Structure-Dependent Activity of Plant-Derived Sweeteners

molecules

Review Structure-Dependent Activity of Plant-Derived Sweeteners

Serhat Sezai Ҫiçek

Department of Pharmaceutical Biology, Kiel University, Gutenbergstraße 76, 24118 Kiel, Germany; [email protected]; Tel.: +49-431-880-1077

Academic Editors: Wolfgang Sippl and Fidele Ntie-Kang Received: 31 March 2020; Accepted: 21 April 2020; Published: 22 April 2020

Abstract: Human sensation for sweet tastes and the thus resulting over-consumption of sugar in recent decades has led to an increasing number of people suffering from caries, diabetes, and obesity. Therefore, a demand for sugar substitutes has arisen, which increasingly has turned towards natural sweeteners over the last 20 years. In the same period, thanks to advances in bioinformatics and structural biology, understanding of the sweet taste receptor and its different binding sites has made significant progress, thus explaining the various chemical structures found for sweet tasting molecules. The present review summarizes the data on natural sweeteners and their most important (semi-synthetic) derivatives until the end of 2019 and discusses their structure–activity relationships, with an emphasis on small-molecule high-intensity sweeteners.

Keywords: natural product; plant origin; non-caloric sweeteners; mogrosides; stevia glycosides; phyllodulcin

1. Introduction Taste—or the gustatory system—is the essential evolutionary tool to evaluate the composition of foods before ingestion [1,2]. The five taste qualities generally considered to be basic tastes are sweet, sour, salty, bitter, and umami, the latter being typically elicited by L-glutamate [2,3]. Each of these five taste qualities plays an important role, either indicating high electrolyte concentrations (salty), spoiled food (sour), and poisonous plant metabolites (bitter), or giving nutritional information, such as the abundance of carbohydrates (sweet) or proteins (umami) [3,4]. Sweet and umami are the main attractive taste modalities and therefore called the two palatable tastes [5,6]. In particular, the sweet taste has a great relevance, as most people respond positively to the sensation of sweetness [7]. The propensity to sweet foods and the thus resulting over-consumption of sugar in industrial countries has led to a significant number of people suffering from caries, diabetes, and hyperlipidemia [7–9]. Consequently, in the last four decades a market for non-caloric sweeteners and dietary products has evolved, addressing the needs of more than a billion people [10]. Additionally, more and more consumers express their interest for natural ingredients, leading to an increasing demand for natural non-caloric sweeteners [10]. This demand is reflected by scientific publications, which show both an increase in papers dealing with sugar substitutes and a significant rise in publications on natural sweeteners (Figure1).

Molecules 2020, 25, 1946; doi:10.3390/molecules25081946 www.mdpi.com/journal/molecules Molecules 2020, 25, x 2 of 23

complementary pair of hydrogen bond donor and acceptor on the receptor, and that this feature would be the crucial interaction for binding [15]. Other models followed, applying either planar [16] or three-dimensional geometries, the latter deriving from newly found guanidine-based sweeteners [17]. However, due to the ongoing discovery of natural sweeteners with distinct chemical structures, the idea of a general model to predict and quantify the sweetness of compounds was abandoned, now assuming that different classes of sweet molecules might interact with different receptor types [18]. Still, the model of Schallenberger and Acree was furthermore used to describe the linkage Molecules 2020between, 25, 1946 sweetness and structure, as demonstrated in recent rotational studies on sugars and artificial 2 of 23 sweeteners combining laser ablation with Fourier-transform microwave spectroscopy [19–21].

Figure 1.FigureNumber 1. Number of publications of publications with with the the topic topic “sugar substitute” substitute” and and“natur “naturalal sweetener” sweetener” according according to the Web of Science citation indexing service (core collection) from 1945 to 2019. to the Web of Science citation indexing service (core collection) from 1945 to 2019. The identification of the taste 1 receptor family, comprising three taste receptors (TAS1R1–3), Overrevealed the years, that indeed numerous only one compounds receptor type have is respon beensible identified for the sweet with taste, sweetening which is a properties, heterodimer either by phytochemicalof TAS1R2 approaches and TAS1R3 or organic [22,23]. synthesisFurthermore, [11 –the13 ].taste In manyfor umami cases, was natural attributed compounds to the or their glucophoreTAS1R1/TAS1R3 were taken heterodimer, as lead compound, thus showing such that as bo theth taste isovanillyl receptors group share a ofcommon (+)-phyllodulcin, subunit [5,23]. which is the sweetSame principle as other of class amacha C G Protein-co (sweetupled tea), areceptors regional (GPCRs), specialty the sweet in Japan taste[ receptor14]. (+)-Phyllodulcin is composed of and its a cytoplasmic region, a region constituted by seven transmembrane helices, and a large extracellular analogs were furthermore studied to understand the structure–activity relationships of the sweetening region [4,22]. In contrast to some other GPCRs, such as metobotropic glutamate receptors or γ- effect. Theaminobutyric first successful acid type model B receptors, was which established function by as Schallenbergerhomo- and heterodimers, and Acree, TAS1 receptors who hypothesized are that a hydrogenobligatory bond heterodimers donor [2]. and The a hydrogenextracellular bond domain acceptor contains werea Venus interacting fly trap domain with (VFD) a complementary and a pair of hydrogencysteine-rich bond domain donor (CRD), and with acceptor nine cysteine on the residues receptor, forming and that four thisdisulfide feature bonds would and a be fifth the crucial interactiondisulfide for binding bond with [15 the]. OtherVFD. Even models though followed, several GPCRs applying have either already planar been crystallized, [16] or three-dimensional a crystal structure for the sweet taste receptor is still missing [9,22]. However, the good sequence alignment geometries, the latter deriving from newly found guanidine-based sweeteners [17]. However, due to with metabotropic glutamate receptors 1 and 5 allowed a homology 3D-model of the sweet taste the ongoingreceptor discovery and a detailed of natural discussion sweeteners of the different with binding distinct sites chemical [9]. The structures,orthosteric binding the idea pockets of a general model toboth predict have a and volume quantify of about the 4900 sweetness Å³ in their of open compounds form and wasthus allow abandoned, small as nowwell as assuming large that differentsweeteners classes of to sweet bind to molecules the receptor. might Additionally, interact both with cavities different are hydrophilic, receptor types with 45% [18]. (TAS1R2) Still, the model of Schallenbergerand 50% (TAS1R3) and Acree surface was area furthermore being accessible used to to polar describe molecules. the linkageFurthermore, between each subunit sweetness and contains a transmembrane domain (TMD) binding pocket, with a volume of 210 Å³ for TAS1R2 and structure, as demonstrated in recent rotational studies on sugars and artificial sweeteners combining 270 Å³ for TAS1R3, respectively. With the TAS1R3 TMD binding pocket also being present in the laser ablationumami with receptor, Fourier-transform ligands such as the microwave sweetener cyclamate spectroscopy or the sweetness [19–21]. inhibitor lactisol have also Thebeen identification found to enhance of the or taste inhibit 1 receptorthe taste of family, glutamate comprising [24]. Both threebinding taste sites receptorsallow allosteric (TAS1R1–3), revealedmodulation that indeed of onlysmall onemolecules, receptor while type another is responsible allosteric binding for the site sweet is in the taste, CRD which and accessible is a heterodimer for of TAS1R2 andmacromolecules, TAS1R3 [22 such,23]. as Furthermore, the sweet-tasting the tasteproteins for brazzein, umami wasmonellin attributed or thaumatin to the TAS1R1[25]. The /TAS1R3 orthosteric binding sites in the VFD of both subunits can bind ligands, which is the case for e.g., heterodimer, thus showing that both taste receptors share a common subunit [5,23]. Same as other class C G Protein-coupled receptors (GPCRs), the sweet taste receptor is composed of a cytoplasmic region, a region constituted by seven transmembrane helices, and a large extracellular region [4,22]. In contrast to some other GPCRs, such as metobotropic glutamate receptors or γ-aminobutyric acid type B receptors, which function as homo- and heterodimers, TAS1 receptors are obligatory heterodimers [2]. The extracellular domain contains a Venus fly trap domain (VFD) and a cysteine-rich domain (CRD), with nine cysteine residues forming four disulfide bonds and a fifth disulfide bond with the VFD. Even though several GPCRs have already been crystallized, a crystal structure for the sweet taste receptor is still missing [9,22]. However, the good sequence alignment with metabotropic glutamate receptors 1 and 5 allowed a homology 3D-model of the sweet taste receptor and a detailed discussion of the different binding sites [9]. The orthosteric binding pockets both have a volume of about 4900 Å3 in their open form and thus allow small as well as large sweeteners to bind to the receptor. Additionally, both cavities are hydrophilic, with 45% (TAS1R2) and 50% (TAS1R3) surface area being accessible to polar molecules. Furthermore, each subunit contains a transmembrane domain (TMD) binding pocket, with a volume of 210 Å3 for TAS1R2 and 270 Å3 for TAS1R3, respectively. With the TAS1R3 TMD Molecules 2020, 25, 1946 3 of 23 binding pocket also being present in the umami receptor, ligands such as the sweetener cyclamate or the sweetness inhibitor lactisol have also been found to enhance or inhibit the taste of glutamate [24]. Both binding sites allow allosteric modulation of small molecules, while another allosteric binding site is in the CRD and accessible for macromolecules, such as the sweet-tasting proteins brazzein, monellin or thaumatin [25]. The orthosteric binding sites in the VFD of both subunits can bind ligands, which is the case for e.g., sucrose, but most sweeteners mainly interact with the binding site in the VFD of TAS1R2 [18]. Recent findings, furthermore, suggest that the binding site in the VFD of TAS1R3 plays an auxiliary role, showing less discriminating recognition characterized by loosely bound amino acids [6]. Interestingly, the signal transduction (G protein-coupling) is also carried out by the TMD of the TAS1R3 subunit, after having been transmitted from the VFD of TAS1R2, via the VFD of TAS1R3 and the CRD of TAS1R3, respectively [26].

2. Method Literature search was carried out using the Web of Science citation indexing service and the term “sweetener” in combination with the words “natural product”, “naturally occurring” or “plant-derived” resulting in 436 publications. Additionally, the terms “natural” and “sweeteners” were combined, and the results were reduced to the field of plant science and multidisciplinary chemistry, giving 155 hits. The publications were reviewed by title, abstract, and text, and reduced. Additionally, studies from other review articles were collected, which were not found by database search. Chemical structures were divided into compound classes (carbohydrates, amino acids, phenols, and terpenes) and, depending on the number compounds, into subclasses, thereby defining the structure of this review. Compound names (mostly trivial names) and configurations were taken “as is” from the original publications and species names of the natural sources used in the studies were checked using “The Plant List” [27]. The data discussed in this review is additionally summarized in two tables, giving an overview of natural high-intensity sweeteners (Table1) and those which were chemically modified (Table2). Here, compound number, name, and source are given, as well as relative sweetness (RS) values and the concentration of the sucrose solution used for comparison (if available). Furthermore, concomitant bitter tastes (or aftertastes) of the compounds are indicated and divided into slightly bitter (+), bitter (++) and very bitter (+++). (–) means that the compound was reported to exhibit no bitter taste, whereas blank fields indicate the lack of respective data.

Table 1. Natural high-intensity sweeteners.

cpd. Name Source RS c (Sucrose) Bitterness 16 monatin Schlerochiton ilicifolius 1400 5 – 17 (+)-phyllodulcin Hydrangea macrophylla 400 3 + 18 (+)-phyllodulcin 8-O-glucoside Hydrangea macrophylla 19 (+)-diyhdroquercetin 3-acetate Pluchea dodoneifolia 80 2 20 (+)-diyhdro-6-methoxy-kaempferol 3-acetate Tetraneuris turneri 25 2 21 (+)-diyhdro-6-methoxy-luetolin 3-acetate Tetraneuris turneri 15 2 22 (+)-diyhdro-6-methoxy-luteolin Tetraneuris turneri 20 2 23 glycyphyllin Smilax leucophylla ++ 24 phlorizin Symplocos lancifolia 25 trilobatin Symplocos microcalyx 26 naringin Citrus paradisi +++ 27 neohesperidin Citrus x aurantium ++ 28 (+)-haematoxylin Haematoxylum campechianum 120 3 29 selligueain A Selliguea feei 35 2 + 30 perillaldehyde Perilla frutescens 31 (+)-hernandulcin Phyla scaberrima 1000 ++ 32 (+)-4β-hydroxyhernandulcin Phyla scaberrima n.d. 33 mukurozioside IIb Sapindus rarak 1 34 steviolbioside Stevia rebaudiana 100 10 + 35 stevioside Stevia rebaudiana 210 0.6 + 36 rebaudioside E Stevia rebaudiana 174 + Molecules 2020, 25, 1946 4 of 23

Table 1. Cont.

cpd. Name Source RS c (Sucrose) Bitterness 37 rebaudioside B Stevia rebaudiana 150 10 + 38 rebaudioside A Stevia rebaudiana 242 + 39 rebaudioside D Stevia rebaudiana 221 + 40 rebaudioside C Stevia rebaudiana 30 + 41 rebaudioside F Stevia rebaudiana 200 + 42 dulcoside A Stevia rebaudiana 30 30 + 43 rubusoside Rubus chingii 114 + 44 suavioside B Rubus chingii 45 suavioside G Rubus chingii 46 suavioside H Rubus chingii 47 suavioside I Rubus chingii 48 suavioside J Rubus chingii 49 bayunoside Phlomoides betonicoides 500 50 gaudichaudioside A Baccharis gaudichaudiana 55 2 – 51 gaudichaudioside B Baccharis gaudichaudiana ++ 52 gaudichaudioside E Baccharis gaudichaudiana ++ 4β,10α-dimethyl-1,2,3,4,5,10- 53 Pinus sp. 1600 +++ hexahydroﬂuorene- 4α,6-dicarboxylic acid 54 glycyrrhizic acid Glycyrrhiza glabra 93–170 ++ 55 albiziasaponin A Albizia myriophylla 5 56 albiziasaponin B Albizia myriophylla 600 57 periandrin I Periandra mediterranea 93–170 + 58 periandrin III Periandra mediterranea 93–170 + 59 periandrin V Periandra mediterranea 200 + 60 periandrin II Periandra mediterranea 93–170 + 61 periandrin IV Periandra mediterranea 93–170 62 abrusoside A Abrus precatorius 30 – 63 abrusoside B Abrus precatorius 100 – 64 abrusoside C Abrus precatorius 50 – 65 abrusoside D Abrus precatorius 75 – 66 abrusoside E Abrus precatorius 67 cyclocarioside A Cyclocarya paliurus 200 68 cyclocarioside I Cyclocarya paliurus 250 69 pterocaryoside A Cyclocarya paliurus 50 2 + 70 pterocaryoside B Cyclocarya paliurus 100 2 + 71 mogroside III Siraitia grosvenorii – 72 mogroside IIIA1 Siraitia grosvenorii – 73 mogroside IIIA2 Siraitia grosvenorii – 74 mogroside IIIE Siraitia grosvenorii – 75 mogroside IVA Siraitia grosvenorii – 76 mogroside IVE Siraitia grosvenorii 392 1 – 77 siamenoside I Siraitia grosvenorii 563 1 – 78 grosmomoside Siraitia grosvenorii – 79 mogroside V Siraitia grosvenorii 425 1 – 80 isomogroside V Siraitia grosvenorii – 81 neomogroside Siraitia grosvenorii – 82 mogroside VI Siraitia grosvenorii – 83 mogroside VIA Siraitia grosvenorii – 84 mogroside VIB Siraitia grosvenorii – 85 7-oxomogroside IIIE Siraitia grosvenorii 86 7-oxomogroside IV Siraitia grosvenorii 87 7-oxomogroside V Siraitia grosvenorii 88 osladin Polypodium vulgare 500 + 89 polypodoside A Polypodium glycyrrhiza 600 6 + 90 polypodoside B Polypodium glycyrrhiza 91 extensumside C Myriopteron extensum 400 1 92 extensumside E Myriopteron extensum 200 1 93 extensumside F Myriopteron extensum 200 1 94 extensumside H Myriopteron extensum 200 1 95 extensumside D Myriopteron extensum 300 1 96 extensumside G Myriopteron extensum 100 1 97 extensumside I Myriopteron extensum 150 1 98 extensumside J Myriopteron extensum 100 1 99 extensumside K Myriopteron extensum 50 1 100 extensumside L Myriopteron extensum 50 1 Molecules 2020, 25, 1946 5 of 23

Table 2. Synthetic plant-derived high-intensity sweeteners.

cpd. Name Derived From RS c (Sucrose) Bitterness 5a sucralose sucrose 17a 2-(3-hydroxy-4-methoxypheyl)-1,3-benzodioxan phyllodulcin 3000 17b 2-(3-hydroxy-4-methoxypheyl)-1,4-benzodioxan phyllodulcin 450 17c (+)-2-(3-hydroxy-4-methoxypheyl)-1,3-benzoxathian phyllodulcin 18000 17d (+)-2-(3-hydroxy-4-methoxypheyl)-1,3-benzodithian phyllodulcin 20000 (+)-diyhdroquercetin 19a 4 -methyldihydro-quercetin 3-acetate 400 2 – 0 3-acetate 26a naringin dihydrochalcone naringin 300 5 27a neohesperidin dihydrochalcone neohesperidin 1000 5 + 27b 30carboxy-hesperetin neohesperidin 3400 6 + (6aS,11aS)-8-methoxy-11a-methyl-6,11- 28a (+)-haematoxylin 50 3 dihydroindeno[1,2-c]chromene-6a,9-diol 30a perillartine perillaldehyde 370 ++ 4-(methoxymethyl)-1,4-cyclohexadiene-1- 30b perillaldehyde 450 carboxaldehyde syn-oxime 35a steviolbioside sulfopropyl ester sodium salt stevioside 160 + 35b steviolbioside sulfoprop-2-yl ester sodium salt stevioside 120 – 38a rebaudioside B sulfopropyl ester sodium salt rebaudioside A 170 – 43a 13-O-β-maltotriosyl-19-O-β-d-glucosylsteviol rubusoside 298 – 66a abrusoside E 60’-O-methyl ester abrusoside E 150

3. Plant-Derived Sweeteners

3.1. Carbohydrates

3.1.1. Sugars Carbohydrates are an important dietary source, with mono- and disaccharides providing fast energy. However, not only their content on calories made this compounds valuable to humans, also their sweet taste let them collect and culture diﬀerent forms of this nutrients for thousands of years [11]. First historical records were found for honey, which consist largely of D-glucose (1) and D-fructose (2), the two monomers of the sucrose disaccharide (5) (Figure2)[ 11,28]. D-glucose (1) only has RS of 0.5 to 0.8 compared to sucrose (5), while the RS of D-fructose (2) is ranging from 1.1 to 1.7, making it the sweetest naturally occurring sugar [11,29]. The mixture of the two monosaccharides, which is called “invert sugar”, is about 0.8 times as sweet as sucrose (5), while honey itself is sweeter than sugar, because of higher content of D-fructose (2) (32 to 38%) compared to D-glucose (1) (28 to 31%) [28,30]. Apart from 1 and 2, additional monosaccharides are known to exhibit a sweet taste, such as D-allulose (3), which is also known as D-psicose, and D-tagatose (4)[31,32]. Showing RS values of 0.7 and 0.9, respectively, and being generally recognized as safe (GRAS) by the Food and Drug Administration, bothMolecules compounds 2020, 25, x can be used as sugar substitutes. 6 of 23

FigureFigure 2. 2.Chemical Chemical structuresstructures of sweet-tasting sugars sugars and and derivatives. derivatives.

Sucrose (5), also known as saccharose, or table sugar, has been used to flavor food for centuries; being first harvested from sugarcane it is meanwhile also obtained from the sugar beet [11]. Despite its many advantages, such as its easy availability from natural sources or its use as preservative, consumption of high amounts of sucrose (5) can cause health problems, such as caries, obesity, or diabetes, and therefore led to the search for alternative sweeteners, either from plant origin or by (semi)synthetic synthesis [11,32,33]. One of the most used sweeteners is sucralose (5a), which is obtained by chlorination of sucrose (5) and exhibits a similar taste, but with an about 600 times higher intensity [32]. Though sucralose (5a) underwent extensive safety evaluations, there have been some concerns that the compound leads to a decrease in beneficial gut bacteria and to thus provoke adverse effect in the gastrointestinal tract [34,35]. Other sweet-tasting disaccharides used in food industry are maltose (6) and trehalose (7), though both compounds show less than half of the sweetness of sucrose (5) [27,36,37]. Regarding receptor interaction, glucose (1), sucrose (5), and sucralose (5a) in contrast to many other sweeteners, are known to bind to the orthosteric binding sites of both subunits [38]. Due to the structural similarities of the other saccharides, their relatively similar intensities, and the fact that the orthosteric binding sites are accessible to polar molecules, it can be assumed that these molecules show similar binding [9]. However, the reason for the much higher intensities of sucralose (5a) and other chlorinated sucrose derivatives seems to derive from the much higher affinity to the TAS1R3 subunit than its counterpart sucrose (5), while at the same time showing comparable affinities to the TAS1R2 binding site [38]. Another reason is that the orthosteric binding sites show only 45% and 50% polar surface, respectively, meaning that additional lipophilic groups might enhance the sweetening effect and contribute to the much higher intensities of sucralose (5a) and other chlorinated sucrose derivatives [9,39]. Here, in particular the chlorine atom at the C1′ of the furanose moiety showed favorable hydrophobic interactions [40]. A study on the conformational behavior of ketohexoses by new spectroscopic techniques and laser ablation methods, found that the most abundant sweet-tasting conformers all show the same intramolecular H-bond network, with the anomeric hydroxy-group always in axial orientation and the primary hydroxy-group pointing towards the cyclically bound oxygen [19]. Apart from the same conformational signature, these two hydrogen atoms were found to be exclusively responsible for the sweet taste. Therefore, the primary hydroxy-group is acting as proton donor and the axial hydroxy-group as proton acceptor, thus corroborating the theory of Schallenberger and Acree [15].

Molecules 2020, 25, 1946 6 of 23

Sucrose (5), also known as saccharose, or table sugar, has been used to flavor food for centuries; being first harvested from sugarcane it is meanwhile also obtained from the sugar beet [11]. Despite its many advantages, such as its easy availability from natural sources or its use as preservative, consumption of high amounts of sucrose (5) can cause health problems, such as caries, obesity, or diabetes, and therefore led to the search for alternative sweeteners, either from plant origin or by (semi)synthetic synthesis [11,32,33]. One of the most used sweeteners is sucralose (5a), which is obtained by chlorination of sucrose (5) and exhibits a similar taste, but with an about 600 times higher intensity [32]. Though sucralose (5a) underwent extensive safety evaluations, there have been some concerns that the compound leads to a decrease in beneficial gut bacteria and to thus provoke adverse effect in the gastrointestinal tract [34,35]. Other sweet-tasting disaccharides used in food industry are maltose (6) and trehalose (7), though both compounds show less than half of the sweetness of sucrose (5)[27,36,37]. Regarding receptor interaction, glucose (1), sucrose (5), and sucralose (5a) in contrast to many other sweeteners, are known to bind to the orthosteric binding sites of both subunits [38]. Due to the structural similarities of the other saccharides, their relatively similar intensities, and the fact that the orthosteric binding sites are accessible to polar molecules, it can be assumed that these molecules show similar binding [9]. However, the reason for the much higher intensities of sucralose (5a) and other chlorinated sucrose derivatives seems to derive from the much higher affinity to the TAS1R3 subunit than its counterpart sucrose (5), while at the same time showing comparable affinities to the TAS1R2 binding site [38]. Another reason is that the orthosteric binding sites show only 45% and 50% polar surface, respectively, meaning that additional lipophilic groups might enhance the sweetening effect and contribute to the much higher intensities of sucralose (5a) and other chlorinated sucrose derivatives [9,39]. Here, in particular the chlorine atom at the C10 of the furanose moiety showed favorable hydrophobic interactions [40]. A study on the conformational behavior of ketohexoses by new spectroscopic techniques and laser ablation methods, found that the most abundant sweet-tasting conformers all show the same intramolecular H-bond network, with the anomeric hydroxy-group always in axial orientation and the primary hydroxy-group pointing towards the cyclically bound oxygen [19]. Apart from the same conformational signature, these two hydrogen atoms were found to be exclusively responsible for the sweet taste. Therefore, the primary hydroxy-group is acting as proton donor and the axial hydroxy-group as proton acceptor, thus corroborating the theory of Schallenberger and Acree [15].

3.1.2. Sugar Alcohols Apart from mono- and disaccharides, also their reduced forms, the so-called sugar alcohols or polyols, can exhibit a sweet taste and are used as alternative sweeteners [10,11,32,33]. The smallest polyol is glycerol (8), though the compound is not used because of its sweetening eﬀect (Figure3)[ 7]. Erythritol (9) instead, is widely used in food industry, because of its high stability, its clean sweet taste and the lack of unpleasant aftertastes [10,33]. Its sweetness is reported to be about 0.6 to 0.7 times that of sucrose (5), but combination with other sweeteners can increase its sweetness by up to 30%. Other commonly used sugar alcohols are mannitol (10), sorbitol (11) and xylitol (12)[11,33]. While mannitol (10) and sorbitol (11) show RS values of 0.5 to 0.7, xylitol has a sweetness equal to that of sucrose (5) and thus is the sweetest sugar alcohol known. Maltitol (13) is a twelve-carbon sugar alcohol, which is also occurring in nature but same as other sugar alcohols is usually obtained by catalytic hydrogenation [11]. Its RS value is 0.9 and thus higher than that of most other sugar alcohols, but it also has a considerably higher glycemic index (35) than its congeners [11,33]. Molecules 2020, 25, x 7 of 23 taste and the lack of unpleasant aftertastes [10,33]. Its sweetness is reported to be about 0.6 to 0.7 times that of sucrose (5), but combination with other sweeteners can increase its sweetness by up to 30%. Other commonly used sugar alcohols are mannitol (10), sorbitol (11) and xylitol (12) [11,33]. While mannitol (10) and sorbitol (11) show RS values of 0.5 to 0.7, xylitol has a sweetness equal to that of sucrose (5) and thus is the sweetest sugar alcohol known. Maltitol (13) is a twelve-carbon sugar alcohol, which is also occurring in nature but same as other sugar alcohols is usually obtained by Moleculescatalytic2020 hydrogenation, 25, 1946 [11]. Its RS value is 0.9 and thus higher than that of most other sugar alcohols,7 of 23 but it also has a considerably higher glycemic index (35) than its congeners [11,33].

Figure 3. Chemical structures of sweet-tasting sugar alcohols.

Sugar alcoholsalcohols didifferffer from from their their saccharidic saccharidic counterparts counterparts by onlyby only two two protons, protons, which which results results from hydrogenationfrom hydrogenation of one of carbonyl one carbonyl group. group. Therefore, Therefor thesee, these compounds compounds can be can expected be expected to share to share the same the bindingsame binding site, whichsite, which is also is also reflected reflected by theirby their intensities. intensities. However, However, the the use use of of sugar sugar alcoholsalcohols does not so much result from their intense sweetness,sweetness, but from their similar taste to sucrose ( 5) and their additional values, e.g., protection of tooth decay in chewing gums [[11].11]. However, when it comes to their use as substitutes in dietary foods, their caloric values as well as their glycemic index must be taken into consideration. Apart Apart from from that, that, most most sugar sugar alcohols show low absorption in the upper gastrointestinal tract and are subsequently fermented by colon bacteria, leading to flatulenceflatulence and digestion problems [[33].33]. Same asas thethe abovementionedabovementioned ketohexoses, ketohexoses, sorbitol sorbitol (11 ()11 and) and the the linkage linkage between between its sweetness its sweetness and structureand structure were investigatedwere investigated in a rotational in a rotational study using study laser using ablation laser and ablation Fourier-transform and Fourier-transform microwave spectroscopymicrowave spectroscopy [20]. Interestingly, [20]. Interestingly, in all three observed in all three conformers observed the conformers hydroxy-groups the hydroxy-groups at positions 2 toat 6positions were involved 2 to 6 inwere a circular involved intramolecular in a circular hydrogen intramolecular bond network,hydrogen whereas bond network, the hydroxy-group whereas the at positionhydroxy-group 1 showed at noposition interaction 1 showed with any no partinteraction of the molecule. with any Regarding part of the the molecule. model of Regarding Schallenberger the andmodel Acree of Schallenberger and the abovementioned and Acree studyand the on abovemen sweet-tastingtioned ketohexoses, study on sweet-tasting the hydroxy-group ketohexoses, at position the 1hydroxy-group again fulfils the at roleposition of the 1 protonagain fulfils donor, the while role theof the hydroxy-group proton donor, at while position the 2hydroxy-group serves as proton at acceptor,position 2 the serves latter as being proton accomplished acceptor, the by latter the intramolecular being accomplished hydrogen by the bonds intramolecular and the thus hydrogen resulting dispositionbonds and the of thethus oxygen resulting atom disposit [15,19ion,20]. of the oxygen atom [15,19,20].

3.2. Amino Acids Most naturally occurring amino acids are tasteless or bitter, except alanine and glycine ( 14)) (Figure4 4))[ [41].41]. In In fact, fact, the the word word glycine glycine even even deri derivesves from from the theGreek Greek word word of sweet of sweet [7]. Of [7 ].the Of bitter the bitter amino acids, interestingly, their unnatural D-forms show a sweet taste, such as d-phenylalanine, amino acids, interestingly, their unnatural D-forms show a sweet taste, such as D-phenylalanine, D- d-tyrosine, or d-tryptophan (15)[41]. Additionally, there are several sweet synthetic or natural tyrosine, or D-tryptophan (15) [41]. Additionally, there are several sweet synthetic or natural peptides, such as the artificialartificial sweetener aspartame or monatin ( 16), which is the firstfirst natural high potency sweetener to be presented in this review [[42–44].42–44]. Monatin ( 16) was isolated from the roots of Schlerochiton ilicifolius,, a South African species ofof thethe AcanthaceaeAcanthaceae familyfamily inin 1992.1992. Evaluation of the compound sweetening eeffectffect with 5 and 10% ( w/v) sucrose solutions gave RS values of of 1400 1400 and and 1200, 1200, respectively, and thusthus aa promisingpromising newnew naturalnatural sweetener.sweetener. Even more so, as in subsequent studies three out of four diastereomers were found to be sweet and chemo-enzymatic approaches for the synthesis of the more potent enantiomers were established [45,46]. MoleculesMolecules 2020 2020, ,25 25, ,x x 88 ofof 2323 threeMoleculesthree outout2020 ofof, 25 fourfour, 1946 diastereomersdiastereomers werewere foundfound toto bebe sweetsweet andand chemo-enzymaticchemo-enzymatic approachesapproaches forfor8 ofthethe 23 synthesissynthesis ofof thethe moremore potentpotent enantiomersenantiomers werewere establishedestablished [45,46].[45,46].

FigureFigure 4. 4. ChemicalChemical structures structures of of sweet-tas sweet-tastingsweet-tastingting amino amino acids acids and and monatin.monatin.

D-tryptophanD-tryptophan ((1515)) was was found foundfound to toto bind bindbind to toto the thethe same samesame binding binding site sitesite as as aspartame aspartame and and thus thus exclusively exclusively tototo the the VFD VFD of of the the TAS1R2 TAS1R2 subunitsubunit [[26,40].[26,40].26,40]. Monatin Monatin ( ((1616)) is is a a dipeptidedipeptide consisting consisting of of tryptophan tryptophan and and glycine,glycine, which which is is also also part part of of the the synthetic synthetic dipe dipeptidedipeptideptide aspartame, aspartame, and and therefore therefore binding binding to to the the same same bindingbinding pocket pocket must must bebe assumed.assumed.

3.3.3.3. Phenols Phenols 3.3.1. Phyllodulcin and Derivatives 3.3.1.3.3.1. PhyllodulcinPhyllodulcin andand DerivativesDerivatives The next high potency sweetener to be discussed is (+)-phyllodulcin (17) (Figure5), which is TheThe nextnext highhigh potencypotency sweetenersweetener toto bebe discusseddiscussed isis (+)-phyllodulcin(+)-phyllodulcin ((1717)) (Figure(Figure 5),5), whichwhich isis thethe the sweetening principle of amacha, a traditional Japanese preparation [14]. Amacha means “sweet sweeteningsweetening principleprinciple ofof amacha,amacha, aa traditionaltraditional JapaneJapanesese preparationpreparation [14].[14]. AmachaAmacha meansmeans “sweet“sweet tea”tea” tea” and is obtained by fermenting leaves of Hydrangea macrophylla (Hydrangeaceae), which leads to andand isis obtainedobtained byby fermentingfermenting leavesleaves ofof HydrangeaHydrangea macrophyllamacrophylla (Hydrangeaceae),(Hydrangeaceae), whichwhich leadsleads toto thethe the enzymatic hydrolyses of (+)-phyllodulcin-8-O-β-d-glucoside (18) into the much sweeter aglycone enzymaticenzymatic hydrolyseshydrolyses ofof (+)-phyllodulcin-8-(+)-phyllodulcin-8-OO--ββ--DD-glucoside-glucoside ((1818)) intointo thethe muchmuch sweetersweeter aglyconeaglycone 1717.. 17. Having already been isolated in 1929, (+)-phyllodulcin (17) served as lead compound for the HavingHaving alreadyalready beenbeen isolatedisolated inin 1929,1929, (+)-phyllodulcin(+)-phyllodulcin ((1717)) servedserved asas leadlead compoundcompound forfor thethe development of new sweeteners and for models to describe their structure–activity relationships [47]. developmentdevelopment of of newnew sweetenerssweeteners and and forfor models models toto dedescribescribe theirtheir structure–activity structure–activity relationshipsrelationships [47].[47]. These studies focused on the isovanillyl group, which is responsible for the sweetening eﬀect and TheseThese studiesstudies focusedfocused onon thethe isovanillylisovanillyl group,group, whwhichich isis responsibleresponsible forfor thethe sweeteningsweetening effecteffect andand resulted in the synthesis of several derivatives [48]. First attempts revealed two dioxane derivatives resultedresulted inin thethe synthesissynthesis ofof severalseveral derivativesderivatives [4[48].8]. FirstFirst attemptsattempts revealedrevealed twotwo dioxanedioxane derivativesderivatives (17a and 17b), of which compound 17a showed an RS value of 3000 but low stability in aqueous ((17a17a andand 17b17b),), ofof whichwhich compoundcompound 17a17a showedshowed anan RSRS valuevalue ofof 30003000 butbut lowlow stabilitystability inin aqueousaqueous media [13]. Compound 17b instead remained stable, but the sweetening eﬀect decreased to “only” 450 mediamedia [13].[13]. CompoundCompound 17b17b insteadinstead remainedremained stable,stable, butbut thethe sweeteningsweetening effecteffect decreaseddecreased toto “only”“only” times that of glucose. However, subsequent synthesis of oxathiane (17c) and dithiane (17d) derivatives 450450 timestimes thatthat ofof glucose.glucose. However,However, subsequentsubsequent synthesissynthesis ofof oxathianeoxathiane ((17c17c)) andand dithianedithiane ((17d17d)) with RS values of 18,000 and 20,000, for the sweet (R)-enantiomer. derivativesderivatives withwith RSRS valuesvalues ofof 1818,000,000 andand 20,000,20,000, forfor thethe sweetsweet ((RR)-enantiomer.)-enantiomer.

FigureFigure 5. 5. ChemicalChemical structures structures of of sweet sweet dihydroisocoumarins dihydroisocoumarins andand derivatives.derivatives.

NeohesperidinNeohesperidin dihydrochalcone, dihydrochalcone, a a commonly commonly us useduseded semi-synthetic semi-synthetic sweetener, sweetener, was was found found to to interactinteractinteract with withwith the thethe TMD TMDTMD binding bindingbinding pocket pocketpocket in inin the thethe TAS1R3 TAS1R3TAS1R3 subunit subunit and and thus thus on onon the thethe samesame sitesite asas cyclamatecyclamate andand the the sweetness sweetness inhibitor inhibitor lactisollactisol [49].[[49].49]. Because Because neohesperidin neohesperidin dihydrochalc dihydrochalconedihydrochalconeone belongsbelongs toto thethe classclass of isovanillyl derivatives, also (+)-phyllodulcin (17) and its analogs are expected to bind to this site.

Molecules 2020, 25, x 9 of 23 Molecules 2020, 25, 1946 9 of 23 of isovanillyl derivatives, also (+)-phyllodulcin (17) and its analogs are expected to bind to this site. The allosteric receptor modulation via this binding site furthermore explains the extremely high RS Thevalues allosteric of some receptor of the synthesized modulation compounds. via this binding site furthermore explains the extremely high RS values of some of the synthesized compounds. 3.3.2. Flavanonols 3.3.2. Flavanonols Another compound class which yielded sweet-tasting constituents is the class of flavanonols, also Anotherreferred compoundto as diyhdroflavonols class which yielded(Figure sweet-tasting 6). Of this compound constituents class is the so classfar, four of flavanonols, natural sweet- also referredtasting molecules to as diyhdroflavonols have been isolated. (Figure (+)-Dihydroquercetin6). Of this compound acetate class so(19 far,) was four isolated natural from sweet-tasting the young moleculesshoots of Tessaria have been dodoneifolia isolated. (Asteraceae), (+)-Dihydroquercetin which was acetate meanwh (19)ile was classified isolated frominto the the genus young Pluchea shoots of[50].Tessaria The compound dodoneifolia showed(Asteraceae), an RS value which of was 80 meanwhilecompared to classified a 2% (w/v) into sucrose the genus solution.Pluchea However,[50]. The a compoundsynthetic analog showed (19a an) showing RS value a of 4′ 80-methoxy-group compared to a and 2% (thusw/v) an sucrose isovanillyl solution. moiety, However, exhibited a synthetic a RS of analog400. Because (19a) showing only the aracemic 40-methoxy-group form was synthesized and thus an it isovanillyl can be assumed moiety, that exhibited the (+)-enantiomer a RS of 400. Because shows onlytwice the the racemic intensity. form Further was synthesized flavanonols it can (20–22 be assumed) have thatbeen the isolated (+)-enantiomer from Hymenoxys shows twice turneri the intensity.(Asteraceae), Further which flavanonols now belongs (20–22 to the) have genus been Tetraneuris isolated from, but Hymenoxystheir RS values turneri were(Asteraceae), only in the whichrange nowof 15 belongsto 25 [51]. to the genus Tetraneuris, but their RS values were only in the range of 15 to 25 [51].

Figure 6. Chemical structures of sweet-tasting ﬂavanonols.flavanonols. 3.3.3. Dihydrochalcones 3.3.3. Dihydrochalcones The class of dihydrochalcones contains three semi-synthetic high potent sweeteners (26a, 27a, The class of dihydrochalcones contains three semi-synthetic high potent sweeteners (26a, 27a, and 27b), though their precursor molecules are ranging from slightly sweet (23–25) to even bitter and 27b), though their precursor molecules are ranging from slightly sweet (23–25) to even bitter molecules (26 and 27) (Figure7). Compounds 23 to 25 are naturally occurring dihydrochalcones, molecules (26 and 27) (Figure 7). Compounds 23 to 25 are naturally occurring dihydrochalcones, soso-called phloretin derivatives. The ﬁrst of these compounds, glycyphyllin (23) was isolated from called phloretin derivatives. The first of these compounds, glycyphyllin (23) was isolated from Smilax Smilax glycyphylla (Smilaccaceae), which is meanwhile named Smilax leucophylla, and was reported to glycyphylla (Smilaccaceae), which is meanwhile named Smilax leucophylla, and was reported to have have a bittersweet taste, similar to that of licorice [11]. Phlorizin (24) was obtained from Symplocos a bittersweet taste, similar to that of licorice [11]. Phlorizin (24) was obtained from Symplocos lancifolia lancifolia (Symplocaceae), while trilobatin (25) was isolated form Symplocos microcalyx, a species with (Symplocaceae), while trilobatin (25) was isolated form Symplocos microcalyx, a species with an an unresolved taxonomic status [52,53]. unresolved taxonomic status [52,53].

Molecules 2020, 25, 1946 10 of 23 Molecules 2020, 25, x 10 of 23

FigureFigure 7.7. Chemical structures ofof sweet-tastingsweet-tasting dihydrochalconesdihydrochalcones andand theirtheir precursors.precursors.

TheThe sweeteningsweetening potentialpotential ofof naringeninnaringenin dihydrochalconedihydrochalcone ((26a26a)) andand especiallyespecially neohesperidinneohesperidin dihydrochalconedihydrochalcone ( 27a)) was was already already detected detected in in1969, 1969, when when the therespective respective flavanones flavanones (26 and (26 27and) were27) weretreated treated with withhot hotalkali alkali [54]. [54 While]. While compound compound 26a26a showsshows an an RS RS value value of of 300,300, neohesperidinneohesperidin dihydrochalconedihydrochalcone ( 27a(27a)) shows shows a a relative relative sweetness sweetness of of 1000 1000 compared compared to ato 5% a 5% (w/v (w/v)) sucrose sucrose solution. solution. The higherThe higher intensity intensity of compound of compound27a derives 27a derives from the from isovanillyl the isovanillyl feature, which feature, is missing which inis naringeninmissing in dihydrochalconenaringenin dihydrochalcone (26a). Though (26a neohesperidin). Though neohesperidin dihydrochalcone dihydrochalcone is used as sweetener is used as in sweetener a wide range in a ofwide foods range and beveragesof foods and it has beverages some limitations, it has some such lim as aitations, slow onset such and as aa lingering slow onset aftertaste and a [12lingering,55,56]. Toaftertaste improve [12,55,56]. these temporal To improve deficits these different temporal modification deficits different of neohesperidin modification have of been neohesperidin carried out, withhave onebeen compound, carried out, 30 -carboxyhesperetinwith one compound, dihydrochalcone 3′-carboxyhesperetin (27b), even dihydrochalcone showing a higher (27b RS), even 3400 showing compared a tohigher a 6% sucroseRS 3400 solution compared [57]. to However, a 6% sucrose the characteristic solution [57]. lingering However, aftertaste the wascharacteristic not improved. lingering aftertasteSame was as above, not improved. the isovanillyl feature of compounds 27a and 27b was responsible for the much higherSame sweetening as above, e fftheect isovan comparedillyl feature to compounds of compounds lacking 27a this and feature 27b was (23–26a responsible). Compared for the tomuch the previouslyhigher sweetening discussed effect phenolic compared compound to compounds classes, the lacking dihydrochalcone this feature type(23–26a seems). Compared to stimulate to thethe allostericpreviously binding discussed pocket phenolic to a much compound higher extentclasses, than the otherdihydrochalcone plant-derived type phenols. seems to stimulate the allosteric binding pocket to a much higher extent than other plant-derived phenols. 3.3.4. Condensed Phenols 3.3.4. Condensed Phenols The last class of phenolic compounds to be discussed consists of two condensed phenols (28 and 29The) and last oneclass synthetic of phenolic analog compounds (28a) (Figure to be 8discussed). The first consists compound of two is condensed ( +)-haematoxylin phenols ( (2828 and), a constituent29) and one of Haematoxylumsynthetic analog campechianum (28a) (Figure(Fabaceae) 8). The and first shows compound an RS value is (+)-haematoxylin of 120 compared to(28 a), 3% a (constituentw/v) sucrose of solution Haematoxylum [58]. To campechianum improve the sweetening (Fabaceae) eandffect shows a synthetic an RS analog value wasof 120 prepared, compared which to a displayed3% (w/v) sucrose an isovanillyl solution feature [58]. onToone improve of the the two sweeten aromaticing rings. effect However, a synthetic the relativeanalog was sweetness prepared, did notwhich alter displayed significantly, an isovanillyl now showing feature an RS on value one ofof 50 the for two the racemate.aromatic rings. However, the relative sweetness did not alter significantly, now showing an RS value of 50 for the racemate.

Molecules 2020, 25, 1946 11 of 23

MoleculesMolecules 20202020,, 2525,, xx 1111 ofof 2323

FigureFigure 8. 8. ChemicalChemical structures structures of of sweet-tasting sweet-tasting condensed condensed phenols. phenols.

TheThe second second condensed condensed phenol phenol is is aa proanthocyanidineproanthocyanproanthocyanidineidine isolated isolated from from thethe IndonesianIndonesian medicinalmedicinal plantplant SelligueaSelliguea feeifeeifeei (Polypodiaceae) (Polypodiaceae)(Polypodiaceae) [ 59[59].[59].]. The TheThe bittersweet bittersweetbittersweet rhizomes, rhrhizomes,izomes, which whichwhich are areare used usedused for forfor the thethe treatment treatmenttreatment of ofrheumatismof rheumatismrheumatism and andand as tonic,asas tonic,tonic, yielded yieldedyielded selligueain selligueainselligueain A (29 AA), ((2929 which),), whichwhich showed showedshowed a RS aa of RSRS 35, ofof when 35,35, whenwhen compared comparedcompared to a 2% toto a(aw 2%2%/v) ( sucrose(ww//vv)) sucrosesucrose solution. solution.solution. Structure–activityStructure–activity relationships relationships of of these these three three compoundscompounds areare didifficult.difficult.fficult. Due Due to to their their phenolicphenolic naturenature and and it it must must be be assumed assumed that that they they share share th thethee samesame bindingbinding pocket pocket as as thethe previouslypreviously discussed discussed phenols.phenols. However, However, the the fact fact that that compound compound 28a28a diddid not not show show significantly significantlysignificantly higher higher RS RS values values than than its its congenercongener as as well well as as the the size size of of compound compound 2929 somehowsomehow contradict contradict this this theory. theory. On On the the contrary, contrary, the the relativelyrelatively low low intensities intensities observed observed for for these these molecules molecules in in comparison comparison to to e.g., e.g., diyhdrochalcones diyhdrochalcones could could hinthint towards towards the the same same bindingbinding sitesite butbut muchmuch lowerlower eefficacies.efficacies.fficacies.

3.4. Monoterpenes 3.4.3.4. MonoterpenesMonoterpenes 30 TheThe classclass ofofof monoterpenes, monoterpenes,monoterpenes, so soso far, far,far, only onlyonly yielded yieldeyielde onedd oneone natural naturalnatural sweetener, sweetener,sweetener, which whichwhich is perillaldehyde isis perillaldehydeperillaldehyde ( ) Perilla ((Figure(3030)) (Figure(Figure9). The 9).9). compound TheThe compoundcompound is the isis sweet thethe sweetsweet principle principleprinciple of perilla ofof perillaperilla oil, which oil,oil, whichwhich is obtained isis obtainedobtained by distillation byby distillationdistillation of ofof frutescens PerillaPerilla frutescensfrutescens(Lamiaceae) (Lamiaceae)(Lamiaceae) [11]. The [11].[11]. compound TheThe compoundcompound itself, same itself,itself, as samesame the oil, asas only thethe oil, showsoil, onlyonly a slightlyshowsshows aa sweet slightlyslightly taste, sweetsweet but syn taste,thetaste, respective butbut thethe respectiverespective-oxime, synsyn perillartine-oxime,-oxime, perillartineperillartine (30a) possesses ((30a30a)) possesses apossesses RS of about aa RSRS 370 ofof about toabout that 370370 of sucrose toto thatthat ofof (5 )[sucrosesucrose11,60, 61 ((55].)) [11,60,61].However,[11,60,61]. perillartine However,However, perillartineperillartine (30a) has an ((30a appreciable30a)) hashas anan appreciableappreciable bitterness and bitternessbitterness a low water andand solubility,aa lowlow waterwater thus solubility,solubility, restricting thusthus its restrictinguserestricting for many itsits useuse applications forfor manymany [applicationsapplications61]. Further [61].[61]. studies FurtherFurther focusing studiesstudies on focusing perillartinefocusing onon analogsperillartineperillartine yielded analogsanalogs compound yieldedyielded compound30bcompound, which showed30b30b,, whichwhich an improvedshowedshowed anan RS improvedimproved of 450 times RSRS toofof that 450450 oftimestimes sucrose toto thatthat (5) coupledofof sucrosesucrose with ((55)) superior coupledcoupled water withwith superiorsolubilitysuperior waterwater [61]. solubilitysolubility [61].[61].

FigureFigure 9. 9. ChemicalChemical structures structures of of sweet-tasti sweet-tastingsweet-tastingng monoterpenes monoterpenes and and derivatives. derivatives.

ApartApart from from the the TMD TMD binding binding pocket pocket of of the the TAS1 TAS1R3TAS1R3R3 subunit, subunit, also also the the TAS1R2 TAS1R2 subunit subunit has has an an allostericallosteric binding binding sitesite [9,26].[9,26]. [9,26]. TheThe The latterlatter latter isis is saidsaid said toto to havehave have aa a volumevolume volume ofof of aboutabout about 210210 210 Å³Å³ Å and,3and,and, moreover,moreover, moreover, hashas has aa surfaceasurface surface areaarea area ofof of 38%38% 38% accessibleaccessible accessible toto to polarpolar polar moleculesmolecules molecules [9].[9]. [9]. Thus,Thus, Thus, onlyonly only veryvery very smallsmall and and less less hydrophilichydrophilic moleculesmolecules are are known known toto bindbind toto thisthis region,region, suchsuch asas ((+)-perillartin(+)-perillartin+)-perillartin (30a30a)) and and its its analogs. analogs.

3.5.3.5. SesquiterpenesSesquiterpenes AnotherAnother twotwo sweet-tastingsweet-tasting volatilevolatile oiloil constituentsconstituents belongbelong toto thethe classclass ofof sesquiterpenessesquiterpenes andand areare namednamed (+)-hernandulcin(+)-hernandulcin ((3131)) andand (+)-4(+)-4ββ-hydroxyhernandulcin-hydroxyhernandulcin ((3232)) (Figure(Figure 10)10) [62,63].[62,63]. HavingHaving beenbeen

Molecules 2020, 25, 1946 12 of 23

3.5. Sesquiterpenes Another two sweet-tasting volatile oil constituents belong to the class of sesquiterpenes and Molecules 2020, 25, x 12 of 23 are named (+)-hernandulcin (31) and (+)-4β-hydroxyhernandulcin (32) (Figure 10)[62,63]. Having been isolated from Lippia dulcis (Verbenaceae), which was later reclassiﬁed as Phyla scaberrima, isolated from Lippia dulcis (Verbenaceae), which was later reclassified as Phyla scaberrima, (+)- (+)-hernandulcin (31) was found to be about 1000 times as sweet as sucrose (5)[62]. Moreover, the hernandulcin (31) was found to be about 1000 times as sweet as sucrose (5) [62]. Moreover, the compound showed a bitter aftertaste, very low wa waterter solubility and was found to decompose upon β 31 heating [62,64]. [62,64]. Its natural derivative ((+)-4+)-4β-hernandulcin-hernandulcin ( (31)) also also showed showed a a sweet sweet taste taste and better water solubility, though a detailed characterizationcharacterization could not be performed due to low substance quantities [[63].63].

Figure 10. ChemicalChemical structures of sweet-tasting sesquiterpenes.

Another very interesting sesquiterpenoid, mukurozioside IIb ( 33) was isolated from the fruit of Sapindus rarak (Sapindaceae) [[63].63]. Evaluation Evaluation of of the comp compound’sound’s sweetening potential revealed that its sweetness was comparable to that of sucrose ( 5) and that the effect eﬀect derived from the high content (6.3%) in the fruit pulp. With regard to receptor binding, compounds 31 and 32 are likely to bind to the same site as the monoterpenes, as both compounds are of small size and show rather pronounced lipophilicity. The structure of compound 33, instead, indicates binding at the orthosteric site in the VFD of TAS1R2. As the linkage of sugar moietiesmoieties attachedattached remindsreminds of of other other speciﬁc specific TAS1R2 TAS1R2 binders, binders, such such as as steviosides, steviosides, it itis is likely likely that that also also compound compound33 33acts acts exclusively exclusively at at this this subunit subunit [26 [26].].

3.6. Diterpenes Diterpenes

3.6.1. Kauran-Type Kauran-Type Diterpenoids Stevia glycosides glycosides belong belong to to the the most most importan importantt natural natural sweeteners sweeteners and and currently currently are aremaybe maybe the mostthe most popular popular plant-derived plant-derived sugar sugar substitutes. substitutes. These These Kauran-type Kauran-type diterpenoids diterpenoids are present are present in the in leavesthe leaves of Stevia of Stevia rebaudiana rebaudiana (Asteraceae),(Asteraceae), which which has has been been used used for for centuries centuries by by the the indigenous population of Paraguay [12]. [12]. After After stevioside stevioside ( (3535)) was was first ﬁrst isolated isolated from from the the leaves leaves of of “Yerba “Yerba dulce” dulce” in 1931, additional glycosides were were discovered sinc sincee the 1970s, of which nine were studied for their sweetening properties properties ( (3434, ,36–4236–42) )(Figure (Figure 11) 11 )[[65–68].65–68 ].Of Ofthe the total total amount amount of glycosides, of glycosides, which which can varycan vary between between 4 and 4 and20% 20%of dried of dried leaves, leaves, stevioside stevioside (35) (and35) andrebaudioside rebaudioside A (38 A) (are38) present are present in the in highestthe highest concentrations concentrations [67,69]. [67 ,The69]. latter The lattercompound compound has the has GRAS the GRASstatus and status is marketed and is marketed as rebiana as [68,70].rebiana [It68 is,70 also]. It isthe also sweetest the sweetest among among the glycosides the glycosides and andsaid said to tohave have the the least, least, though though still considerable, bitterbitter andand lingering lingering aftertaste aftertaste [11 [11].]. To To get get rid rid of theseof these drawbacks drawbacks several several derivatives derivatives were prepared,were prepared, of which of which compound compound35a was 35a still was slightly still slightly bitter, whilebitter,compounds while compounds35b and 35b38a andwere 38a devoid were devoidof bitterness of bitterness [71]. However, [71]. However, their sweetness their sweetne was alsoss somewhatwas also somewhat decreased. decreased. Regarding theRegarding sweetening the potentialsweetening of potential the natural of the glycosides natural glycosides the compounds the compounds showing branchedshowing branched sugar moieties sugar moieties (37–39) were (37– 39sweeter) were than sweeter each than of their each non-branched of their non-branched counterparts counterparts (34–36). Within (34–36 each). Within group, each the componentgroup, the withcomponent a monoglucosidic with a monoglucosidic ester was the sweetest,ester was which the weresweetest, stevioside which ( 35were) and stevioside rebaudioside (35 A) (and38). rebaudioside A (38). Compounds containing an α-L-rhamnose moiety (40 and 42), showed significantly decreased RS values of 30, while substitution with a β-D-xylose moiety only slightly altered relative sweetness (41). The reported RS values of stevia glycosides are in line with a recent molecular docking study, which found the lowest binding energy – and thus highest sweetness – for

Molecules 2020, 25, 1946 13 of 23

Compounds containing an α-l-rhamnose moiety (40 and 42), showed signiﬁcantly decreased RS values of 30, while substitution with a β-d-xylose moiety only slightly altered relative sweetness (41). The reported RS values of stevia glycosides are in line with a recent molecular docking study, which found Molecules 2020, 25, x 13 of 23 the lowest binding energy—and thus highest sweetness—for rebaudioside A (38) and the highest rebaudiosidebinding energy A for(38 rebaudioside) and the highest C (40)[ binding72]. The authors,energy for furthermore, rebaudioside suggested C (40) the [72]. rhamnose The authors, moiety furthermore,as a good discriminator suggested the between rhamnose bitter moiety and sweet as a good tastes. discriminator between bitter and sweet tastes.

FigureFigure 11. 11. ChemicalChemical structures structures of of sweet-tas sweet-tastingting kauran-type kauran-type diterpenoids. diterpenoids.

AnotherAnother seriesseries of of kauran-type kauran-type glycosides glycosides (43–48 (43–48) was) isolated was isolated from Rubus from suavissimus Rubus suavissimus(Rosaceae), (Rosaceae),which is now which classiﬁed is now as Rubusclassified chingii as var.Rubussuavissimus chingii var.[73 suavissimus,74]. Rubusoside [73,74]. (43 Rubusoside) was found to(43 have) was a foundRS of aboutto have 115 a RS time of thatabout of 115 sucrose time (that5), but of sucrose a slightly (5 bitter), but a aftertaste. slightly bitter Semi-synthetic aftertaste. modiﬁcationSemi-synthetic of modificationrubusoside with of rubusoside a maltotriosyl with moiety a maltotriosyl (43a) increased moiety the (43a RS) increased and at the the same RS time and reducedat the same the bittertime reducedtaste [75 ].the Unfortunately, bitter taste [75]. RS valuesUnfortunately, of the other RS glycosidesvalues of the were other not obtained.glycosides However, were not onlyobtained. those However,compounds only with those glucose compounds moieties inwith positions glucose 13 moie and 19ties were in positions found sweet, 13 and namely 19 were suaviosides found sweet, B (44), namelyG(45), Isuaviosides (46), H (47), B and (44), J (G48 ()[4574), ].I (46), H (47), and J (48) [74].

3.6.2.3.6.2. Lanostan- Lanostan- and and Gibban-Type Gibban-Type Diterpenoids Diterpenoids Also,Also, lanostan-type triterpenoidstriterpenoids havehave been been found found to to potentially potentially exhibit exhibit a sweeta sweet taste, taste, though though not notas many as many as reported as reported for kauran-typefor kauran-type diterpenoids. diterpenoids. One One of these of these diterpenoids, diterpenoids, bayunoside bayunoside (49) ( was49) wasfound found to possess to possess a RS ofa aboutRS of 500about times 500 of times that ofof glucose that of (Figure glucose 12 (Figure)[11]. However, 12) [11]. theHowever, compound, the compound, which was isolated from Salvia digitaloides (Lamiaceae), meanwhile named Phlomoides betonicoides, furthermore showed a lingering aftertaste lasting for more than one hour and therefore restricting its use for many applications.

Molecules 2020, 25, 1946 14 of 23 which was isolated from Salvia digitaloides (Lamiaceae), meanwhile named Phlomoides betonicoides, furthermore showed a lingering aftertaste lasting for more than one hour and therefore restricting its useMolecules for many2020, 25 applications., x 14 of 23

Figure 12. Chemical structures of sweet-tasting lanostan- and gibban-type diterpenoids.

Sweet-tasting gaudiachaudiosidesgaudiachaudiosides A ( 50), B B ( 51), and E (52) were isolatedisolated fromfrom BaccharisBaccharis gaudichaudiana (Asteraceae) [64]. [64]. Of these compounds,compounds, gaudichaudioside A ( 50) showed a pleasantpleasant and sweet taste that was 55 times sweeter than a 2% (w/v) sucrose solution.solution. Gaudichaudioside B ( 51) and EE (52(52),), instead, instead, showed showed a sweet-bittera sweet-bitter taste, taste, while while additional additional compounds compounds of this of group this group were eitherwere neutraleither neutral or entirely or entirely bitter. bitter. The sweetsweet gibbangibban derivativederivative53 53was was isolated isolated from from pine pine tree tree rosin rosin and and was was found found to to exhibit exhibit a RS a RS of aboutof about 1600 1600 times times of of that that of of glucose, glucose, but but also also a a relative relative bitterness bitterness ofof 1515 timestimes of that of caffeine caﬀeine [76]. [76]. Because of the fact that only one ofof fourfour possiblepossible isomersisomers was found to bebe sweetsweet andand subsequentsubsequent reports on the hormone-like eeffectﬀect on plant growth,growth, the research on thisthis compoundcompound was not further conducted [[47].47].

3.7. Triterpenes

3.7.1. Oleanan-Type Triterpenoids 3.7.1. Oleanan-Type Triterpenoids Glycyrrhzicic acid (54), is the sweet principle in the roots of Glycyrrhiza glabra (Fabaceae), where it Glycyrrhzicic acid (54), is the sweet principle in the roots of Glycyrrhiza glabra (Fabaceae), where is contained in amounts of 6–14% as calcium, magnesium and potassium salt (Figure 13)[10,12]. The it is contained in amounts of 6–14% as calcium, magnesium and potassium salt (Figure 13) [10,12]. mixture of the diﬀerent salts is also referred to as glycyrrhizin, whereas the crude extract is known as The mixture of the different salts is also referred to as glycyrrhizin, whereas the crude extract is licorice [10]. The ammonium salt of glycyrrhizic acid (54) hast GRAS status for the use as ﬂavoring known as licorice [10]. The ammonium salt of glycyrrhizic acid (54) hast GRAS status for the use as compound (but not as sweetener) in the US, while the Committee of Food in the EU suggests to flavoring compound (but not as sweetener) in the US, while the Committee of Food in the EU consume not more than 100 mg per day, due to the possibility of pseudoaldosteronism [12]. The suggests to consume not more than 100 mg per day, due to the possibility of pseudoaldosteronism relative sweetness of glycyrrhizin compared to sucrose (5) is about 50, while for glycyrrhizic acid [12]. The relative sweetness of glycyrrhizin compared to sucrose (5) is about 50, while for glycyrrhizic (54) RS values of 93 to 170 have been reported [12,77]. Apart from its sweetness, glycyrrhizic acid acid (54) RS values of 93 to 170 have been reported [12,77]. Apart from its sweetness, glycyrrhizic acid (54) is characterized by a licorice aftertaste, and shows both a slow onset as well as long sweetness lingering [10], thus making it an ideal candidate for studying receptor kinetics. Attempts to alter the compound’s taste profile by modifying the saccharide units, have, so far, been unsuccessful [78,79].

Molecules 2020, 25, 1946 15 of 23

(54) is characterized by a licorice aftertaste, and shows both a slow onset as well as long sweetness lingering [10], thus making it an ideal candidate for studying receptor kinetics. Attempts to alter the

Moleculescompound’s 2020, 25 taste, x proﬁle by modifying the saccharide units, have, so far, been unsuccessful [7815, 79of ].23

FigureFigure 13. 13. ChemicalChemical structures structures of of sweet-tasti sweet-tastingng oleanan-type oleanan-type triterpenoids. triterpenoids.

StemsStems of of AlbiziaAlbizia myriophylla myriophylla (Fabaceae)(Fabaceae) are are used used in in traditional traditional medicine medicine in in Thailand Thailand and and Vietnam Vietnam andand were were found found to to contain contain sweet-tasting sweet-tasting oleanan-ty oleanan-typepe triterpenoids, triterpenoids, namely namely albiziasaponin albiziasaponin A A and and B B ((5555 andand 5656))[ [80].80]. While While albiziasaponin albiziasaponin A A ( (5555),), which which is is bearing bearing a a lactone lactone ring, ring, only only exhibited exhibited a a relative relative sweetnesssweetness ofof 55 compared compared to to sucrose sucrose (5), ( albiziasaponin5), albiziasaponin B (56 B), ( which56), which has a freehas carboxylica free carboxylic acid, showed acid, showedan RS value an RS of value 600. Other of 600. sweet-tasting Other sweet-tasting oleanan-type oleanan-type triterpenoids triterpenoids were isolated were fromisolated the rhizomesfrom the rhizomesof Periandra of Periandra dulcis (Lamiaceae), dulcis (Lamiaceae), which meanwhile which meanwhile was reclassified was reclassified as Periandra as Periandra mediterranea mediterranea[64,81]. [64,81].The plant, The which plant, is which commonly is commonly known as known Brazilian aslicorice Brazilian aff ordedlicorice five afforded sweet-tasting five sweet-tasting compounds, compounds,periandrin I toperiandrin V (57–61 ).I to While V (57–61 the sweetness). While the of sweetness periandrins of Iperiandrins to IV was comparable I to IV was comparable to glycyrrhizic to glycyrrhizicacid (54), periandrin acid (54 V), ( 59periandrin), which contains V (59), a terminalwhich contains xylose moiety a terminal instead xylose of glucuronic moiety acid instead showed of glucuronica RS of 200 acid times showed compared a RS to of sucrose 200 times (5)[ compared82]. to sucrose (5) [82]. ComparisonComparison of the eight oleanan-typeoleanan-type triterpenoidstriterpenoids ( 54–61(54–61)) makes makes clear clear that that the the carboxylic carboxylic acid acid is iscrucial crucial for thefor sweeteningthe sweetening effect, aseffect, demonstrated as demonstrated by the low by sweetness the low ofsweetness the lactone of albiziasaponin the lactone albiziasaponinA(55). In contrast, A (55 the). In keto-group contrast, inthe position keto-group 11 of in glycyrrhizic position 11 acid of glycyrrhizic (54) does not acid seem (54 essential) does not as seemnone ofessential the periandrins as none of (57–61 the periandrins) shows this ( feature,57–61) shows though this being feature, equally though sweet. being Within equally the group sweet. of Withinperiandrins, the group the one of periandrins, compound showing the one acompound xylose moiety showing (59) turneda xylose out moiety to be slightly(59) turned sweeter. out to This be slightlyeffect was sweeter. already This observed effect was for xylosealready bearing observed rebaudioside for xylose Fbearing (41), which rebaudioside was much F ( sweeter41), which than was its muchrhamnose sweeter bearing than analog its rhamnose rebaudioside bearing C (analog40). Another rebaudioside effect observed C (40). Another with the steviaeffect observed glycosides with was thethat stevia branching glycosides trisaccharide was that moieties branching were trisacchar sweeteride than moieties their disaccharidic were sweeter analogs. than their If thedisaccharidic branching analogs.sugar moiety If the is branching also the reason sugar formoiety the significantly is also the reason higher sweetnessfor the significantly of albiziasaponin higher sweetness B (56) cannot of albiziasaponin B (56) cannot be concluded, because additional features distinguish this compound, such as the hydroxy-group in position 24.

3.7.2. Cycloartan- and Dammaran-Type Triterpenoids Four sweet-tasting cycloartanoids, abrusosides A to D (62–65), were isolated from the leaves of Abrus precatorius (Fabaceae), showing RS values of 30, 100, 50, and 75, respectively (Figure 14) [83,84].

Molecules 2020, 25, 1946 16 of 23 be concluded, because additional features distinguish this compound, such as the hydroxy-group in position 24.

3.7.2. Cycloartan- and Dammaran-Type Triterpenoids Four sweet-tasting cycloartanoids, abrusosides A to D (62–65), were isolated from the leaves of Molecules 2020, 25, x 16 of 23 Abrus precatorius (Fabaceae), showing RS values of 30, 100, 50, and 75, respectively (Figure 14)[83,84]. Abrusoside BB ( 63(63),), which which was was the sweetestthe sweetest of the of four the compounds, four compounds, is bearing is abearing glucosylated a glucosylated glucuronic glucuronicacid methyl acid ester methyl moiety. ester This moiety. is interesting, This is interesting, because abrusoside because abrusoside E, which was E, which isolated was in laterisolated study in laterand showedstudy and only showed slight sweetness, only slight became sweetness, about be 150came times about sweeter 150 times than sucrosesweeter upon than semi-syntheticsucrose upon semi-syntheticesteriﬁcation of esterification its glucuronic of acid its glucuronic moiety [85 acid]. moiety [85].

O O R2 O OH O H

R R O 1 O OH O

67 R1=5-O-acetyl- -L-ara, R2= -L-rha 62 R= -D-glc 68 R1= -L-ara, R2= -D-qui 2 63 R= -D-glcA-6-CH3 - -D-glc 64 R= -D-glc2- -D-glc 65 R= -D-glcA2- -D-glc OH 66 R= -D-glc2- -D-glcA OR 2 66a R= -D-glc - -D-glcA-6-CH3 O OH HO

69 R= -D-qui 70 R= -L-ara Figure 14. ChemicalChemical structures structures of sweet-tasting cycloartan- and dammaran-type triterpenoids.

Phytochemical investigations on on Pterocarya paliurus (Juglandaceae), which which was was reclassified reclassiﬁed as Cyclocarya paliurus , yielded dammaran-type triterpenoids triterpenoids cyclocaryoside cyclocaryoside A A ( (6767)) and and I I ( 68) as well as 3,4-seco-dammaranes pterocaryosidespterocaryosides A A and and B B (69 (69and and70 )[7086) [86–88].–88]. While While the the seco-dammaranes seco-dammaranes69 and 69 and70 showed 70 showed a relative a relative sweetness sweetness of 50 and of 100 50 compared and 100 tocompared a 2% (w/v )to sucrose a 2% solution,(w/v) sucrose cyclocaryosides solution, A(cyclocaryosides67) and I (68) wereA (67 reported) and I with (68) RS were values reported of 200 andwith 250, RS respectively. values of Interestingly,200 and 250, compounds respectively.68 Interestingly,and 69 were found compounds to contain 68 and a β- 69d-quinovose were found moiety to contain in position a β-D-quinovose C-12, which moiety is a rarely in position encountered C-12, whichdeoxyhexose is a rarely in plants. encountered deoxyhexose in plants.

3.7.3. Cucurbitane-Type Cucurbitane-Type Triterpenoids Mogrosides (71–87 (71–87) )belong belong to tothe the most most intense intense and most and popular most popular triterpenoid triterpenoid sweeteners sweeteners (Figure 15)(Figure [12,68]. 15)[ These12,68]. compounds These compounds are the sweetening are the sweetening principle principle of Siraitia of Siraitia grosvenorii grosvenorii (Cucurbitaceae)(Cucurbitaceae) fruit, alsofruit, known also known as monk as fruit monk or fruit luo han or luo guo han [68]. guo The [68 total]. The amount total of amount mogrosides of mogrosides in the dried in thefruit dried was determinedfruit was determined with 3.8%, with while 3.8%, the while amount the of amount major oftriterpenoid major triterpenoid mogroside mogroside V (79) was V ( 79found) was to found vary betweento vary between 0.5 and 0.51.4% and [89]. 1.4% Having [89]. Havingbeen used been for used centuries for centuries in China in as China medicinal as medicinal plant and plant herbal and sweetener,herbal sweetener, extracts extracts containing containing mogroside mogroside V (79) are V ( 79presently) are presently used in used Japan in and Japan the and United the UnitedStates, where they are classified with the GRAS status [12,68,89]. Of the more than 50 triterpenoids isolated so far, only those bearing an α-hydroxy-group in position 11 and at least three sugar moieties exhibit a sweet taste [89–93]. Compounds without a hydroxy-group or with a keto-group in position 11 lose their sweetness and become bitter instead [89,91–93]. Though RS values are only found for the major components, at least some conclusions can be drawn. Mogroside V (79), which consists of five sugar moieties (a disaccharide in position C-3 and a branched trisaccharide in position C-24) has an RS value of 425 compared to a 1% (w/v) sucrose solution and is sweeter than mogrosides IVe (76) and VI (82), which consist either of two disaccharides (76) or two branched trisaccharides, respectively. Siamenoside I (77), which has only four sugar moieties has an RS value of 563 and is considered the

Molecules 2020, 25, 1946 17 of 23

States, where they are classiﬁed with the GRAS status [12,68,89]. Of the more than 50 triterpenoids isolated so far, only those bearing an α-hydroxy-group in position 11 and at least three sugar moieties exhibit a sweet taste [89–93]. Compounds without a hydroxy-group or with a keto-group in position 11 lose their sweetness and become bitter instead [89,91–93]. Though RS values are only found for the major components, at least some conclusions can be drawn. Mogroside V (79), which consists of ﬁve sugar moieties (a disaccharide in position C-3 and a branched trisaccharide in position C-24) has an RS value of 425 compared to a 1% (w/v) sucrose solution and is sweeter than mogrosides IVe (76) and VI (82), which consist either of two disaccharides (76) or two branched trisaccharides, Molecules 2020, 25, x 17 of 23 respectively. Siamenoside I (77), which has only four sugar moieties has an RS value of 563 and is sweetestconsidered of the sweetestmogrosides. of the However, mogrosides. the four However, sugar moieties the four of sugar siamenoside moieties I of (77 siamenoside) are not equally I (77) dividedare not equallyinto two divided disaccharides into two disaccharidesbut split into buta monoglucoside split into a monoglucoside at position atC-3 position and a C-3branched and a trisaccharidebranched trisaccharide at position atC-24. position Mogroside C-24. MogrosideV (79) was also V (79 modeled) was also against modeled stevioside against (35 stevioside) in a binding (35) modelin a binding of the modelVFD at of the the TAS1R2 VFD at subunit the TAS1R2 [26]. subunitThe authors [26]. found The authors a strong found linkage a strong for mogroside linkage for V (mogroside79) and an V additional (79) and an hydrophilic additional interaction hydrophilic of interaction the 11-hydroxy-group, of the 11-hydroxy-group, which could which explain could the higherexplain intensity the higher compared intensity to compared stevioside to stevioside(35). Unfortunately, (35). Unfortunately, the authors the confused authors confusedthe sugar themoieties sugar ofmoieties mogroside of mogroside V (79), and V (therefore79), and thereforeno conclusions no conclusions for the glycosylation for the glycosylation pattern can pattern be made can from be made this study.from this study.

Figure 15. ChemicalChemical structures structures of sweet-tasting cucurbitan-type triterpenoids.

3.7.4. Steroids Steroids The last compound class to be discussed in this review are sweet-tasting steroids ( 8888––100100)) (Figure 1616).). Investigation Investigation of of two two fern fern species species of of the the Polypodiaceae Polypodiaceae family, family, namely Polypodium vulgare and PolypodiumPolypodium glycyrrhiza glycyrrhiza yieldedyielded compounds compounds osladin osladin (88 ()88 as) aswell well as polypodiosides as polypodiosides A and A andB (89– B (9089–90) [94–96].)[94– Osladin96]. Osladin (88) and (88 )polypodioside and polypodioside A (89) Awere (89 )both were found both to found be potent to be sweeteners, potent sweeteners, with RS withvalues RS of values 500 and of 500 600, and respectively. 600, respectively. However, However, their their poor poor availability availability coupled coupled with with low low water solubility and a licorice aftertaste minimizedminimized their commercial potentialpotential [[12].12].

Molecules 2020, 25, 1946 18 of 23 Molecules 2020, 25, x 18 of 23

FigureFigure 16.16. ChemicalChemical structuresstructures ofof sweet-tastingsweet-tasting steroids.steroids.

TenTen pregnanepregnane glycosidesglycosides ((9191––100100)) werewere isolatedisolated fromfrom thethe pericarpspericarps ofof MyriopteronMyriopteron extensumextensum (Apocynaceae)(Apocynaceae) andand namednamed extensumsidesextensumsides C–LC–L [[97].97]. TheThe compoundscompounds showedshowed RSRS valuesvalues ofof 5050 toto 400400 comparedcompared toto aa 1%1% ((ww//vv)) sucrosesucrose solutionsolution andand interestinginteresting substitutionsubstitution patternpattern withwith severalseveral methylatedmethylated deoxydeoxy hexoses.hexoses. These unusual sugar moieties were onlyonly linked to the hydroxy-grouphydroxy-group at positionposition 3, whereaswhereas thethe saccharidicsaccharidic moietymoiety atat positionposition 1616 waswas assembledassembled ofof glucoseglucose molecules.molecules. OfOf thesethese moietiesmoieties thethe highesthighest RS RS values values were were observed observed for for compounds compounds with with a disaccharide a disaccharide (91 (and91 and 95).95 Apart). Apart from from that, that,the terminal the terminal sugar sugar of the of themoiety moiety at atposition position 3 3pl playsays an an important important role, role, with with signiﬁcantlysignificantly higherhigher intensitiesintensities forfor compoundscompounds containingcontainingβ β--dD-cymarose-cymarose insteadinstead ofofβ β--dD-olandrose.-olandrose.

4.4. Conclusions AA totaltotal of of 115 115 plant-derived plant-derived compounds compounds with reported with reported sweetness sweetness (including (including 17 synthetic 17 derivatives synthetic butderivatives not counting but not the counting two precursors the two naringinprecursors and nari neohesperidin)ngin and neohesperidin) have been have found been in the found literature in the andliterature were and discussed were discussed in this review. in this review. Depending Depending on the on number the number of similar of similar compounds, compounds, more more or feweror fewer conclusions conclusions on on their their structure–activity structure–activity relationships relationships were were possible. possible. In In particular, particular, diterpenesditerpenes andand triterpenestriterpenes werewere foundfound suitable,suitable, asas forfor thesethese twotwo compoundcompound (sub)classes(sub)classes severalseveral sweet-tastingsweet-tasting constituentsconstituents werewere reported.reported. AA keykey issueissue withwith regardregard toto thethe developmentdevelopment ofof newnew naturalnatural highlyhighly intenseintense sweetenerssweeteners isis certainlycertainly thethe bitterbitter aftertasteaftertaste observedobserved forfor manymany compounds.compounds. ThisThis aftertasteaftertaste seemsseems toto occuroccur independentlyindependently from from the the respective respective binding binding site, site which, which may may be be due due to strongto strong similarities similarities with with the bitterthe bitter taste taste receptors. receptors. Though Though also also being being part part of of the the GPCRs, GPCRs, the the TAS2R TAS2R bitter bitter taste-receptor taste-receptor familyfamily encompassedencompassed aroundaround 25 25 members, members, and and therefore therefore detailed detailed structure–activity structure–activity relationships relationships will be will much be moremuch complex more complex [98]. [98]. Still,Still, somesome conclusionsconclusions cancan bebe derivedderived fromfrom thethe tastetaste profilesprofiles ofof thethe alreadyalready knownknown sweeteners.sweeteners. OneOne goodgood exampleexample isis thethe groupgroup ofof mogrosides,mogrosides, ofof whichwhich moremore thanthan 5050 membersmembers areare known,known, bothboth ofof bitterbitter andand extremelyextremely sweet sweet tasting tasting properties. properties. Here, Here, a a change change of of the the functional functional group group at positionat position 11 of11 theof the aglycone aglycone significantly significantly alters alters the the taste taste profile. profile. Whereas Whereas mogrosides mogrosides bearing bearing an anα α-hydroxy-group-hydroxy-group are intensely sweet, compounds with a keto-function taste bitter, a fact also apparent for glycyrrhizic

Molecules 2020, 25, 1946 19 of 23 are intensely sweet, compounds with a keto-function taste bitter, a fact also apparent for glycyrrhizic acid (54). The keto-function of the latter compound was, furthermore, found not to be relevant for its sweet flavor by comparison with other oleanan-type triterpenoids being devoid of this feature. Moreover, an impact of the sugar moieties can be deduced from the current review. On the one hand side it seems that branching of three sugar moieties at one end of the molecule is favorable for a sweet taste, whereas at the other end a monosaccharide is superior, as observed for mogrosides as well as stevia glycosides. However, for oleanan-type triterpenoids, the distal carboxylic acid substitutes the missing second sugar moiety and was, furthermore, found crucial for the sweetening effect. Interestingly, the sweetest of the eight investigated oleanan-type triterpenoids, albiziasaponin B (56), was also the only one with a branched sugar moiety at position 3. However, not only does the quantity and arrangement of sugar moieties affect the amount of sweetness. Xylose moieties in many cases were found equally sweet as their glucose counterparts, whereas substitution with rhamnose in most cases led to a decrease in sweetness, as observed for dulcoside A (42) and rebaudioside C (40). Similarly, though more specifically, in the case of extensumsides a terminal cymarose moiety led to much higher sweetness than the C-3 epimeric oleandrose. A comparison of diterpenoids with triterpenoids revealed overall higher intensities for the triterpene-type sweeteners, with some exceptions, such as e.g., bayunoside (49), which showed an RS value 500. A dynamical modeling study of the TAS1R2/TAS1R3 sweet taste-receptor binding either the diterpenoid stevioside (35) or the triterpenoid mogroside V (79) found that both the additional sugar moieties as well as the 11-hydroxy-group of mogroside V (79) caused stronger hydrogen bonds with adjacent hydrophilic residues [26]. The same results were obtained in a molecular docking study comparing mogroside V (79) with stevioside (35) and ten other terpenoid sweeteners, resulting in the lowest binding energy for mogroside V (79) among all investigated compounds [72].

Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Munger, S.D.; Meyerhof, W. The Molecular Basis of Gustatory Transduction. In Handbook of Olfaction and Gustation; Wiley: Hoboken, NJ, USA, 2015; pp. 685–700. 2. Behrens, M.; Meyerhof, W. G Protein–Coupled Taste Receptors; Elsevier BV: Amsterdam, Netherlands, 2016; pp. 227–244. 3. DeSimone, J.A.; Dubois, G.E.; Lyall, V. Chemical Modulators of Taste. In Handbook of Olfaction and Gustation; Wiley: Hoboken, NJ, USA, 2015; pp. 665–684. 4. Spaggiari, G.; Di Pizio, A.; Cozzini, P. Sweet, umami and bitter taste receptors: State of the art of in silico molecular modeling approaches. Trends Food Sci. Technol. 2020, 96, 21–29. [CrossRef] 5. Zhao, G.Q.; Zhang, Y.; Hoon, M.A.; Chandrashekar, J.; Erlenbach, I.; Ryba, N.J.; Zuker, C.S. The Receptors for Mammalian Sweet and Umami Taste. Cell 2003, 115, 255–266. [CrossRef] 6. Nuemket, N.; Yasui, N.; Kusakabe, Y.; Nomura, Y.; Atsumi, N.; Akiyama, S.; Nango, E.; Kato, Y.; Kaneko, M.K.; Takagi, J.; et al. Structural basis for perception of diverse chemical substances by T1r taste receptors. Nat. Commun. 2017, 8, 15530. [CrossRef][PubMed] 7. Temussi, P. The Sweet Taste Receptor: A Single Receptor with Multiple Sites and Modes of Interaction. In Advances in Food and Nutrition Research; Elsevier BV: Amsterdam, Netherlands, 2007; Volume 53, pp. 199–239. 8. Sharma, V.K.; Ingle, N.A.; Kaur, N.; Yadav, P.; Ingle, E.; Charania, Z. Sugar Substitutes and Health: A Review. J. Adv. Oral Res. 2016, 7, 7–11. [CrossRef] 9. Chéron, J.-B.; Golebiowski, J.; Antonczak, S.; Fiorucci, S. The anatomy of mammalian sweet taste receptors. Proteins: Struct. Funct. Bioinform. 2017, 85, 332–341. [CrossRef][PubMed] 10. Dubois, G.E.; Prakash, I. Non-Caloric Sweeteners, Sweetness Modulators, and Sweetener Enhancers. Annu. Rev. Food Sci. Technol. 2012, 3, 353–380. [CrossRef][PubMed] 11. Kinghora, A.D.; Soejarto, D.D.; Inglett, G.E. Sweetening agents of plant origin*. Crit. Rev. Plant Sci. 1986, 4, 79–120. [CrossRef] Molecules 2020, 25, 1946 20 of 23

12. Behrens, M.; Meyerhof, W.; Hellfritsch, C.; Hofmann, T. Sweet and Umami Taste: Natural Products, Their Chemosensory Targets, and Beyond. Angew. Chem. Int. Ed. 2011, 50, 2220–2242. [CrossRef] 13. Bassoli, A.; Merlini, L.; Morini, G. Isovanillyl sweeteners. From molecules to receptors. Pure Appl. Chem. 2002, 74, 1181–1187. [CrossRef] 14. Çiçek, S.S.; Vitalini, S.; Zidorn, C. Natural Phenyldihydroisocoumarins: Sources, Chemistry and Bioactivity. Nat. Prod. Commun. 2018, 13, 279–288. [CrossRef] 15. Shallenberger, R.S.; Acree, T.E. Molecular Theory of Sweet Taste. Nature 1967, 216, 480–482. [CrossRef] 16. Kier, L.B. A Molecular Theory of Sweet Taste. J. Pharm. Sci. 1972, 61, 1394–1397. [CrossRef][PubMed] 17. Nofre, C. New hypotheses for the GPCR 3D arrangement based on a molecular model of the human sweet-taste receptor. Eur. J. Med. Chem. 2001, 36, 101–108. [CrossRef] 18. Dubois, G.E. Molecular mechanism of sweetness sensation. Physiol. Behav. 2016, 164, 453–463. [CrossRef] [PubMed] 19. Bermúdez, C.; Peña, I.; Mata, S.; Alonso, J.L. Sweet Structural Signatures Unveiled in Ketohexoses. Chem. Eur. J. 2016, 22, 16829–16837. [CrossRef][PubMed] 20. Alonso, E.R.; León, I.; Kolesniková, L.; Alonso, J.L. The Structural Signs of Sweetness in Artificial Sweeteners: A Rotational Study of Sorbitol and Dulcitol. ChemPhysChem 2018, 19, 3334–3340. [CrossRef][PubMed] 21. Alonso, E.R.; León, I.; Kolesnikova, L.; Alonso, J.L. Rotational Spectrum of Saccharin: Structure and Sweetness. J. Phys. Chem. A 2019, 123, 2756–2761. [CrossRef] 22. Leach, K.; Gregory, K.J.; Leach, K. Molecular insights into allosteric modulation of Class C G protein-coupled receptors. Pharmacol. Res. 2017, 116, 105–118. [CrossRef] 23. Li, X.; Staszewski, L.; Xu, H.; Durick, K.; Zoller, M.; Adler, E. Human receptors for sweet and umami taste. Proc. Natl. Acad. Sci. USA 2002, 99, 4692–4696. [CrossRef] 24. Xu, H.; Staszewski, L.; Tang, H.; Adler, E.; Zoller, M.; Li, X. Different functional roles of T1R subunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. USA 2004, 101, 14258–14263. [CrossRef] 25. Jiang, P.; Ji, Q.; Liu, Z.; Benard, L.M.J.; Margolskee, R.F.; Max, M.; Snyder, L.A. The Cysteine-rich Region of T1R3 Determines Responses to Intensely Sweet Proteins. J. Boil. Chem. 2004, 279, 45068–45075. [CrossRef] [PubMed] 26. Kim, S.-K.; Chen, Y.; Abrol, R.; Goddard, W.A.; Guthrie, B. Activation mechanism of the G protein-coupled sweet receptor heterodimer with sweeteners and allosteric agonists. Proc. Natl. Acad. Sci. USA 2017, 114, 2568–2573. [CrossRef][PubMed] 27. The Plant List. Available online: http://www.theplantlist.org/ (accessed on 31 March 2020). 28. Eteraf-Oskouei, T.; Najafi, M.; Alvarez-Suarez, J.M.; Tulipani, S.; Romandini, S.; Bertoli, E.; Battino, M. Contribution of honey in nutrition and human health: A review. Mediterr. J. Nutr. Metab. 2010, 3, 15–23. 29. Rose, D.J.; Anderson, J.J. Sweeteners in our society. World Rev. Nutr. Diet. 1983, 41, 200–231. 30. Inglett, G. Sweeteners - a Review. Food Technol. 1981, 35, 37. 31. Hossain, A.; Yamaguchi, F.; Matsuo, T.; Tsukamoto, I.; Toyoda, Y.; Ogawa, M.; Nagata, Y.; Tokuda, M. Rare sugar d-allulose: Potential role and therapeutic monitoring in maintaining obesity and type 2 diabetes mellitus. Pharmacol. Ther. 2015, 155, 49–59. [CrossRef] 32. Shankar, P.; Ahuja, S.; Sriram, K. Non-nutritive sweeteners: Review and update. Nutrition 2013, 29, 1293–1299. [CrossRef] 33. Grembecka, M. Sugar alcohols – their role in the modern world of sweeteners: A review. Eur. Food Res. Technol. 2015, 241, 1–14. [CrossRef] 34. Abou-Donia, M.B.; El-Masry, E.M.; Abdel-Rahman, A.A.; McLendon, R.E.; Schiffman, S.S. Splenda Alters Gut Microflora and Increases Intestinal P-Glycoprotein and Cytochrome P-450 in Male Rats. J. Toxicol. Environ. Health Part A 2008, 71, 1415–1429. [CrossRef] 35. Qin, X. What made Canada become a country with the highest incidence of inflammatory bowel disease: Could sucralose be the culprit? Can. J. Gastroenterol. 2011, 25, 511. [CrossRef] 36. O’Brien-Nabors, L. Alternative Sweeteners, Fourth Edition; CRC Press: Boca Raton, FL, USA, 2011; ISBN 978-1-4398-4614-8. 37. Elbein, A.D.; Pan, Y.; Pastuszak, I.; Carroll, D. New insights on trehalose: A multifunctional molecule. Glycobiology 2003, 13, 17–27. [CrossRef][PubMed] 38. Nie, Y.; Vigues, S.; Hobbs, J.R.; Conn, G.L.; Munger, S.D. Distinct Contributions of T1R2 and T1R3 Taste Receptor Subunits to the Detection of Sweet Stimuli. Curr. Boil. 2005, 15, 1948–1952. [CrossRef][PubMed] Molecules 2020, 25, 1946 21 of 23

39. Servant, G.; Tachdjian, C.; Tang, X.-Q.; Werner, S.; Zhang, F.; Li, X.; Kamdar, P.; Petrovic, G.; Ditschun, T.; Java, A.; et al. Positive allosteric modulators of the human sweet taste receptor enhance sweet taste. Proc. Natl. Acad. Sci. USA 2010, 107, 4746–4751. [CrossRef][PubMed] 40. Masuda, K.; Koizumi, A.; Nakajima, K.-I.; Tanaka, T.; Abe, K.; Misaka, T.; Ishiguro, M. Characterization of the Modes of Binding between Human Sweet Taste Receptor and Low-Molecular-Weight Sweet Compounds. PLoS ONE 2012, 7, e35380. [CrossRef][PubMed] 41. Solms, J.; Vuataz, L.; Egli, R.H. The taste ofl- andd-amino acids. Cell. Mol. Life Sci. 1965, 21, 692–694. [CrossRef] 42. Mazur, R.H.; Schlatter, J.M.; Goldkamp, A.H. Structure-taste relationships of some dipeptides. J. Am. Chem. Soc. 1969, 91, 2684–2691. [CrossRef] 43. Iwamura, H. Structure–sweetness relationship of L-aspartyl dipeptide analogues. A receptor site topology. J. Med. Chem. 1981, 24, 575–583. [CrossRef] 44. Vleggaar, R.; Ackerman, L.G.J.; Steyn, P.S. Structure elucidation of monatin, a high-intensity sweetener isolated from the plant Schlerochiton ilicifolius. J. Chem. Soc. Perkin Trans. 1992, 1, 3095–3098. [CrossRef] 45. Bassoli, A.; Borgonovo, G.; Busnelli, G.; Morini, G.; Merlini, L. Monatin, Its Stereoisomers and Derivatives: Modeling the Sweet Taste Chemoreception Mechanism. Eur. J. Org. Chem. 2005, 2005, 2518–2525. [CrossRef] 46. Rousseau, A.L.; Buddoo, S.R.; Gordon, G.E.R.; Beemadu, S.; Kupi, B.G.; Lepuru, M.J.; Maumela, M.C.; Parsoo, A.; Sibiya, D.M.; Brady, D. Scale-Up of a Chemo-Biocatalytic Route to (2R,4R)- and (2S,4S)-Monatin. Org. Process. Res. Dev. 2011, 15, 249–257. [CrossRef] 47. Bassoli, A.; Borgonovo, G.; Morini, G. Lost and found in sweeteners: forgotten molecules and unsolved problems in the chemistry of sweet compounds. Flavour Fragr. J. 2011, 26, 269–273. [CrossRef] 48. Bassoli, A.; Drew, M.G.B.; Hattotuwagama, C.K.; Merlini, L.; Morini, G.; Wilden, G.R. Quantitative Structure-Activity Relationships of Sweet Isovanillyl Derivatives. Quant. Struct. Relationships 2001, 20, 3–16. [CrossRef] 49. Winnig, M.; Bufe, B.; Kratochwil, N.A.; Slack, J.P.; Meyerhof, W. The binding site for neohesperidin dihydrochalcone at the human sweet taste receptor. BMC Struct. Boil. 2007, 7, 66. [CrossRef] 50. Nanayakkara, N.P.D.; Hussain, R.A.; Pezzuto, J.M.; Soejarto, D.D.; Kinghorn, A.D. Potential sweetening agents of plant origin. 13. An intensely sweet dihydroflavonol derivative based on a natural product lead compound. J. Med. Chem. 1988, 31, 1250–1253. [CrossRef][PubMed] 51. Gao, F.; Wang, H.; Mabry, T.J.; Kinghorn, A.D. Dihydroflavonol sweeteners and other constituents from Hymenoxys turneri. Phytochemistry 1990, 29, 2865–2869. [CrossRef] 52. Tanaka, T.; Yamasaki, K.; Kohda, H.; Tanaka, O.; Mahato, S. Dihydrochalcone-glucosides as Sweet Principles of Symplocos ssp. Planta Medica 1980, 40, 81–83. [CrossRef] 53. Tanaka, T.; Kawamura, K.; Kohda, H.; Yamasaki, K.; Tanaka, O. Glycosides of the leaves of Symplocos spp. (Symplocaceae). Chem. Pharm. Bull. 1982, 30, 2421–2423. [CrossRef] 54. Horowitz, R.M.; Gentili, B. Taste and structure in phenolic glycosides. J. Agric. Food Chem. 1969, 17, 696–700. [CrossRef] 55. Inglett, G.E.; Krbechek, L.; Dowling, B.; Wagner, R. Dihydrochalcone Sweeteners? Sensory and Stability Evaluation. J. Food Sci. 1969, 34, 101–104. [CrossRef] 56. Dubois, G.E.; Crosby, G.A.; Stephenson, R.A.; Wingard, R.E. Dihydrochalcone sweeteners. Synthesis and sensory evaluation of sulfonate derivatives. J. Agric. Food Chem. 1977, 25, 763–772. [CrossRef] 57. Whitelaw, M.L.; Daniel, J.R. Synthesis and sensory evaluation of ring-substituted dihydrochalcone sweeteners. J. Agric. Food Chem. 1991, 39, 44–51. [CrossRef] 58. Arnoldi, A.; Bassoli, A.; Borgonovo, G.; Merlini, L. Synthesis and sweet taste of optically active (–)-haematoxylin and of some ( )-haematoxylin derivatives. J. Chem. Soc. Perkin Trans. 1995, 1, 2447–2453. ± [CrossRef] 59. Baek, N.-I.; Chung, M.-S.; Shamon, L.; Kardono, L.B.S.; Tsauri, S.; Padmawinata, K.; Pezzuto, J.M.; Soejarto, D.D.; Kinghorn, A.D. Selligueain A, a Novel Highly Sweet Proanthocyanidin from the Rhizomes of Selliguea feei. J. Nat. Prod. 1993, 56, 1532–1538. [CrossRef][PubMed] 60. Acton, E.M.; Stone, H.; Leaffer, M.A.; Oliver, S.M. Perillartine and some derivatives: Clarification of structures. Cell. Mol. Life Sci. 1970, 26, 473–474. [CrossRef] 61. Acton, E.; Stone, H. Potential new artificial sweetener from study of structure-taste relationships. Science 1976, 193, 584–586. [CrossRef] Molecules 2020, 25, 1946 22 of 23

62. Compadre, C.; Pezzuto, J.; Kinghorn, A.D.; Kamath, S. Hernandulcin: an intensely sweet compound discovered by review of ancient literature. Science 1985, 227, 417–419. [CrossRef][PubMed] 63. Kinghorn, A.D.; Kaneda, N.; Baek, N.-I.; Kennelly, E.J.; Soejarto, D.D. Noncariogenic intense natural sweeteners. Med. Res. Rev. 1998, 18, 347–360. [CrossRef] 64. Kinghorn, A.D.; Soejarto, D.D. Discovery of terpenoid and phenolic sweeteners from plants. Pure Appl. Chem. 2002, 74, 1169–1179. [CrossRef] 65. Briedel, M.; Lavielle, R. Sur le principe sucré des feuilles de Kaá-êh-é (Stevia rebaudiana B). C. R. Acad. Sci. Paris 1931, 192, 1123–1125. 66. Crammer, B.; Ikan, R. Properties and syntheses of sweetening agents. Chem. Soc. Rev. 1977, 6, 431. [CrossRef] 67. Geuns, J.M. Stevioside. Phytochemistry 2003, 64, 913–921. [CrossRef] 68. Soejarto, D.; Addo, E.M.; Kinghorn, A.D. Highly sweet compounds of plant origin: From ethnobotanical observations to wide utilization. J. Ethnopharmacol. 2019, 243, 112056. [CrossRef][PubMed] 69. Makapugay, H.; Nanayakkara, N.; Kinghorn, A.D. Improved high-performance liquid chromatographic separation of the Stevia rebaudiana sweet diterpene glycosides using linear gradient elution. J. Chromatogr. A 1984, 283, 390–395. [CrossRef] 70. Prakash, I.; Dubois, G.; Clos, J.; Wilkens, K.; Fosdick, L. Development of rebiana, a natural, non-caloric sweetener. Food Chem. Toxicol. 2008, 46, S75–S82. [CrossRef][PubMed] 71. DuBois, G.E.; Stephenson, R.A. Diterpenoid sweeteners. Synthesis and sensory evaluation of stevioside analogs with improved organoleptic properties. J. Med. Chem. 1985, 28, 93–98. [CrossRef] 72. Acevedo, W.; Ramírez-Sarmiento, C.A.; Agosin, E. Identifying the interactions between natural, non-caloric sweeteners and the human sweet receptor by molecular docking. Food Chem. 2018, 264, 164–171. [CrossRef] 73. Tanaka, T.; Kohda, H.; Tanaka, O.; Chen, F.-H.; Chou, W.-H.; Leu, J.-L. Rubusoside (β- d -Glucosyl Ester of 13- O -β- d -Glucosyl-steviol), a Sweet Principle of Rubus chingii Hu (Rosaceae). Agric. Boil. Chem. 1981, 45, 2165–2166. [CrossRef] 74. Ohtani, K.; Aikawa, Y.; Kasai, R.; Chou, W.-H.; Yamasaki, K.; Tanaka, O. Minor diterpene glycosides from sweet leaves of Rubus suavissimus. Phytochemistry 1992, 31, 1553–1559. [CrossRef] 75. Darise, M.; Mizutani, K.; Kasai, R.; Tanaka, O.; Kitahata, S.; Okada, S.; Ogawa, S.; Murakami, F.; Chen, F.-H. Enzymic transglucosylation of rubusoside and the structure-sweetness relationship of steviol-bisglycosides. Agric. Biol. Chem. 1984, 48, 2483–2488. 76. Tahara, A.; Nakata, T.; Ohtsuka, Y. Biological Sciences: New Type of Compound with Strong Sweetness. Nature 1971, 233, 619–620. [CrossRef] 77. Kim, N.-C.; Kinghorn, A.D. Highly sweet compounds of plant origin. Arch. Pharmacal Res. 2002, 25, 725–746. [CrossRef][PubMed] 78. Esaki, S.; Konishi, F.; Kamiya, S. Synthesis and taste of some glycosides of glycyrrhetic acid. Agric. Biol. Chem. 1978, 42, 1599–1600. 79. Brieskorn, C.H.; Lang, J. 18 β-Glycyrrhetinsäure und süßer Geschmack. Arch. der Pharm. 1978, 311, 1001–1009. [CrossRef] 80. Yoshikawa, M.; Morikawa, T.; Nakano, K.; Pongpiriyadacha, Y.; Murakami, T.; Matsuda, H. Characterization of New Sweet Triterpene Saponins fromAlbizia myriophylla1. J. Nat. Prod. 2002, 65, 1638–1642. [CrossRef] [PubMed] 81. Hashimoto, Y.; Ishizone, H.; Suganuma, M.; Ogura, M.; Nakatsu, K.; Yoshioka, H. Periandrin I, a sweet triterpene-glycoside from periandra dulcis. Phytochemitry 1983, 22, 259–264. [CrossRef] 82. Suttisri, R.; Chung, M.-S.; Kinghorn, A.D.; Sticher, O.; Hashimoto, Y. Periandrin V, a further sweet triterpene glycoside from Periandra dulcis. Phytochemistry 1993, 34, 405–408. [CrossRef] 83. Choi, Y.-H.; Kinghorn, A.D.; Shi, X.; Zhang, H.; Teo, B.K. Abrusoside A: a new type of highly sweet triterpene glycoside. J. Chem. Soc., Chem. Commun. 1989, 887. [CrossRef] 84. Choi, Y.-H.; Hussain, R.A.; Pezzuto, J.M.; Kinghorn, A.D.; Morton, J.F. Abrusosides A-D, Four Novel Sweet-Tasting Triterpene Glycosides from the leaves of Abrus precatorius. J. Nat. Prod. 1989, 52, 1118–1127. [CrossRef] 85. Kennelly, E.J.; Cai, L.; Kim, N.-C.; Kinghorn, A.D. Abrusoside e, a further sweet-tasting cycloartane glycoside from the leaves of Abrus precatorius. Phytochemistry 1996, 41, 1381–1383. [CrossRef] 86. Yang, D.J.; Zhong, Z.C.; Xie, Z.M. Studies on the sweet principles from the leaves of Cyclocarya paliurus (Batal.) Iljinskaya. Acta Pharm. Sin. 1992, 27, 841–844. Molecules 2020, 25, 1946 23 of 23

87. Shu, R.G.; Xu, C.R.; Li, L.N. Studies on the sweet principles from the leaves of Cyclocarya paliurus (Batal.) Iljinsk. Acta Pharm. Sin. 1995, 30, 757–761. 88. Kennelly, E.J.; Cai, L.; Long, L.; Shamon, L.; Zaw, K.; Zhou, B.-N.; Pezzuto, J.M.; Kinghorn, A.D. Novel Highly Sweet Secodammarane Glycosides from Pterocarya paliurus. J. Agric. Food Chem. 1995, 43, 2602–2607. [CrossRef] 89. Li, C.; Lin, L.-M.; Sui, F.; Wang, Z.-M.; Huo, H.-R.; Dai, L.; Jiang, T.-L. Chemistry and pharmacology of Siraitia grosvenorii: A review. Chin. J. Nat. Med. 2014, 12, 89–102. [CrossRef] 90. Niu, B.; Ke, C.-Q.; Li, B.-H.; Li, Y.; Yi, Y.; Luo, Y.; Shuai, L.; Yao, S.; Lin, L.-G.; Li, J.; et al. Cucurbitane Glucosides from the Crude Extract of Siraitia grosvenorii with Moderate Eﬀects on PGC-1α Promoter Activity. J. Nat. Prod. 2017, 80, 1428–1435. [CrossRef][PubMed] 91. Pawar, R.S.; Krynitsky, A.J.; Rader, J.I. Sweeteners from plants—with emphasis on Stevia rebaudiana (Bertoni) and Siraitia grosvenorii (Swingle). Anal. Bioanal. Chem. 2013, 405, 4397–4407. [CrossRef] 92. Matsumoto, K.; Kasai, R.; Ohtani, K.; Tanaka, O. Minor cucurbitane-glycosides from fruits of Siraitia grosvenori (Cucurbitaceae). Chem. Pharm. Bull. 1990, 38, 2030–2032. [CrossRef] 93. Kasai, R.; Matsumoto, K.; Nie, R.L.; Zhou, J.; Tanaka, O. Glycosides from Chinese medicinal plant, Hemsleya panacis-scandens, and structure-taste relationship to cucurbitane glycosides. Chem. Pharm. Bull. 1988, 36, 234–243. [CrossRef] 94. Jizba, J.; Dolejš, L.; Herout, V.; Šorm, F. The structure of osladin - the sweet principle of the rhizomes of polypodium vulgare L. Tetrahedron Lett. 1971, 12, 1329–1332. [CrossRef] 95. Kim, J.; Pezzuto, J.M.; Soejarto, D.D.; Lang, F.A.; Kinghorn, A.D. Polypodoside A, an Intensely Sweet Constituent of the Rhizomes of Polypodium glycyrrhiza. J. Nat. Prod. 1988, 51, 1166–1172. [CrossRef] 96. Kim, J.; Kinghorn, A.D. Further steroidal and ﬂavonoid constituents of the sweet plant, Polypodium glycyrrhiza. Phytochemistry 1989, 28, 1225–1228. [CrossRef] 97. Sun, G.; Dai, Q.; Zhang, H.-X.; Li, Z.-J.; Du, Z.-Z. New Sweet-Tasting C21-Pregnane Glycosides from Pericarps of Myriopteron extensum. J. Agric. Food Chem. 2016, 64, 9381–9389. [CrossRef][PubMed] 98. Behrens, M.; Briand, L.; A De March, C.; Matsunami, H.; Yamashita, A.; Meyerhof, W.; Weyand, S. Structure–Function Relationships of Olfactory and Taste Receptors. Chem. Senses 2018, 43, 81–87. [CrossRef] [PubMed]

© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).