. SSLA directed further study of toxicity and carcinogenicity of saccharin by the Secretary of the Department of Health and Human Services (DHHS). In addition, it was stipulated that the following warning statement must appear on the label of food products: "USE OF THIS PRODUCT MAY BE HAZARD- OUS TO YOUR HEALTH. THIS PRODUCT CONTAINS SACCHARIN WHICH HAS BEEN DETERMINED TO CAUSE CANCER IN LABORATORY ANIMALS." The SSLA has been extended four times since it was passed in 1977.

S O R B I T O L

Physical and Chemical Chemical category: carbohydrate Molecular formula: C6HI406 (188) Molecular weight: 182.2 (188) Molecular structure:

C H 2 O H

- - O H

H O - -

- - O H

m - O H

C H 2 O H

Solubility: Sorbitol is very soluble in water: 235 g/100 g water at 25°C (67) 220 g/100 g water at 20°C (19) It is more soluble than sucrose in water (1 I). It is slightly soluble in acetic acid, alcohol and methylalco-

hol (188). Stability (pH, heat): Sorbitol is very hygroscopic. It is stable at high tempera-

tures with a melting point of 96-97°C (67). It is stable to sterilization by autoclaving (188). Sorbitol is more heat stable than the corresponding mono- and disaccharides and is more resistant to microbial degradation than the sugars (67). It does not caramelize due to the absence of free carbonyl groups (67).

Degradation route, in vivo: Sorbitol is slowly absorbed from the gastrointestinal tract

by passive diffusion. Part is oxidized, mainly in the liver, to fructose, catalyzed by sorbitol dehydrogenase (67,188,245). The fructose is then converted to fructose-l-phosphate, catalyzed by fructokinase in a step that is not dependent on or regulated by insulin. The fructose-l-phosphate is metabolized in the liver to dihydroxyacetone phosphate and glyceralde- hyde. The dihydroxyacetone phosphate is metabolized either to pyruvate or to glucose and glycogen, depending on the metabolic state of the person (67). Some sorbitol can undergo direct conversion to glucose in the presence of the enzyme aldose reductase (67,245). The portion that reaches the distal intestine is partially or completely fermented by the intestinal flora to short-chain volatile fatty acids which are then absorbed from the gut and metabolized (245). The presence of sorbitol in the small intestine can result in osmotic diarrhea, especially in large amounts.

Sweetness Potency: P.5 = 0.82, P.1o = 0.53, where P.5, and P.1o are

the potencies in water at a sweetness equivalent to 3% sucrose, 5070 sucrose, and 10°7o sucrose, respectively.

Typical concentration (range): Sorbitol is used as a vehicle for oral and topical liquids at

25-90070 and as a diluent for injectables below 10-25°70 (188). In a recent survey of excipients in 91 pharmaceutic antibi-

otic preparations, sorbitol was found to be used in 7.7070. It was present as the single sweetener or in combination with sucrose and/or saccharin in the liquid formulations of five antibiotics. Information on quantities of sweeteners is not readily available, but a search by Hill et al. (104) indicated use of sorbitol in the following ratio: sucrose : sorbitol : saccharin 500 : 70 : 1 in sulfamethoxazole 200 mg-trimethoprim 40 mg/ 5ml combination suspension (104). Sorbitol was also a sweetener in liquid respiratory preparations, iron supplementation liquids, and an H2-receptor antagonist liquid, either alone or in combination with sucrose, saccharin, or invert sugar. Quantity levels revealed diverse ratios: sorbitol :saccharin 400 : 1, 133 : 1,867 : 1, and 140 : 1; with sucrose : sorbitol the ratios were 3.2 : 1, 1.2 : 1, and 8.6 : 1 with various drug entities (104).

18,

16 ~

14"

12"

10"

6

4

S o r b l t o l

1 o 20 3 o ,'o Concent ra t ion (%)


Response indicates sucrose sweetness equivalent in percent.

Limits of Use

Dose restrictions: Sorbitol has a designation of GRAS, generally regarded as

safe, by the US FDA. If consumption of a sorbitol-containing food is likely to result in daily ingestion of more than 50 g, the label must contain the statement "Excess consumption may have a laxative effect" (60).

Sorbitol may be used in food at levels not to exceed Good Manufacturing Practices, which currently have the following maximum levels: 99o7o for hard candy and cough drops, 75o70 for chewing gum, 980-/0 for soft candy, 300-/0 for jams and jellies (nonstandardized), 170-/0 for baked goods and baking mixes, and 120-/0 for other foods (67).

The EEC has proposed a directive that specifies the acceptable maximum level of sweeteners that may be used as food additives to give specific foods a sweet taste. Sorbitol, designated E 420, may be used in all foodstuffs, excluding water- based flavoured nonalcoholic drinks, at a maximum level of quantum satis (46). It has become increasingly accepted that sugar alcohols have a lower caloric value than sucrose and glucose due to their incomplete absorption and metabolism. A caloric value of 2.4 kcal/g was allocated for all sugar alcohols in a recent Directive of the Council of the European Community (10).

Physiological and toxicological effects: Studies show that sorbitol can increase the absorption of

minerals from the intestinal tract, specifically calcium and iron. Mechanisms of action and clinical significance of such effects require further study (60).

Sorbitol has been tested in multigenerational studies in rats, mice, hamsters, and rabbits and no teratogenic or micro- scopic abnormalities were found (245).

There is disagreement on the advisability of dietary substitution of sorbitol for sucrose and other quickly absorbed carbohydrates in the diabetic patient. Opponents argue that sorbitol is capable of being converted to glucose and eventually requires insulin for its metabolism. Also, the accumulation of sorbitol intracellularly causes local edema and contributes to the development of diabetic complications, such as cataracts and polyneuritis. The proponents' research indicates that, although glucose is produced in sorbitol metabolism, there is such a delay that hyperglycemia is insignificant compared to the metabolism of other carbohydrates (67).

Current Use

Source(s): natural Sorbitol is found in significant amounts in many fruits

such as cherries and other stone fruits, (67,79) but it is not economically feasible to extract sorbitol commercially.

Sorbitol is prepared commercially by hydrogenation of glucose at high hydrogen pressure (70-140 atm) at 120-160°C with Raney nickel as a catalyst (67). When sucrose is the raw material, sorbitol (75°/o) and mannitol (250"/0) are produced simultaneously by the hydrogenation of the invert sugar (67).

Commercial products and concentration(s): Sorbitol Solution (U.S.P.) an aqueous solution containing

not less than 64070 w/w of sorbitol (188) Sorbitol intravenous infusion (B.P.) Sorbitol injection. A

sterile solution of sorbitol for parenteral use in water for injection.

Sorbitol Solution 70% (B.P.) Frequency/advantages: Sorbitol is used extensively in food, primarily baked goods,

frozen dairy products, soft candy and chewing gum as a bulk sweetening agent and for its humectant properties as a sof- tener and texturizer. It is used in combination with saccharin to decrease the bitter taste of the latter. Sorbitol is a sugar substitute in food items for diabetics (188).

Sorbitol is used as an excipient in pharmaceutical preparations for its unreactive nature (e.g., with amino acids), nonfer- mentability, lower cariogenicity and viscosity (19).

There is interest in the use of sorbitol as a sweetener in dentifrice products because it is much less cariogenic than sucrose due to its slow fermentation by oral microorganisms. This results in lower acidity of the plaque melieu that dissolves dental enamel calcium and phosphate, which eventually results in cavities (19,60,245). There also may be decreased ad- hesiveness of the microbial cells due to decreased formation of insoluble glucans that contribute to the matrix of microorganisms (60).

Availability (manufacturers): Numerous international sources under proprietary names Regulatory status: Sorbitol has a designation of GRAS by the US FDA. The Joint Expert Committee on Food Additives designated

an ADI of "not specified" in 1982 (10). Sorbitol is permitted in various countries with some legisla-

tive or regulatory limitations. It is listed as a permitted sweetener in Japan, South Africa, Sweden, and the United States. It is not listed as a sweetener but is permitted, except when standard prohibits use, in UK and Federal Republic of Ger- many. It is listed as a sweetener but permitted only in certain foodstuffs in Belgium, Denmark, Australia, Finland, Greece, Norway, Spain, and Switzerland. It is listed as a food additive only in France, Italy, Netherlands, Brazil, and Canada, and is not described in Austria (67).

History

Sorbitol was first isolated in 1872 from the berries of the mountain ash tree by French chemist Joseph Brussingault. Sorbitol is widely distributed in nature (67). Originally listed as GRAS in 1959, sorbitol was used at a maximum level of 7°-/0 in foods for special dietary use. In 1961 its category of use was changed to nutrient and /or dietary supplement. It was also regulated as a food additive for use as a stabilizer and nutrient sweetener with a maximum of 15 g/serving or 40 g/day; labeling was required to warn of laxative effects at higher intake (86). After comprehensive safety review by the Select Committee on GRAS Substances in 1974, the FDA af- firmed sorbitol as GRAS for a large number of technical effects and specified maximum levels in various foods (86).

STEVIOSIDE

Physical and Chem:cal

Chemical category: diterpene glycoside 13 - [(2 - O-/3 - D - glucopyranosyl - c~ - D - glucopyranosyl) oxy]

kaur- 16-en- 18-oic acid/3-D-glucopyranosyl ester (31) Molecular formula: C38 H~o O~8 Molecular weight: 804.9



O~ ~,0

~ =CM~I

Solubility: Sparingly soluble in water, 9300 ppm in water (72). Slightly

soluble in alcohol (31). Stability (pH, heat): Stevioside is stable at 100°C when maintained in solution

at pH ranging from 3 to 9. It decomposes at pH values of 10 and greater (128).

Degradation route, in vivo: In rat, most of an orally administered dose of stevioside is

degraded by intestinal flora to steviol, steviobioside and glucose which are then absorbed in the cecum. Glucose is then metabolized and excreted as CO2 and water. Steviol is conju- gated in the liver and excreted in the bile. Stevioside is also excreted unchanged in the feces (128). Metabolism in humans has not been investigated to date (128).

Sweetness

Potency: Pw2 = 193, P,5 = 120, P,75 = 59, where Pw2, Pws, and P,75 are the potencies in water at a sweetness equivalent to 2070 sucrose, 5°70 sucrose, and 7.5°70 sucrose, respectively (65).


15"

~ 10"

5'

0 0

Stevioslde

500 I000 Concentration Ippm)

xs'oo


Temporal properties: The AT (appearance time) for stevioside is 6 s which is

similar to sucrose (4 s); however, the extinction time (ET) is 22 s which is far greater than the ET for sucrose which is 14 s (126). Stevioside also exhibits some bitterness (128). Stevioside is synergistic with glycyrrhizin, aspartame, calcium cyclamate, and acesulfame-K (128),

Limits o f Use

Physiological and toxicological effects: There has been a long history of exposure to Stevia rebau-

diana in Paraguay. However, a related compound, steviol, the aglycone obtained on enzymatic hydrolysis of stevioside, has raised safety concerns even though no data exists that steviol is produced in vivo in humans. Steviol is an inhibitor of oxida- tive phosphorylation and is mutagenic following activation with liver microsomes (126). Steviol has been shown to be a by-product of stevioside in metabolic studies in the rat (128). Rodents dosed with Stevia rebaudiana extracts also exhibit reduced spermatogenesis, medullary cell proliferation in the adrenal glands, inflammatory lesions in the trachea and lungs, and pigmentation and increased hematogenesis of the spleen (128). It may also have hypotensive effects.

Current Use

Source(s): natural, extracted dulce, Stevia rebaudiana

from the leaves of yerba

SUCRALOSE

Physical and Chemical Chemical category: carbohydrate 1,6-dichloro- 1,6-dideoxy-/3-D- fructo fur anosyl-4-chloro-4-de-

oxy-(x-D-galactopyranoside (3 I) l ' ,4,6'-trichlorogalactosucrose (3 I) Molecular formula: CI2I-IIgCI3Os (31) Molecular weight: 397.64 (31) Molecular structure:

CH = OH CH zCl O H

I I O I ] H OH OH H

Solubility: The solubility of sucralose in water ranges from 28.3 g/100

rnl at 20°C to 66 g/100 ml at 60°C. Its solubility in ethanol ranges from 9.5 g/100 ml at 2°C to 18.9 g/100 ml at 60°C (159).

Stability (pH, heat): Sucralose is stable in aqueous solutions with no measurable

loss after one year of storage at pH 4.0, 6.0, and 7.5. At pH 3.0, there is less than 4°70 loss of sucralose after one year of storage. Sucrnlose is not susceptible to enzymatic hydrolysis (215). Sucralose is more stable to heat and acid than sugar, and does not break down under high heat conditions used for baked goods (107).

Degradalion route, in vivo: Sucralose does not hydrolyze or dechlorinate following inges-

tion in human, rat, mouse, dog, or rabbit (159). It does not provide any calories (18).

Known ineompatabilities: Sucralose has a good compatibility prof'de (242). The reactiv-

ity of sucralose as a 1.0070 solution was evaluated with 0.1070


concentrations of niacinamide, monosodium glutamate, hydrogen peroxide, sodium metabisulfite, acetaldehyde, ethyl acet- oacetate, or ferric chloride. The resulting data demonstrate that sucralose is relatively inactive chemically. Interactions between sucralose and any other food components are unlikely (159).

Sweetness

Potency: Pw2 = 614, P,5 = 636, P , t0 = 385 where Pw2, Pws, and Pwt0 are the potencies in water at a sweetness equivalent to 2070 sucrose, 50/0 sucrose, and 10°70 sucrose, respectively (65).

Typical concentration (range) Typical use levels range from 50 to 2000 ppm.

157

g I 0

5

Sucralose

I & 0 - - - - - ' " =1 - . -

0 1000 250 500 750 Coneealtrat/on (ppm)


Temporal properties: Sucralose has a good quality sweetness with no unpleasant

aftertaste (8,242). The sweetness time/intensity profile indicates that sucralose has a slower rate of sweetness decay than sucrose (159). The temporal properties of sweetness, onset, and lingering are similar to those of aspartame (158).

Limits of Use

Dose restrictions: not yet determined Taste saturation: The sweetness of sucralose does not increase

beyond 16070 sucrose (65). Physiological and toxicological effects: Sucralose is not acutely toxic, with an LDso > 16 g/kg in

mice and > 10 g/kg in the rat. It has no known mutagenic, teratogenic, carcinogenic, or neurotoxic activity and has no known effect on reproductive performance. No adverse effects applicable to man were observed in long-term sucralose studies in the mouse, the rat, and the dog at dose levels which are 700 (rat) and 400 (dog) times, respectively, the projected maximal mean dally human consumption of sucralose. However, under slow breakdown which can occur especially at low pH (4070 per year), two monosaccharides are produced. One (l,6-dichloro- 1,6-dideoxyfructose) is mutagenic in in vitro assays and cova- lently attaches to tissue macromolecules including DNA. The projected maximal mean daily human intake of sucralose is 100 mg or 2.3 mg/kg at the 90th percentile of projected intake level (92). Sucralose is also noncariogeuic (159).

The WHO JECFA recommended a temporary ADI level of 0-15 mg/kg/day in June 1990 (88).

Current Use

Source(s): Sucralose is synthesized from sucrose by replacement of three

hydroxyl groups with chlorine atoms at the 4-,1 ' - and 6 ' - posi- tions (the 4-position on the galactose moiety and the 1- and 6-

position on the fructose moiety) (159). This is achieved as follows: First, the primary alcohols of sucrose are blocked with trityl chloride and the remaining alcohol groups are protected with acetate. Next, the trityl protecting groups are removed and an acetate group is caused to migrate from the 4-position to the 6-position. Then the exposed alcohol groups are chlorinated. Finally, synthesis of sucralose is achieved by removal of the acetate groups (158).

Commercial products and concentration(s): The Food Additive Petition submitted to the US FDA by

McNeil Specialty Products Co. in 1987 was for the following application categories: baked goods and baking mixes, beverages and beverage bases, chewing gum, coffee and tea, confections and frostings, dairy products analogs, fats and oils (salad dress- ings), frozen dairy desserts and mixes, fruits and water ices, gelatins and puddings, jams and jellies, milk products, processed fruits and fruit juices, sugar substitutes, and sweet sauces, toppings, and syrups (155).

Frequency/advantages: Sucralose is not currently approved in the US. It is approved

in Canada and is used as a table-top sweetener. Availability (manufacturers): Redpath Specialty Products, Toronto, Canada McNeil Specialty Products, Inc. a subsidiary of Johnson &

Johnson, New Brunswick, New Jersey, USA Vending approval in the US and currently marketed as Splenda 'M in Canada)

Tate and Lyle PLC, London, UK (pending approval) Regulatory status: Sucralose is currently not approved by the FDA, but the Food

Additive Petition was filed in 1987. The FAO/WHO JECFA granted an ADI of 15 mg/kg/day in June 1990 (88). Sucralose was approved by the Health Protection Branch in Canada with permission for use in 13 categories as of September 1991. The EEC lists sucralose as not toxicologically acceptable (159).

History

During the 1970s, a program to develop new chemical entities from sugar was established at T a t e & Lyle. During this program, it was found that selective halogenation of sucrose greatly enhanced sweetness potency (110). Of the halogenated sugars, sucralose (1,6 - dichioro - 1 , 6 - dideoxy-/3- D - fructofur- anosyl-4-chloro-4-deoxy-t~-D-galactopyranoside) was ultimately selected for development and commercialization.

SUCROSE


Chemical category: carbohydrate Molecular formula: C,2H2201~ Molecular weight: 342.3 (31) Molecular structure:

c oH

2OH I-I O H O H i - I

Solubility: 1 g/0.5 ml water at 25°C (31,188) Can be prepared as unsaturated, saturated and supersatu-

rated solutions


Stability (pH, heat): Sucrose has good stability at room temperature and moder-

ate relative humidity. It absorbs up to 1070 moisture which makes its subject to microbial attack. Sucrose has a melting point of 186°C. Dry sugar melts when heated above 160°C, resulting in caramelization.

Degradation route, in vivo: Sucrose is hydrolyzed in the small intestine by sucrase to

glucose and fructose, which are then absorbed; sucrose is excreted unchanged in the urine when given intravenously (188).

Known incompatabilities: Potential contaminant trace heavy metals are incompatibile

with ascorbic acid; contamination with sulfite can cause color changes in coated tablets when the sucrose is used as an excipient.

Sweetness

Typical concentration (range): Cola-type beverages are sweetened to approximately 10070

sucrose (242). The response curve is linear by definition within the concentration limit of 15070 (65).

i g

15'

I0 '

5'

Sucrose

~ o

0 0 5 I0 15

Concentration (%)


Temporal properties: Sucrose is used as the comparison standard for expert taste

panels. Sucrose is described as having a pure, clean sweetness with few perceived secondary attributes, such as bitterness and astringency. Its sweetness time-intensity profile indicates a quick perception of sweetness and a sharp cut-off, (230) with an appearance time (AT) of 4 s and an extinction time (ET) of 14 s (126).

Limits of Use

Dose restrictions: Sucrose must be administered with care to diabetics. The

primary health concerns with the ingestion of sucrose are ex- cessive calories and tooth decay. Sucrose is also contraindi- cated in patients with fructose intolerance, glucose-galactose malabsorption syndrome, or sucrase-isomaltase deficiency.

Physiological and toxicological effects: LDs0 is 35 + 7 g/kg males and 29.7 g/kg in females

Current Use

Source(s): natural For commercial quantities, sucrose is extracted from sugar

cane and sugar beets. It is used as a sweetening agent and demulcent, lozenge base, tablet coating, and for syrup solutions.

Commercial products and concentration(s): Syrup (U.S.N.F.) sucrose 85070 w/v in water

THAUMATIN


Chemical category: several distinct proteins Molecular formula: protein Molecular weight: T~ = 22,209 (99), T2 = 22,293 Molecular structure: protein Solubility: Thaumatin is extremely soluble in water. It has good solu-

bility in aqueous alcohols, propylene glycol and in higher alcohols such as sorbitol (99).

Stability (pH, heat): The Talin protein (thaumatin) molecule is most stable to

heat between pH 2.7 and 6.0, with an optimum around 2.8- 3.0. At higher pH values, Talin protein becomes less stable to heat but at ambient temperatures is stable at pH up to 8-9. It can be pasteurized or sterilized at ultrahigh temperatures. Because Talin protein stability is enhanced at lower pH levels, it can be heated at 100 ° for several hours without sweetness loss, thus making it suitable for typical soft drinks, which have pH2.8 -3.5 (99).

Known incompatabilities: The thaumatin molecule has an overall positive ionic charge that allows it to form salts or polymers with suitably shaped, negatively charged ingredients such as food gums and synthetic colors. This interaction can result in molecular aggregation, dimerization, polymerization, reduction in sweet taste or actual precipitation. The thaumatin molecule can ionically bond to synthetic colors having sulfonate groups; the interaction is strongly dependent on concentration, pH, and temperature (99). Research efforts to make thaumatin more compatible have resulted in co-drying thaumatin with 6-20 parts of gum arabic or other weakly acidic polymer. The resulting bulkier product also has the added advantage of being denser and more easily handled and mixed (99).

Sweetness

Potency: Pw2 = 22,500, Pw5 = 14,200, Pw75 = 278 where Pw2 is the potency in water at a sweetness equivalent to 2070 sucrose, Pw5 equivalent to 5070 sucrose and Pw75 equivalent to 7.5070 sucrose (65).


15"

~ I0"

0 0

T h a u m a t l n

n * * I . I 20 4O

Concentration {ppm)

X _

I

t 6O



Temporal properties: Talin's flavor profile indicates a delay in perceiving the

sweetness, a slow buildup to maximum intensity, and a long, lingering sweet aftertaste without any unpleasant aftertaste (99).

Limits of Use

Dose restrictions: Thaumatin was assigned an ADI of "not specified" in the 29th Report of the Joint F A O / W H O Expert Committee on Food Additives in 1986.

Current Use

Source(s): Thaumatin is a natural product from the fruit of Thauma-

tococcus daniellii, a plant that grows abundantly in the rain forest belt of West Africa, primarily in Ghana, Ivory Coast, Togo, and Sierra Leone. After the fruit is collected, the arils are removed, frozen and transported to the UK for processing. The details of the extraction and purification process are proprietary, but basically, a solely aqueous extraction process using filtration and ultrafiltration are used (99).

The Unilever Research Laboratorium of The Netherlands has undertaken to produce Talin protein or its constituent proteins through genetic engineering. They have two patents for the recombinant DNA production of thaumatin and pre- prothaumatin. This technology is not yet applicable to production quantities (99).

Commercial products and concentration(s): Used in chewing gum in the US and Europe.

Availability (manufacturers): Trade name: Talin Regulatory status: Talin has been permitted as a natural protein in Japan since

1979, with recommended codes of practice and specification/ analysis published by the Japanese Food Additive Federation. A petition for its use in medicines is under consideration.

In the UK, the Talin protein has been listed as a safe excipient in medicines when accompanied by the appropriate product license as determined by the UK Committee on the Safety of Medicines in 1981. In 1982 the Food Additive and Contami- nants Committee agreed to regulate Talin protein as a sweetener. The Sweeteners in Food Regulations took effect in 1983 permitting the use of Talin protein in foods, drinks, and dietary products with the sole exception of baby foods (99). In the United States the Flavor Extract Manufacturers Associa- tion (FEMA) has designated Talin protein as GRAS (generally regarded as safe) as a flavor adjunct in chewing gum (99).

The F A O / W H O JECFA set specifications in 1983; in 1985 they agreed it was safe for food use and designated at ADI of "not specified" in its 29th Report. Petitions are pending in various other countries for use of Talin protein as a flavor enhancer and sweetener.

History

The sweetness of the red, pyramid-shaped fruit of the tropical plant was first described in the literature by British physi- cian-amateur botanist W.F. Daniell in 1855. John Joseph Ben- nett, F.R.S. first classified the plant as from the species Phrynium danielli but it was later classified as Thaumatococ- cus. No doubt the extreme sweetness of the fleshy top to the

seeds inside the fruit was known by natives for many centuries; it was used before sugar cane was introduced into West Africa (99).

XYLITOL


Chemical category: carbohydrate Molecular formula: C5I-II205 Molecular weight: 152.15 (31) Molecular structure:

CH2OH

- - - - O i l

H O - - -

- - - O H

CH2OH

Solubility: 169 g/100 g water (11) 1.2 g/100 g ethanol solution (31) Less soluble than sucrose (11). Stability (pH, heat): Stable at 120 ° and under normal food processing condi-

tions with no caramelization, caramelization occurs if heated for several minutes near boiling point of 216°C (760 mmHg) (11).

Degradation route, in vivo: Xylitol, like other polyols, is slowly absorbed from the

intestine because there is no specific transport system for facil- itated transport. After ingestion, approximately one-third of the ingested portion of xylitol is absorbed and enters the liver, where it is metabolized via the glucuronate-pentose phosphate cycle. The remainder of xylitol reaches the distal gut where it undergoes indirect metabolism via extensive fermentation by intestinal flora, yielding gas (H2, CH4, and COz) plus short- chain volatile fatty acids which are subsequently absorbed from the gut. In the liver, xylitol is converted to D-xylulose and then to fructose-6-phosphate without requiring insulin; insulin is necessary for utilization and storage of the glucose formed from further metabolism (11,188).

Known incompatabilities: Xylitol cannot be used in products requiring yeast as a

leavening agent because it inhibits growth and fermentative properties of yeast (1 I).

Sweetness

Potency: Pw5 = 0.97, PwJo = 0.77, where Pws, and Pw~0 are the potencies in water at a sweetness equivalent to 507o sucrose, and 10°7o sucrose, respectively.

Relative: Xylitol is considered to he isosweet with sucrose (11). All

other sugar alcohols are less sweet than sucrose. Typical concentration (range):


Xylitol has been used in an aspar tame:xyl i to l mixture 2.4 : 97.6 to provide a taste more like that of sucrose and to improve the stability of aspartame (7).

1 8 '

16,

14,

12'

10,

8

6

4

2

0 o 5

Xylitol

i"0 I ~ 20 25 Concentration (%)

Response indicates sucrose sweetness equivalent in percent. Temporal properties: There is a cooling effect when crystalline xylitol is dissolved

in the mouth because there is loss of heat when sugar alcohols are dissolved in water. This cooling effect is not perceived in products in which xylitol is already dissolved (11). Xylitol has a lower viscosity than maltitol (11).

Limits of Use

Dose restrictions: Doses larger than 50 g can cause osmotic diarrhea (249). Physiological and toxicological effects: There is the suggestion that sugar alcohols have a lower

caloric value than sucrose and glucose due to their incomplete absorption and metabolism. A caloric value of 2.4 kcal/g was allocated for all sugar alcohols in a recent Directive of the Council of the European Community (l 0).

Xylitol is not fermented by most oral microorganisms and plaque pH is not reduced upon exposure to xylitol (11). Xylitol is generally considered noncariogenic and may be anticariogenic. Under certain conditions, it reduces the cariogenic potential of sucrose (60). The possible anticariogenic effects are produced by several processes, including inhibition of acid formation from glucose by Steptococcus mutans (I 1).

Xylitol taken orally does not increase blood glucose or insulin levels, probably because conversion of xylitol to glucose is very slow (11). It may effect mineral absorption and utilization (60).

Xylitol has been found to have low acute toxicity by all routes of administration. It is not embryotoxic, teratogenic, mutagenic, or clastogenic (11).

Current Use

Source(s):

Xylitol is produced synthetically by chemical conversion of xylan which has been extracted from birchwood, almond shells, straw, corn cobs, or wastes from the pulp and paper industries. Xylan is first hydrolyzed to xylose which is then hydrogenated to xylitol in the presence of a nickel catalyst (11).

Commercial products and concentration(s): Xylitol is used as a sweetener in noncariogenic confection-

ely such as chewing gum, candies, chocolate and gum drops, and in foods for diabetics. It is also used in pharmaceutical preparations, including tablets, throat lozenges, multivitamin tablets, cough syrup, and toothpaste (1 I).

Eutrit, Klinit, and Kylit all are products in Japan. Frequency/advantages: Xylitol is more chemically inert than sucrose and thus phar-

maceutical preparations made with xylitol have good shelf life because they neither ferment nor mold. Xylitol is a good car- rier for tablets because of its low melting point (92-96°C). Solid dispersions of some drugs with xylitol showed a faster release than micronized drugs. Xylitol is also used in parenteral nutrition because it has little effect on insulin and thus does not suppress lipolysis (11).

Availability (manufacturers): Xylitol Injection (Jap.P) a sterile solution of xylitol in Wa-

ter for Injection Xylotin Company Regulatory status: The JECFA assigned an ADI of "not specified" in its 27th

report in 1983. This is a favorable designation. The proposed directive from the EEC specifies acceptable maximum levels of sweeteners that may be used as food additives to give specific foods a sweet taste. For xylitol, label designation E 967, the proposed level is quantum satis (46). In the United States, it is approved as a food additive for special dietary or nutritional uses as long as the amount used does not exceed that needed to produce the intended effect (60). It is approved for pharmaceutical use in many countries, including Austria, Canada, Denmark, Germany, Italy, Norway, Portugal, Spain, Japan, and Switzerland. However, it is not approved for pharmaceutical use in the US or UK (11).

History

Xylitol is found naturally in small amounts in a variety of fruits and vegetables. It is also formed naturally in the body as an intermediate in glucose metabolism through the glucuronate cycle in the liver (11,31). It was first synthesized and described by Emil Fischer and his associates in 1891 (11).

Documents

Sweetners State of Knowledge Review