Phytochemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Phytochemistry journal homepage: www.elsevier.com/locate/phytochem
Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach Mayank, Vikas Jaitak ⇑ Centre for Chemical and Pharmaceutical Sciences, Central University of Punjab, Bathinda (Pb) 151001, India
a r t i c l e
i n f o
Article history: Received 6 April 2015 Received in revised form 8 May 2015 Accepted 14 May 2015 Available online xxxx Keywords: Steviol glycosides Homology model TIR2 T1R3 Ramachandran plot PROCHECK ERRAT Multiple point stimulation model
a b s t r a c t Docking studies were performed on natural sweeteners from Stevia rebaudiana by constructing homology models of T1R2 and T1R3 subunits of human sweet taste receptors. Ramachandran plot, PROCHECK results and ERRAT overall quality factor were used to validate the quality of models. Furthermore, docking results of steviol glycosides (SG’s) were correlated significantly with data available in the literature which enabled to predict the exact sweetness rank order of SG’s. The binding pattern indicated that Asn 44, Ans 52, Ala 345, Pro 343, Ile 352, Gly 346, Gly 47, Ala 354, Ser 336, Thr 326 and Ser 329 are the main interacting amino acid residues in case of T1R2 and Arg 56, Glu 105, Asp 215, Asp 216, Glu 148, Asp 258, Lys 255, Ser 104, Glu 217, Leu 51, Arg 52 for T1R3, respectively. Amino acids interact with SG’s mainly by forming hydrogen bonds with the hydroxyl group of glucose moieties. Significant variation in docked poses of all the SG’s were found. In this study, we have proposed the mechanism of the sweetness of the SG’s in the form of multiple point stimulation model by considering the diverse binding patterns of various SG’s, as well as their structural features. It will give further insight in understanding the differences in the quality of taste and will be used to improve the taste of SG’s using semi-synthetic approaches. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Human sweet taste receptor (STR) is a heterodimer composed of two subunits, T1R2 and TIR3 which belongs to class C of G-protein coupled receptor (C-GPCR) family (Li et al., 2002). Other members of this class include eight metabotropic glutamate (mGlu1–8), two heterodimeric-aminobutyric acid B (GABAB), one calciumsensing (CaR), one promiscuous L-a-amino acid (GPRC6A) and many orphan receptors (Brauner-Osborne et al., 2007). Each subunit of STR consists of a large amino-terminal domain (ATD) which is joined to a seven transmembrane helical domain (TMD) by extracellular cysteine-rich domain (CRD) that is common to all classes of C-GPCR (Brauner-Osborne et al., 2007; Pin et al., 2003). ATD of C-GPCR is the biggest domain containing approximately 500–600 residues and holds active sites for endogenous ligands (Brauner-Osborne et al., 2007). Binding sites are also observed in TMD, which are responsible for allosteric modulation (BraunerOsborne et al., 2007). Multiple binding sites present in receptors seem to create diversity in their functioning (Masuda et al., 2012). Sucrose, glucose, sucralose and related sugars bind to active
⇑ Corresponding author. E-mail address: [email protected] (V. Jaitak).
sites of ATD of both T1R2 as well as T1R3 subunits whereas dipeptidyl sweeteners such as aspartame, neotame bind only to ATD of T1R2 and can be considered as orthosteric agonists (Nie et al., 2005; Xu et al., 2004). Cyclamate class of sweeteners bind to TMD of the T1R3 subunit. Sweet proteins show interaction with CRD along with ATD of both the subunits (Fernstrom et al., 2012; Masuda et al., 2012). Thus, along with the presence of multiple binding sites, the ability of each subunit of the heterodimeric STR to induce independent taste signaling creates more functional diversity. Nowadays, sweeteners are gaining importance because of huge demand in food as well as in the pharmaceutical industries. Consumption of sucrose and related sugar promotes inappropriate positive caloric balance, excessive weight gain and obesity (Persson et al., 1992; Agurs-Collins et al., 2009). In addition, this dietary habit is responsible for diabetes (Levine et al., 1990; Weiderpass et al., 1997; Michel et al., 2003), dental caries (Sheiham, 2001), candidiasis (Pizzo et al., 2000), inflammatory bowel disease (Persson et al., 1992) and cancer (Patel et al., 2005). Synthetic sweeteners such as aspartame, saccharin, sucralose and acesulfame potassium are available, but they are also not free from health related side effects (Sardesai and Waldshan, 1991). In a society in which the challenge of maintaining a healthy caloric balance is overwhelming for over half of the population, non-caloric natural sweeteners offer hope to those who wish to
http://dx.doi.org/10.1016/j.phytochem.2015.05.006 0031-9422/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
2
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
avoid numerous diseases associated with excessive sugar consumption (Ogden et al., 2006). Stevia rebaudiana (Bartoni) family Asteraceae is an herbaceous perennial plant indigenous to Paraguay and Brazil, where its leaves have been used by local Guarani Indians as a natural sweetener for hundreds of years (Jaitak et al., 2008). 150 species of stevia are known, but among them only S. rebaudiana has sweet tasting properties. This plant is of worldwide importance because its leaves are used as non-nutritive, high potency natural sweetener in several countries. It is now cultivated in several locations in the world such as Asia, Canada, China, Brazil and Paraguay. In 2011 the consumption of stevia extract in Japan and Korea was around 200 and 115 tons/year respectively (Jaitak et al., 2011). High sweet intensity, as well as several health benefits are the main reasons for such an enormous demand of this plant. Its water extract has several beneficial effects on human health and it also has hypoglycemic, hypotensive as well as antioxidant properties. Leaves of S. rebaudiana are useful in diabetes as a natural substitute for sugar, used in baking, inhibits the formation of cavities and plaque in teeth, nontoxic, cardiotonic and is also effective against several microbes such as Streptococcus mutans, Pseudomonas aeruginosa and Proteus vulgaris. Its leaves contain multiple ent-kaurene diterpene glycosides which contain steviol skeleton and exhibit characteristic organoleptic properties (Jaitak et al., 2011; Ceunen and Geuns, 2013). Collectively all of them are known as steviol glycosides (SG’s). S. rebaudiana leaves generally contain 6–10% of stevioside, 2–4% of rebaudioside-A and other minor glycosides up to 1–2%. Unlike sucrose and artificial sweeteners, SG’s are safer and free from various side effects (Jaitak et al., 2009a). Sweetness of SG’s may be due to interaction of associated sugar molecules with STR possibly by binding to ATD as it hold active site for sugars including glucose. This can be justified by the observation that the magnitude of sweetness differ significantly among SG’s with little variation in their glucose molecules. Furthermore, along with magnitude of sweetness, taste quality of various SG’s is also different with rebaudioside D, A and E showing better sweetness quality compared to other SG’s. Other SG’s have shown comparatively poorer taste quality which restricts its use for human consumption and limits their application in food and pharmaceutical products (Purkayastha, 2012). Multiple semi-synthetic approaches have been applied to improve the taste profile of SG’s but we are still short of the desired outcome (de Oliveira et al., 2007; Jaitak et al., 2009b). The binding pattern of SG’s to the ATD domain of STR is an important investigation aspect for exploring key binding interactions responsible for sweetness. Thus, functional groups necessary for sweetness may be identified and structure as a whole can be modified to improve the sweetness profile of SG’s. The crystal structure of STR was not available and thus by homology modeling method, we have constructed 3D models of TIR2 and T1R3 subunits. On the basis of E-value, percentage identity and phylogenetic analysis, the A chain of the metabotropic glutamate receptor mGluR5 complex with glutamate (PDB ID 3LMK) have been chosen as a template to construct the models. Docking studies of eight SG’s having known sweetness profile were performed to explore important binding interactions.
2. Results 2.1. Phylogenetic analysis and template selection Template for comparative modeling was chosen on the basis of E-value of BLAST search, percentage identity with the target sequence and phylogenetic analysis. Phylogenetic analysis of T1R2 and T1R3 subunits (Fig. 1) indicated that both were closely related to chain A of the ligand binding domain of metabotropic
glutamate receptor (mGluR5) complexes with glutamate (PDB ID 3LMK). A chain of 3LMK showed sequence identity of 25%, E-value of 2e 28 with respect to T1R2 and sequence identity of 28%, E-value of 1e 22 with respect to T1R3. Thus, chain A of 3LMK seems to be an ideal template and by using this protein, the homology model of T1R2 and TIR3 was constructed. 2.2. Homology modeling Homology models of T1R2 and T1R3 (Fig. 2) subunits were constructed using the Prime-v34012 module of Schrodinger LLC (NY, USA) which is an automated protein modeling tool. Quality of final models were evaluated using Ramachandran Plot (RAMPAGE) (Lovell et al., 2003), PROCHECK (Overall Procheck G-scores) (Laskowski et al., 1993) and ERRAT (Overall quality factor) (Colovos and Yeates, 1993; Paital et al., 2013). Ramachandran plot (Fig. 3) assessment indicated that 89.8% of residues were present in most favored region, 8.1% in allowed region and 2% in outliers region in case of T1R2 whereas 90% of residues were present in most favored region, 9.1% in allowed and just 0.9% in outlier region in case of T1R3. Overall average G factor was found to be 0.41 for T1R2 and 0.15 for T1R3. T1R2 and T1R3 models yielded the ERRAT overall quality factors of 63.830 and 86.486, respectively. 2.3. Docking simulation study Table 1 summarizes the results of docking study in which lowest dock score represents the best binding ability of the ligand. Complex formed between ligand and receptor provided a lot of information such as important amino acids needed for key interactions including hydrogen bonds, salt bridges, lipophilic, p–p and p-cation interactions. In the present study, SG’s with known sweetness profile were docked against homology models of T1R2 and T1R3 subunits of human STR. Results indicated that rebaudioside A showed best binding ability towards TIR2 and T1R3 receptor followed by rebaudioside E and rebaudioside D. The dock score of rebaudioside A was found to be 12.333 and 7.995 kcal/mol for TIR2 and T1R3 receptor, respectively. In case of rebaudioside E, dock score of 10.658 and 7.841 kcal/mol were obtained respectively for T1R2 and T1R3 receptor while rebaudioside D which was the third bestdocked ligand had shown dock score of 9.764 and 7.767 kcal/mol respectively for T1R2 and T1R3 receptor. Other compounds with decreasing order of binding ability were found to be rebaudioside B, stevioside, steviolbioside and dulcoside. The poor binding affinity of isosteviol and steviol was found due to low dock score and in fact these compounds were not shown to possess any sweetness properties. The cumulative dock scores of all the SG’s were obtained by adding the individual dock score of T1R2 as well as T1R3 as represented in Table 1. 3. Discussion SG’s are among the most investigated compounds because of high sweetness intensity, many health benefits and high availability in the plant. Taste quality of SG’s are poor except rebaudioside D, A, and E which restricts its use for human consumption and limits its application in food and pharmaceutical products (Purkayastha, 2012). The knowledge of the binding pattern of SG’s with STR seems to be the most important starting point towards the solution of this problem. Investigation of the structure and associated sweetness of SG’s has revealed that glucose moieties at C-19 and C-13 position have a substantial impact on their sweetness profile. This fact indicates that ATD of STR, which binds to natural sugars, including glucose seems to play an active role in SG mediated sweetness. Another important point in glucose
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
3
Fig. 1. Phylogenetic tree of T1R2 (a), T1R3 (b) subunits of human sweet taste receptor.
molecules at C-13 and C-19 subunits of SG is that they are distantly apart from each other. This observation indicates that ATD of STR may have different binding points distantly apart from each other which can be stimulated to get sweetness signaling. Aglycone part of SG thus seems to provide glucose moieties at C-13 and C-19 position in such an orientation that different binding sites get stimulated in synchronized way to get intense and typical sweet taste. The distant amino acid residues which get stimulated by SG’s may include orthosteric as well as allosteric active site. Phylogenetic analysis indicates that chain A of 3LMK is most closely associated
with both the subunits of STR. Previously various models of STR have been generated using templates of 1EWK but, chain A of 3LMK seems to be more appropriate according to phylogenetic analysis although both of them belong to the same group in classification of metabotropic glutamate receptor family (Conn and Pin, 1997; Walters and Hellekant, 2006). Homology model was constructed for both the subunits of STR (T1R2 and T1R3), screened for possible binding sites, and docking studies were performed to find the best possible correlation with experimental data. The Ramachandran plot, PROCHECK overall G factor and ERRAT overall
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
4
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
Fig. 2. Homology model of T1R2 (a) and T1R3 (b) generated by Prime (Maestro 9.6).
Fig. 3. Ramachandran plot for T1R2 (a) and T1R3 (b).
quality factor based validation was done to finalize both the models and further docking studies were performed. Docking results revealed that rebaudioside A showed the best score of 12.334 and 7.995 kcal/mol for T1R2 and T1R3, respectively. Rebaudioside E, D, stevioside and rebaudioside B showed dock score of 10.658, 9.764, 9.711, 9.100 kcal/mol for T1R2 and 7.995, 7.841, 7.767, 6.664, 7.814 kcal/mol for T1R3, respectively. Binding interactions of steviolbioside and dulcoside were less as compared to other SG’s. The dock score of steviolbioside and dulcoside was found to be 8.96, 6.910 for T1R2 and 6.657, 6.545 kcal/mol for T1R3, respectively. Interaction patterns (Figs. 4 and 5) indicated that hydrogen bondings (backbone and side chain) with the hydroxyl group of glucose moieties are the key interactions which are responsible for sweetness signaling.
It is quite possible that both the subunits triggered signaling independent of each other and the resulting sweetness is due to the cumulative effect of both the subunits. Collective binding affinity can be calculated by adding dock score of both T1R2 as well as TIR3 (Table 1). Along with individual docking results, collective dock score was also found to follow the same trend as observed in several experimental results. The sweet and bitter tastes, when implemented together suppress each other and thus mutual suppression tends to increase the minimum onset threshold concentration for both of them (Hellfritsch et al., 2012). SG’s have also shown varying degree of bitterness along with intense sweet taste (Hellfritsch et al., 2012). The bitterness threshold of SG’s seems to be determined majorly by associated sweetness because it was the most
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
5
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx Table 1 Dock scores, protein–ligand interactions of T1R2 and T1R3 for SG’s, correlation between dock scores, bitterness threshold and relative bitterness of SG’s. D1 + D2# (kcal/mol)
Protein ligand interactions (T1R2)
Protein–ligand interactions (T1R3)
Bitterness threshold*,#
Relative bitterness at 1 mM*
7.995
20.329
1.3
7.841
18.599
Arg 56, Glu 105, Glu 148, Asp 215, Asp 216, Lys 255, Asp 258 Leu 51,Glu 148, Asp 216, Glu 217
194 lM
10.658
Not available
Not available
Rebaudioside D
9.764
7.767
17.531
0.6
Rebaudioside B Stevioside
9.100 9.711
7.814 6.664
16.914 16.375
137 lM 112 lM
1.1 1.4
6
Steviolbioside
8.096
6.657
14.753
84 lM
0.9
7
Dulcoside
6.910
6.545
13.455
49 lM
1.7
8 9
Isosteviol Steviol
4.198 3.426
3.188 2.152
7.386 5.578
Glu 105, Glu 148, Glu 217, Arg 247, Lys 255 Ser 104, Glu 105, Asp 215 Arg 52, Arg 56, Glu 105, Asp 216, Glu 217, Lys 255 Arg 56, Ser 104, Glu 105, Asp 216 Arg 52, Arg 56, Glu 105, Asp 215, Lys 255, Arg 56 Asp 215
162 lM
4 5
Asn 44, Asn 52, Pro 343, Ala 345 Asn 44, Gly 47, Ile 325, Ala 345, Gly 346 Ile 325, Thr 326, Ser 336 Asn 44, Asn 42 Asn 44, Ile 352, Ala 354 Gly 47, Asn 52, Ser329, Ala345 Asn 44, Trp 341
– –
– –
S. no.
Ligand
Dock score of T1R2 (D1)# (kcal/mol)
1
Rebaudioside A
12.334
2
Rebaudioside E
3
Dock score of T1R3 (D2)# (kcal/mol)
Gly 346 Gly 346
Bitterness threshold: minimum dilution of SG’s where associated bitterness is detectable by taste receptors in human sensory studies. Relative bitterness: comparison of bitterness of SG’s when implemented at 1 mM concentration in human sensory studies. * Hellfritsch et al. (2012). # Higher the bitterness threshold, higher is the binding affinity.
Fig. 4. Interaction pattern of SG’s with T1R2 receptor.
dominating factor in this case (Schiffman et al., 1994). Thus, binding affinity of SG’s towards STR must have a positive correlation with bitterness threshold of SG’s. The exact positive correlation
has been obtained between our experimental docking results and previously reported bitterness threshold for several SG’s which further validate our experimental results (Table 1).
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
6
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
Fig. 5. Interaction pattern of SG’s with T1R3 receptor.
Multiple reports have indicated that rebaudioside A is the sweetest steviol glycoside which is also confirmed by our study (Prakash et al., 2008). The relative bitterness of rebaudioside D is less as compared to rebaudioside A (Table 1). Thus, mutual suppression of sweet taste is comparatively less in case of rebaudioside D; hence in-spite of low dock score rebaudioside D may be able to show minimum taste recognition threshold, as well as maximum sweetness intensity better than that of rebaudioside A as observed in some human sensory experimental results (Hellfritsch et al., 2012). Although bitterness threshold and related data is unavailable for rebaudioside E but its dock score is comparatively better than that of rebaudioside D. Thus, sweetness intensity of rebaudioside E seems to be more than that of rebaudioside D, which is also observed in multiple other experimental results (Puri et al., 2011; Purkayastha, 2012). Analyzing dock scores as well as relative bitterness of rebaudioside B, stevioside, steviolbioside and dulcoside clearly indicated that rebaudioside B is the sweetest among all the four compounds followed by stevioside, steviolbioside and dulcoside respectively which has also been reported earlier (Puri et al., 2011; Purkayastha, 2012). Non-sweet compounds, isosteviol and steviol are expected to have the least binding affinity towards STR and the same trend is obtained in our dock study. Protein–ligand interaction profile revealed that Asn 44, Ans 52, Ala 345, Pro 343, Ile 352, Gly 346, Gly 47, Ala 354, Ser 336, Thr 326, Ser 329 and Arg 56, Glu 105, Asp 215, Asp 216, Glu 148, Asp 258, Lys 255, Ser 104, Glu 217, Leu 51, Arg 52, Arg 247 were involved in the hydrogen bond interactions for T1R2 and TIR3, respectively. Interaction pattern and pose of each individual SG in term of glycone as well as aglycone part is very dissimilar as compared to others in the docking study (Fig. 6), which is quite reasonable
because all of them have unique sweetness properties, differing from other SGs. The present data also supports the above-mentioned assumption of the presence of different binding points within ATD of STR. Thus, several sets of amino acid residues are present in ATD, which are distantly apart from each other and by stimulating different set of residues one can produce characteristic sweet taste. Thus, on the basis of structural aspect as well as obtained docking poses of SG’s we have described the diverse taste of SG’s in the form of a multiple point stimulation model of SG’s for sweetness (Fig. 7). ATD domain of STR consists of several groups of active amino acids, namely active site A, B, C, D, E and so on. Monosaccharide moieties at C-13 and C-19 of SG’s stimulate different combination of active sites to produce characteristic sweet taste. Thus, diversity in the sweet taste of SG’s seems to be due to stimulation of primary active site residues by monosaccharide subunits at the C-13 position and some other type of allosteric modulation by C-19 monosaccharide subunits or vice versa. Least interaction and dock scores obtained for steviol (Figs. 4 and 5) indicated that aglycone part is only involved in providing glucose molecules an orientation enabling these glucose moieties to selectively interact with specific set of active sites to produce a characteristic taste.
4. Conclusion Homology models of T1R2 and T1R3 subunits were constructed using chain A of 3LMK protein and validated by using Ramachandran plot, PROCHECK and ERRAT server. Furthermore, docking of SG’s of known sweetness profile with modeled proteins
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
7
Fig. 6. Non-overlapping and diverse poses of SG’s docked on T1R2 (a) and T1R3 (b).
Fig. 7. Multiple point stimulation model of the sweetness mechanism of SG’s based on structure, varied sweet taste and diverse docked poses.
showed results which were in accordance with the existing experimental data. This high degree of correlation with existing data further justified our assumption that SG’s binds to ATD of STR to produce sweet taste. Several groups of researchers have shown different opinions regarding the rank order of sweetness intensity of multiple SG’s but with our docking scores we are able to predict the exact sweetness strength of various SG’s. Rebaudioside A was found to have maximum sweetness intensity followed by rebaudioside E, rebaudioside D, rebaudioside B, stevioside, steviolbioside and dulcoside respectively. Furthermore, Asn 44, Ans 52, Ala 345, Pro 343, Ile 352, Gly 346, Gly 47, Als 354, Ser 336, Thr 326 and Ser 329 are the main amino acids showing hydrogen bonding interaction with SG’s in case of T1R2 and Arg 56, Glu 105, Asp 215, Asp 216, Glu 148, Asp 258, Lys 255, Ssr 104, Glu 217, Leu 51 and Arg 52 amino acid residues have shown hydrogen bonding in case of T1R3. On the basis of dissimilarity in interaction pattern, poses of individual docked SG’s, its structural aspect and impact of C-13 and C-19 glucose molecules, we have concluded that multiple groups of amino acids so-called active sites are present in the ATD domain of STR and can be stimulated in different combinations to get diverse sweet taste. Using this model of sweetness we can easily explain how one SG is able to produce a unique sweet taste
which is different from other SG’s. The interaction of a particular SG with a specific set of active sites by mean of C-13 and C-19 glucose molecules is responsible for producing a specific taste. This, observation further justified that how an individual SG with difference only in sugar moieties at C-13 and/or C-19 positions of steviol are able to produce unique taste quality. The multiple point stimulation model is an important aspect for further exploring the different taste related points in ATD of STR. It will also be useful to enhance the taste quality and sweetness index of SG’s by further modifying structures and thus enabling their binding towards a specific point in STR. Thus, by improving dock score, we may be able to increase the sweetness intensity of SG’s and by manipulating the binding pattern taste quality can be improved.
5. Experimental 5.1. Ligand preparation SG’s were drawn using ChemBioDraw-12 (Fig. 8) and ligands were prepared using LegPrep wizard of Maestro 9.6. In ligand preparation step, 2D structure of the ligand is sufficiently modified
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
8
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
Fig. 8. Structures of SG’s used in the study.
to produce a better representative model of the actual ligand which is necessary for better docking results. 5.2. Template selection Amino acid sequences of T1R2 and T1R3 receptors were retrieved from NCBI (Accession: XP_94400 NP_689418.27, GI:112789566 and NP_689414.1, GI:91206396 respectively). Template required for homology modeling was taken from Protein Data Bank (PDB). A BLAST search of PDB database was done for suitable templates of T1R2 and T1R3. The BLAST search results were further used to create a phylogenetic tree using ClustalW2 to determine the most appropriate template for T1R2 and T1R3. ‘‘A’’ chain of the ligand binding domain of the Metabotropic glutamate receptor mGluR5 complexed with glutamate (PDB ID 3LMK), which is closest to the sequences of T1R2 and T1R3 in the phylogenetic tree, have suitable E-value and sequence identity of 25% with respect to T1R2 and 28% with respect to T1R3 was selected for homology modeling of both the subunits. 5.3. Homology modeling and protein preparation Homology models of T1R2 and T1R3 were constructed using the prime-v34012 module of the Schrodinger LLC (NY, USA) which is an automated protein modeling tool. The primary sequence of T1R2 and T1R3 in FASTA format was imported into structure prediction wizard present in Prime tool. BLAST search was performed and A chain of 3LMK being closest in the phylogenetic tree of TIR2 and T1R3, having suitable E-value and having a sequence identity of 25% and 28%, respectively was selected for homology modeling. Thus, using A chain of 3LMK, initial target-template alignment was generated and subsequently secondary structure prediction was performed by secondary structure prediction module (SSP) integrated with Prime. Prime structure prediction module is provided with two model building methods and we have selected energybased method to build the models. The generated model was further subjected to loop and side-chain optimization. The restrained minimization of final models of T1R2 and TIR3 were performed by
protein preparation wizard using the OPLS_2005 force field with default cut off root mean square distance (RMSD) of 0.30 Å. Finally, refined homology models were evaluated using RAMPAGE, PROCHECK and ERRAT web servers. The final proteins were now ready for docking procedure. 5.4. Grid generation Fully prepared and restrain minimized model of T1R2 and T1R3 were screened for possible binding sites using SitMap in Maestro 9.6 which had provided best five possible binding sites using default settings. The grid was generated on all the five predicted binding sites using default settings to define the area in the protein where docking was performed. 5.5. Molecular docking Docking was performed on already generated five grid sites using Glide docking application present in Schrödinger suite 2013. Glide is one of the most accurate method for docking studies as observed in several studies (Kellenberger et al., 2004; Zhou et al., 2007). There were three precessions provided with Maestro Glide docking, which include high throughput virtual screening (HTVC), standard precision (SP) and most accurate extra precision (XP). XP docking precision was used to dock SG’s. Further selection and interpretation of results were done on the basis of best agreement with the experimental sweetness profile of SG’s. Acknowledgements Authors are grateful to Department of Science and Technology, New Delhi, India for providing financial assistance during the course of the work. Dually acknowledge to Dr. Alpna Saini, Assistant Professor, Centre for Comparative literature for editing and linguistic correction of the manuscript. Authors are also thankful to the Honorable Vice - Chancellor for providing the necessary facilities at Central University of Punjab, Bathinda, India.
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
References Agurs-Collins, T., Rosenberg, L., Makambi, K., Palmer, J.R., Adams-Campbell, L., 2009. Dietary patterns and breast cancer risk in women participating in the Black Women’s Health Study. Am. J. Clin. Nutr. 90, 621–628. Brauner-Osborne, H., Wellendorph, P., Jensen, A.A., 2007. Structure, pharmacology and therapeutic prospects of family C G-protein coupled receptors. Curr. Drug Targets 8, 169–184. Ceunen, S., Geuns, J.M., 2013. Steviol glycosides: chemical diversity, metabolism, and function. J. Nat. Prod. 76, 1201–1228. Colovos, C., Yeates, T.O., 1993. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 2, 1511–1519. Conn, P.J., Pin, J.-P., 1997. Pharmacology and functions of metabotropic glutamate receptors. Annu. Rev. Pharmacol. 37, 205–237. de Oliveira, B.H., Packer, J.F., Chimelli, M., de Jesus, D.A., 2007. Enzymatic modification of stevioside by cell-free extract of Gibberella fujikuroi. J. Biotechnol. 131, 92–96. Fernstrom, J.D., Munger, S.D., Sclafani, A., de Araujo, I.E., Roberts, A., Molinary, S., 2012. Mechanisms for sweetness. J. Nutr. 142, 1134S–1141S. Hellfritsch, C., Brockhoff, A., Stähler, F., Meyerhof, W., Hofmann, T., 2012. Human psychometric and taste receptor responses to steviol glycosides. J. Agr. Food Chem. 60, 6782–6793. Jaitak, V., Bandna, B.S., Kaul, V., 2009a. An efficient microwave-assisted extraction process of stevioside and rebaudioside-A from Stevia rebaudiana (Bertoni). Phytochem. Anal. 20, 240–245. Jaitak, V., Gupta, A., Kaul, V., Ahuja, P.S., 2008. Validated high-performance thinlayer chromatography method for Steviol glycosides in Stevia rebaudiana. J. Pharm. Biomed. Anal. 47, 790–794. Jaitak, V., Kaul, V.K., Kumar, N., Singh, B., Savergave, L., Jogdand, V., Nene, S., 2009b. Simple and efficient enzymatic transglycosylation of stevioside by bcyclodextrin glucanotransferase from Bacillus firmus. Biotechnol. Lett. 31, 1415–1420. Jaitak, V., Kaul, K., Kaul, V.K., Singh, V., Singh, B., 2011. Stevia rebaudiana: a natural substitute for sugar. In: Genetic Resources, Chromosome Engineering, and Crop Improvement. CRC Press, pp. 885–910. Kellenberger, E., Rodrigo, J., Muller, P., Rognan, D., 2004. Comparative evaluation of eight docking tools for docking and virtual screening accuracy. Proteins 57, 225–242. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Levine, W., Dyer, A.R., Shekelle, R.B., Schoenberger, J.A., Stamler, J., 1990. Post-load plasma glucose and cancer mortality in middle-aged men and women: 12-year follow-up findings of the Chicago Heart Association Detection Project in Industry. Am. J. Epidemiol. 131, 254–262. Li, X., Staszewski, L., Xu, H., Durick, K., Zoller, M., Adler, E., 2002. Human receptors for sweet and umami taste. Proc. Natl. Acad. Sci. U.S.A. 99, 4692–4696. Lovell, S.C., Davis, I.W., Arendall, W.B., de Bakker, P.I., Word, J.M., Prisant, M.G., Richardson, J.S., Richardson, D.C., 2003. Structure validation by Ca geometry: /, w and Cb deviation. Proteins 50, 437–450.
9
Masuda, K., Koizumi, A., Nakajima, K.-I., Tanaka, T., Abe, K., Misaka, T., Ishiguro, M., 2012. Characterization of the modes of binding between human sweet taste receptor and low-molecular-weight sweet compounds. PLoS ONE 7, e35380. Michel, K.B., Solomon, C.G., Hu, F.B., Rosner, B.A., Hankinson, S.E., Colditz, G.A., Manson, J.E., 2003. Type 2 diabetes and subsequent incidence of breast cancer in the Nurses’™ Health Study. Diabetes Care 26, 1752–1758. Nie, Y., Vigues, S., Hobbs, J.R., Conn, G.L., Munger, S.D., 2005. Distinct contributions of T1R2 and T1R3 taste receptor subunits to the detection of sweet stimuli. Curr. Biol. 15, 1948–1952. Ogden, C.L., Carroll, M.D., Curtin, L.R., McDowell, M.A., Tabak, C.J., Flegal, K.M., 2006. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 295, 1549–1555. Paital, B., Kumar, S., Farmer, R., Chainy, G.B.N., 2013. In silico prediction of 3D structure of Mn superoxide dismutase of Scylla serrata and its binding properties with inhibitors. Interdiscip. Sci. Comput. Life Sci. 5, 69–76. Prakash, I., DuBois, G., Clos, J., Wilkens, K., Fosdick, L., 2008. Development of rebiana, a natural, non-caloric sweetener. Food Chem. Toxicol. 46, S75–S82. Patel, A.V., Rodriguez, C., Bernstein, L., Chao, A., Thun, M.J., Calle, E.E., 2005. Obesity, recreational physical activity and risk of pancreatic cancer in a large U.S. cohort. Cancer Epidemiol. Biomarkers Prev. 14, 459–466. Persson, P.G., Ahlbom, A., Hellers, G., 1992. Diet and inflammatory bowel disease: a case-control study. Epidemiology 3, 47–52. Pin, J.-P., Galvez, T., Prézeau, L., 2003. Evolution, structure, and activation mechanism of family 3/C G-protein-coupled receptors. Pharmacol. Ther. 98, 325–354. Pizzo, G., Giuliana, G., Milici, M.E., Giangreco, R., 2000. Effect of dietary carbohydrates on the in vitro epithelial adhesion of Candida albicans, Candida tropicalis, and Candida krusei. New Microbiol. 23, 63–71. Purkayastha, S., 2012. Glucosylated SG composition as a flavor modifier. U.S. Patent Application 14/005,852. Puri, M., Sharma, D., Tiwari, A.K., 2011. Downstream processing of stevioside and its potential applications. Biotechnol. Adv. 29, 781–791. Sardesai, V.M., Waldshan, T.H., 1991. Natural and synthetic intense sweeteners. J. Nutr. Biochem. 2, 236–244. Schiffman, S.S., Gatlin, L.A., Sattely-Miller, E.A., Graham, B.G., Heiman, S.A., Stagner, W.C., Erickson, R.P., 1994. The effect of sweeteners on bitter taste in young and elderly subjects. Brain Res. Bull. 35, 189–204. Sheiham, A., 2001. Dietary effects on dental disease. Public Health Nutr. 4, 569–591. Walters, D.E., Hellekant, G., 2006. Interactions of the sweet protein brazzein with the sweet taste receptor. J. Agr. Food Chem. 54, 10129–10133. Weiderpass, E., Gridley, G., Nyren, O., Ekbom, A., Persson, I., Adami, H.O., 1997. Diabetes mellitus and risk of large bowel cancer. J. Natl Cancer Inst. 89, 660– 661. Xu, H., Staszewski, L., Tang, H., Adler, E., Zoller, M., Li, X., 2004. Different functional roles of T1R subunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. U.S.A. 101, 14258–14263. Zhou, Z., Felts, A.K., Friesner, R.A., Levy, R.M., 2007. Comparative performance of several flexible docking programs and scoring functions: enrichment studies for a diverse set of pharmaceutically relevant targets. J. Chem. Inf. Model. 47, 1599– 1608.
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Contents lists available at ScienceDirect
Phytochemistry journal homepage: www.elsevier.com/locate/phytochem
Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach Mayank, Vikas Jaitak ⇑ Centre for Chemical and Pharmaceutical Sciences, Central University of Punjab, Bathinda (Pb) 151001, India
a r t i c l e
i n f o
Article history: Received 6 April 2015 Received in revised form 8 May 2015 Accepted 14 May 2015 Available online xxxx Keywords: Steviol glycosides Homology model TIR2 T1R3 Ramachandran plot PROCHECK ERRAT Multiple point stimulation model
a b s t r a c t Docking studies were performed on natural sweeteners from Stevia rebaudiana by constructing homology models of T1R2 and T1R3 subunits of human sweet taste receptors. Ramachandran plot, PROCHECK results and ERRAT overall quality factor were used to validate the quality of models. Furthermore, docking results of steviol glycosides (SG’s) were correlated significantly with data available in the literature which enabled to predict the exact sweetness rank order of SG’s. The binding pattern indicated that Asn 44, Ans 52, Ala 345, Pro 343, Ile 352, Gly 346, Gly 47, Ala 354, Ser 336, Thr 326 and Ser 329 are the main interacting amino acid residues in case of T1R2 and Arg 56, Glu 105, Asp 215, Asp 216, Glu 148, Asp 258, Lys 255, Ser 104, Glu 217, Leu 51, Arg 52 for T1R3, respectively. Amino acids interact with SG’s mainly by forming hydrogen bonds with the hydroxyl group of glucose moieties. Significant variation in docked poses of all the SG’s were found. In this study, we have proposed the mechanism of the sweetness of the SG’s in the form of multiple point stimulation model by considering the diverse binding patterns of various SG’s, as well as their structural features. It will give further insight in understanding the differences in the quality of taste and will be used to improve the taste of SG’s using semi-synthetic approaches. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Human sweet taste receptor (STR) is a heterodimer composed of two subunits, T1R2 and TIR3 which belongs to class C of G-protein coupled receptor (C-GPCR) family (Li et al., 2002). Other members of this class include eight metabotropic glutamate (mGlu1–8), two heterodimeric-aminobutyric acid B (GABAB), one calciumsensing (CaR), one promiscuous L-a-amino acid (GPRC6A) and many orphan receptors (Brauner-Osborne et al., 2007). Each subunit of STR consists of a large amino-terminal domain (ATD) which is joined to a seven transmembrane helical domain (TMD) by extracellular cysteine-rich domain (CRD) that is common to all classes of C-GPCR (Brauner-Osborne et al., 2007; Pin et al., 2003). ATD of C-GPCR is the biggest domain containing approximately 500–600 residues and holds active sites for endogenous ligands (Brauner-Osborne et al., 2007). Binding sites are also observed in TMD, which are responsible for allosteric modulation (BraunerOsborne et al., 2007). Multiple binding sites present in receptors seem to create diversity in their functioning (Masuda et al., 2012). Sucrose, glucose, sucralose and related sugars bind to active
⇑ Corresponding author. E-mail address: [email protected] (V. Jaitak).
sites of ATD of both T1R2 as well as T1R3 subunits whereas dipeptidyl sweeteners such as aspartame, neotame bind only to ATD of T1R2 and can be considered as orthosteric agonists (Nie et al., 2005; Xu et al., 2004). Cyclamate class of sweeteners bind to TMD of the T1R3 subunit. Sweet proteins show interaction with CRD along with ATD of both the subunits (Fernstrom et al., 2012; Masuda et al., 2012). Thus, along with the presence of multiple binding sites, the ability of each subunit of the heterodimeric STR to induce independent taste signaling creates more functional diversity. Nowadays, sweeteners are gaining importance because of huge demand in food as well as in the pharmaceutical industries. Consumption of sucrose and related sugar promotes inappropriate positive caloric balance, excessive weight gain and obesity (Persson et al., 1992; Agurs-Collins et al., 2009). In addition, this dietary habit is responsible for diabetes (Levine et al., 1990; Weiderpass et al., 1997; Michel et al., 2003), dental caries (Sheiham, 2001), candidiasis (Pizzo et al., 2000), inflammatory bowel disease (Persson et al., 1992) and cancer (Patel et al., 2005). Synthetic sweeteners such as aspartame, saccharin, sucralose and acesulfame potassium are available, but they are also not free from health related side effects (Sardesai and Waldshan, 1991). In a society in which the challenge of maintaining a healthy caloric balance is overwhelming for over half of the population, non-caloric natural sweeteners offer hope to those who wish to
http://dx.doi.org/10.1016/j.phytochem.2015.05.006 0031-9422/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
2
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
avoid numerous diseases associated with excessive sugar consumption (Ogden et al., 2006). Stevia rebaudiana (Bartoni) family Asteraceae is an herbaceous perennial plant indigenous to Paraguay and Brazil, where its leaves have been used by local Guarani Indians as a natural sweetener for hundreds of years (Jaitak et al., 2008). 150 species of stevia are known, but among them only S. rebaudiana has sweet tasting properties. This plant is of worldwide importance because its leaves are used as non-nutritive, high potency natural sweetener in several countries. It is now cultivated in several locations in the world such as Asia, Canada, China, Brazil and Paraguay. In 2011 the consumption of stevia extract in Japan and Korea was around 200 and 115 tons/year respectively (Jaitak et al., 2011). High sweet intensity, as well as several health benefits are the main reasons for such an enormous demand of this plant. Its water extract has several beneficial effects on human health and it also has hypoglycemic, hypotensive as well as antioxidant properties. Leaves of S. rebaudiana are useful in diabetes as a natural substitute for sugar, used in baking, inhibits the formation of cavities and plaque in teeth, nontoxic, cardiotonic and is also effective against several microbes such as Streptococcus mutans, Pseudomonas aeruginosa and Proteus vulgaris. Its leaves contain multiple ent-kaurene diterpene glycosides which contain steviol skeleton and exhibit characteristic organoleptic properties (Jaitak et al., 2011; Ceunen and Geuns, 2013). Collectively all of them are known as steviol glycosides (SG’s). S. rebaudiana leaves generally contain 6–10% of stevioside, 2–4% of rebaudioside-A and other minor glycosides up to 1–2%. Unlike sucrose and artificial sweeteners, SG’s are safer and free from various side effects (Jaitak et al., 2009a). Sweetness of SG’s may be due to interaction of associated sugar molecules with STR possibly by binding to ATD as it hold active site for sugars including glucose. This can be justified by the observation that the magnitude of sweetness differ significantly among SG’s with little variation in their glucose molecules. Furthermore, along with magnitude of sweetness, taste quality of various SG’s is also different with rebaudioside D, A and E showing better sweetness quality compared to other SG’s. Other SG’s have shown comparatively poorer taste quality which restricts its use for human consumption and limits their application in food and pharmaceutical products (Purkayastha, 2012). Multiple semi-synthetic approaches have been applied to improve the taste profile of SG’s but we are still short of the desired outcome (de Oliveira et al., 2007; Jaitak et al., 2009b). The binding pattern of SG’s to the ATD domain of STR is an important investigation aspect for exploring key binding interactions responsible for sweetness. Thus, functional groups necessary for sweetness may be identified and structure as a whole can be modified to improve the sweetness profile of SG’s. The crystal structure of STR was not available and thus by homology modeling method, we have constructed 3D models of TIR2 and T1R3 subunits. On the basis of E-value, percentage identity and phylogenetic analysis, the A chain of the metabotropic glutamate receptor mGluR5 complex with glutamate (PDB ID 3LMK) have been chosen as a template to construct the models. Docking studies of eight SG’s having known sweetness profile were performed to explore important binding interactions.
2. Results 2.1. Phylogenetic analysis and template selection Template for comparative modeling was chosen on the basis of E-value of BLAST search, percentage identity with the target sequence and phylogenetic analysis. Phylogenetic analysis of T1R2 and T1R3 subunits (Fig. 1) indicated that both were closely related to chain A of the ligand binding domain of metabotropic
glutamate receptor (mGluR5) complexes with glutamate (PDB ID 3LMK). A chain of 3LMK showed sequence identity of 25%, E-value of 2e 28 with respect to T1R2 and sequence identity of 28%, E-value of 1e 22 with respect to T1R3. Thus, chain A of 3LMK seems to be an ideal template and by using this protein, the homology model of T1R2 and TIR3 was constructed. 2.2. Homology modeling Homology models of T1R2 and T1R3 (Fig. 2) subunits were constructed using the Prime-v34012 module of Schrodinger LLC (NY, USA) which is an automated protein modeling tool. Quality of final models were evaluated using Ramachandran Plot (RAMPAGE) (Lovell et al., 2003), PROCHECK (Overall Procheck G-scores) (Laskowski et al., 1993) and ERRAT (Overall quality factor) (Colovos and Yeates, 1993; Paital et al., 2013). Ramachandran plot (Fig. 3) assessment indicated that 89.8% of residues were present in most favored region, 8.1% in allowed region and 2% in outliers region in case of T1R2 whereas 90% of residues were present in most favored region, 9.1% in allowed and just 0.9% in outlier region in case of T1R3. Overall average G factor was found to be 0.41 for T1R2 and 0.15 for T1R3. T1R2 and T1R3 models yielded the ERRAT overall quality factors of 63.830 and 86.486, respectively. 2.3. Docking simulation study Table 1 summarizes the results of docking study in which lowest dock score represents the best binding ability of the ligand. Complex formed between ligand and receptor provided a lot of information such as important amino acids needed for key interactions including hydrogen bonds, salt bridges, lipophilic, p–p and p-cation interactions. In the present study, SG’s with known sweetness profile were docked against homology models of T1R2 and T1R3 subunits of human STR. Results indicated that rebaudioside A showed best binding ability towards TIR2 and T1R3 receptor followed by rebaudioside E and rebaudioside D. The dock score of rebaudioside A was found to be 12.333 and 7.995 kcal/mol for TIR2 and T1R3 receptor, respectively. In case of rebaudioside E, dock score of 10.658 and 7.841 kcal/mol were obtained respectively for T1R2 and T1R3 receptor while rebaudioside D which was the third bestdocked ligand had shown dock score of 9.764 and 7.767 kcal/mol respectively for T1R2 and T1R3 receptor. Other compounds with decreasing order of binding ability were found to be rebaudioside B, stevioside, steviolbioside and dulcoside. The poor binding affinity of isosteviol and steviol was found due to low dock score and in fact these compounds were not shown to possess any sweetness properties. The cumulative dock scores of all the SG’s were obtained by adding the individual dock score of T1R2 as well as T1R3 as represented in Table 1. 3. Discussion SG’s are among the most investigated compounds because of high sweetness intensity, many health benefits and high availability in the plant. Taste quality of SG’s are poor except rebaudioside D, A, and E which restricts its use for human consumption and limits its application in food and pharmaceutical products (Purkayastha, 2012). The knowledge of the binding pattern of SG’s with STR seems to be the most important starting point towards the solution of this problem. Investigation of the structure and associated sweetness of SG’s has revealed that glucose moieties at C-19 and C-13 position have a substantial impact on their sweetness profile. This fact indicates that ATD of STR, which binds to natural sugars, including glucose seems to play an active role in SG mediated sweetness. Another important point in glucose
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
3
Fig. 1. Phylogenetic tree of T1R2 (a), T1R3 (b) subunits of human sweet taste receptor.
molecules at C-13 and C-19 subunits of SG is that they are distantly apart from each other. This observation indicates that ATD of STR may have different binding points distantly apart from each other which can be stimulated to get sweetness signaling. Aglycone part of SG thus seems to provide glucose moieties at C-13 and C-19 position in such an orientation that different binding sites get stimulated in synchronized way to get intense and typical sweet taste. The distant amino acid residues which get stimulated by SG’s may include orthosteric as well as allosteric active site. Phylogenetic analysis indicates that chain A of 3LMK is most closely associated
with both the subunits of STR. Previously various models of STR have been generated using templates of 1EWK but, chain A of 3LMK seems to be more appropriate according to phylogenetic analysis although both of them belong to the same group in classification of metabotropic glutamate receptor family (Conn and Pin, 1997; Walters and Hellekant, 2006). Homology model was constructed for both the subunits of STR (T1R2 and T1R3), screened for possible binding sites, and docking studies were performed to find the best possible correlation with experimental data. The Ramachandran plot, PROCHECK overall G factor and ERRAT overall
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
4
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
Fig. 2. Homology model of T1R2 (a) and T1R3 (b) generated by Prime (Maestro 9.6).
Fig. 3. Ramachandran plot for T1R2 (a) and T1R3 (b).
quality factor based validation was done to finalize both the models and further docking studies were performed. Docking results revealed that rebaudioside A showed the best score of 12.334 and 7.995 kcal/mol for T1R2 and T1R3, respectively. Rebaudioside E, D, stevioside and rebaudioside B showed dock score of 10.658, 9.764, 9.711, 9.100 kcal/mol for T1R2 and 7.995, 7.841, 7.767, 6.664, 7.814 kcal/mol for T1R3, respectively. Binding interactions of steviolbioside and dulcoside were less as compared to other SG’s. The dock score of steviolbioside and dulcoside was found to be 8.96, 6.910 for T1R2 and 6.657, 6.545 kcal/mol for T1R3, respectively. Interaction patterns (Figs. 4 and 5) indicated that hydrogen bondings (backbone and side chain) with the hydroxyl group of glucose moieties are the key interactions which are responsible for sweetness signaling.
It is quite possible that both the subunits triggered signaling independent of each other and the resulting sweetness is due to the cumulative effect of both the subunits. Collective binding affinity can be calculated by adding dock score of both T1R2 as well as TIR3 (Table 1). Along with individual docking results, collective dock score was also found to follow the same trend as observed in several experimental results. The sweet and bitter tastes, when implemented together suppress each other and thus mutual suppression tends to increase the minimum onset threshold concentration for both of them (Hellfritsch et al., 2012). SG’s have also shown varying degree of bitterness along with intense sweet taste (Hellfritsch et al., 2012). The bitterness threshold of SG’s seems to be determined majorly by associated sweetness because it was the most
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
5
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx Table 1 Dock scores, protein–ligand interactions of T1R2 and T1R3 for SG’s, correlation between dock scores, bitterness threshold and relative bitterness of SG’s. D1 + D2# (kcal/mol)
Protein ligand interactions (T1R2)
Protein–ligand interactions (T1R3)
Bitterness threshold*,#
Relative bitterness at 1 mM*
7.995
20.329
1.3
7.841
18.599
Arg 56, Glu 105, Glu 148, Asp 215, Asp 216, Lys 255, Asp 258 Leu 51,Glu 148, Asp 216, Glu 217
194 lM
10.658
Not available
Not available
Rebaudioside D
9.764
7.767
17.531
0.6
Rebaudioside B Stevioside
9.100 9.711
7.814 6.664
16.914 16.375
137 lM 112 lM
1.1 1.4
6
Steviolbioside
8.096
6.657
14.753
84 lM
0.9
7
Dulcoside
6.910
6.545
13.455
49 lM
1.7
8 9
Isosteviol Steviol
4.198 3.426
3.188 2.152
7.386 5.578
Glu 105, Glu 148, Glu 217, Arg 247, Lys 255 Ser 104, Glu 105, Asp 215 Arg 52, Arg 56, Glu 105, Asp 216, Glu 217, Lys 255 Arg 56, Ser 104, Glu 105, Asp 216 Arg 52, Arg 56, Glu 105, Asp 215, Lys 255, Arg 56 Asp 215
162 lM
4 5
Asn 44, Asn 52, Pro 343, Ala 345 Asn 44, Gly 47, Ile 325, Ala 345, Gly 346 Ile 325, Thr 326, Ser 336 Asn 44, Asn 42 Asn 44, Ile 352, Ala 354 Gly 47, Asn 52, Ser329, Ala345 Asn 44, Trp 341
– –
– –
S. no.
Ligand
Dock score of T1R2 (D1)# (kcal/mol)
1
Rebaudioside A
12.334
2
Rebaudioside E
3
Dock score of T1R3 (D2)# (kcal/mol)
Gly 346 Gly 346
Bitterness threshold: minimum dilution of SG’s where associated bitterness is detectable by taste receptors in human sensory studies. Relative bitterness: comparison of bitterness of SG’s when implemented at 1 mM concentration in human sensory studies. * Hellfritsch et al. (2012). # Higher the bitterness threshold, higher is the binding affinity.
Fig. 4. Interaction pattern of SG’s with T1R2 receptor.
dominating factor in this case (Schiffman et al., 1994). Thus, binding affinity of SG’s towards STR must have a positive correlation with bitterness threshold of SG’s. The exact positive correlation
has been obtained between our experimental docking results and previously reported bitterness threshold for several SG’s which further validate our experimental results (Table 1).
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
6
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
Fig. 5. Interaction pattern of SG’s with T1R3 receptor.
Multiple reports have indicated that rebaudioside A is the sweetest steviol glycoside which is also confirmed by our study (Prakash et al., 2008). The relative bitterness of rebaudioside D is less as compared to rebaudioside A (Table 1). Thus, mutual suppression of sweet taste is comparatively less in case of rebaudioside D; hence in-spite of low dock score rebaudioside D may be able to show minimum taste recognition threshold, as well as maximum sweetness intensity better than that of rebaudioside A as observed in some human sensory experimental results (Hellfritsch et al., 2012). Although bitterness threshold and related data is unavailable for rebaudioside E but its dock score is comparatively better than that of rebaudioside D. Thus, sweetness intensity of rebaudioside E seems to be more than that of rebaudioside D, which is also observed in multiple other experimental results (Puri et al., 2011; Purkayastha, 2012). Analyzing dock scores as well as relative bitterness of rebaudioside B, stevioside, steviolbioside and dulcoside clearly indicated that rebaudioside B is the sweetest among all the four compounds followed by stevioside, steviolbioside and dulcoside respectively which has also been reported earlier (Puri et al., 2011; Purkayastha, 2012). Non-sweet compounds, isosteviol and steviol are expected to have the least binding affinity towards STR and the same trend is obtained in our dock study. Protein–ligand interaction profile revealed that Asn 44, Ans 52, Ala 345, Pro 343, Ile 352, Gly 346, Gly 47, Ala 354, Ser 336, Thr 326, Ser 329 and Arg 56, Glu 105, Asp 215, Asp 216, Glu 148, Asp 258, Lys 255, Ser 104, Glu 217, Leu 51, Arg 52, Arg 247 were involved in the hydrogen bond interactions for T1R2 and TIR3, respectively. Interaction pattern and pose of each individual SG in term of glycone as well as aglycone part is very dissimilar as compared to others in the docking study (Fig. 6), which is quite reasonable
because all of them have unique sweetness properties, differing from other SGs. The present data also supports the above-mentioned assumption of the presence of different binding points within ATD of STR. Thus, several sets of amino acid residues are present in ATD, which are distantly apart from each other and by stimulating different set of residues one can produce characteristic sweet taste. Thus, on the basis of structural aspect as well as obtained docking poses of SG’s we have described the diverse taste of SG’s in the form of a multiple point stimulation model of SG’s for sweetness (Fig. 7). ATD domain of STR consists of several groups of active amino acids, namely active site A, B, C, D, E and so on. Monosaccharide moieties at C-13 and C-19 of SG’s stimulate different combination of active sites to produce characteristic sweet taste. Thus, diversity in the sweet taste of SG’s seems to be due to stimulation of primary active site residues by monosaccharide subunits at the C-13 position and some other type of allosteric modulation by C-19 monosaccharide subunits or vice versa. Least interaction and dock scores obtained for steviol (Figs. 4 and 5) indicated that aglycone part is only involved in providing glucose molecules an orientation enabling these glucose moieties to selectively interact with specific set of active sites to produce a characteristic taste.
4. Conclusion Homology models of T1R2 and T1R3 subunits were constructed using chain A of 3LMK protein and validated by using Ramachandran plot, PROCHECK and ERRAT server. Furthermore, docking of SG’s of known sweetness profile with modeled proteins
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
7
Fig. 6. Non-overlapping and diverse poses of SG’s docked on T1R2 (a) and T1R3 (b).
Fig. 7. Multiple point stimulation model of the sweetness mechanism of SG’s based on structure, varied sweet taste and diverse docked poses.
showed results which were in accordance with the existing experimental data. This high degree of correlation with existing data further justified our assumption that SG’s binds to ATD of STR to produce sweet taste. Several groups of researchers have shown different opinions regarding the rank order of sweetness intensity of multiple SG’s but with our docking scores we are able to predict the exact sweetness strength of various SG’s. Rebaudioside A was found to have maximum sweetness intensity followed by rebaudioside E, rebaudioside D, rebaudioside B, stevioside, steviolbioside and dulcoside respectively. Furthermore, Asn 44, Ans 52, Ala 345, Pro 343, Ile 352, Gly 346, Gly 47, Als 354, Ser 336, Thr 326 and Ser 329 are the main amino acids showing hydrogen bonding interaction with SG’s in case of T1R2 and Arg 56, Glu 105, Asp 215, Asp 216, Glu 148, Asp 258, Lys 255, Ssr 104, Glu 217, Leu 51 and Arg 52 amino acid residues have shown hydrogen bonding in case of T1R3. On the basis of dissimilarity in interaction pattern, poses of individual docked SG’s, its structural aspect and impact of C-13 and C-19 glucose molecules, we have concluded that multiple groups of amino acids so-called active sites are present in the ATD domain of STR and can be stimulated in different combinations to get diverse sweet taste. Using this model of sweetness we can easily explain how one SG is able to produce a unique sweet taste
which is different from other SG’s. The interaction of a particular SG with a specific set of active sites by mean of C-13 and C-19 glucose molecules is responsible for producing a specific taste. This, observation further justified that how an individual SG with difference only in sugar moieties at C-13 and/or C-19 positions of steviol are able to produce unique taste quality. The multiple point stimulation model is an important aspect for further exploring the different taste related points in ATD of STR. It will also be useful to enhance the taste quality and sweetness index of SG’s by further modifying structures and thus enabling their binding towards a specific point in STR. Thus, by improving dock score, we may be able to increase the sweetness intensity of SG’s and by manipulating the binding pattern taste quality can be improved.
5. Experimental 5.1. Ligand preparation SG’s were drawn using ChemBioDraw-12 (Fig. 8) and ligands were prepared using LegPrep wizard of Maestro 9.6. In ligand preparation step, 2D structure of the ligand is sufficiently modified
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
8
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
Fig. 8. Structures of SG’s used in the study.
to produce a better representative model of the actual ligand which is necessary for better docking results. 5.2. Template selection Amino acid sequences of T1R2 and T1R3 receptors were retrieved from NCBI (Accession: XP_94400 NP_689418.27, GI:112789566 and NP_689414.1, GI:91206396 respectively). Template required for homology modeling was taken from Protein Data Bank (PDB). A BLAST search of PDB database was done for suitable templates of T1R2 and T1R3. The BLAST search results were further used to create a phylogenetic tree using ClustalW2 to determine the most appropriate template for T1R2 and T1R3. ‘‘A’’ chain of the ligand binding domain of the Metabotropic glutamate receptor mGluR5 complexed with glutamate (PDB ID 3LMK), which is closest to the sequences of T1R2 and T1R3 in the phylogenetic tree, have suitable E-value and sequence identity of 25% with respect to T1R2 and 28% with respect to T1R3 was selected for homology modeling of both the subunits. 5.3. Homology modeling and protein preparation Homology models of T1R2 and T1R3 were constructed using the prime-v34012 module of the Schrodinger LLC (NY, USA) which is an automated protein modeling tool. The primary sequence of T1R2 and T1R3 in FASTA format was imported into structure prediction wizard present in Prime tool. BLAST search was performed and A chain of 3LMK being closest in the phylogenetic tree of TIR2 and T1R3, having suitable E-value and having a sequence identity of 25% and 28%, respectively was selected for homology modeling. Thus, using A chain of 3LMK, initial target-template alignment was generated and subsequently secondary structure prediction was performed by secondary structure prediction module (SSP) integrated with Prime. Prime structure prediction module is provided with two model building methods and we have selected energybased method to build the models. The generated model was further subjected to loop and side-chain optimization. The restrained minimization of final models of T1R2 and TIR3 were performed by
protein preparation wizard using the OPLS_2005 force field with default cut off root mean square distance (RMSD) of 0.30 Å. Finally, refined homology models were evaluated using RAMPAGE, PROCHECK and ERRAT web servers. The final proteins were now ready for docking procedure. 5.4. Grid generation Fully prepared and restrain minimized model of T1R2 and T1R3 were screened for possible binding sites using SitMap in Maestro 9.6 which had provided best five possible binding sites using default settings. The grid was generated on all the five predicted binding sites using default settings to define the area in the protein where docking was performed. 5.5. Molecular docking Docking was performed on already generated five grid sites using Glide docking application present in Schrödinger suite 2013. Glide is one of the most accurate method for docking studies as observed in several studies (Kellenberger et al., 2004; Zhou et al., 2007). There were three precessions provided with Maestro Glide docking, which include high throughput virtual screening (HTVC), standard precision (SP) and most accurate extra precision (XP). XP docking precision was used to dock SG’s. Further selection and interpretation of results were done on the basis of best agreement with the experimental sweetness profile of SG’s. Acknowledgements Authors are grateful to Department of Science and Technology, New Delhi, India for providing financial assistance during the course of the work. Dually acknowledge to Dr. Alpna Saini, Assistant Professor, Centre for Comparative literature for editing and linguistic correction of the manuscript. Authors are also thankful to the Honorable Vice - Chancellor for providing the necessary facilities at Central University of Punjab, Bathinda, India.
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
Mayank, V. Jaitak / Phytochemistry xxx (2015) xxx–xxx
References Agurs-Collins, T., Rosenberg, L., Makambi, K., Palmer, J.R., Adams-Campbell, L., 2009. Dietary patterns and breast cancer risk in women participating in the Black Women’s Health Study. Am. J. Clin. Nutr. 90, 621–628. Brauner-Osborne, H., Wellendorph, P., Jensen, A.A., 2007. Structure, pharmacology and therapeutic prospects of family C G-protein coupled receptors. Curr. Drug Targets 8, 169–184. Ceunen, S., Geuns, J.M., 2013. Steviol glycosides: chemical diversity, metabolism, and function. J. Nat. Prod. 76, 1201–1228. Colovos, C., Yeates, T.O., 1993. Verification of protein structures: patterns of nonbonded atomic interactions. Protein Sci. 2, 1511–1519. Conn, P.J., Pin, J.-P., 1997. Pharmacology and functions of metabotropic glutamate receptors. Annu. Rev. Pharmacol. 37, 205–237. de Oliveira, B.H., Packer, J.F., Chimelli, M., de Jesus, D.A., 2007. Enzymatic modification of stevioside by cell-free extract of Gibberella fujikuroi. J. Biotechnol. 131, 92–96. Fernstrom, J.D., Munger, S.D., Sclafani, A., de Araujo, I.E., Roberts, A., Molinary, S., 2012. Mechanisms for sweetness. J. Nutr. 142, 1134S–1141S. Hellfritsch, C., Brockhoff, A., Stähler, F., Meyerhof, W., Hofmann, T., 2012. Human psychometric and taste receptor responses to steviol glycosides. J. Agr. Food Chem. 60, 6782–6793. Jaitak, V., Bandna, B.S., Kaul, V., 2009a. An efficient microwave-assisted extraction process of stevioside and rebaudioside-A from Stevia rebaudiana (Bertoni). Phytochem. Anal. 20, 240–245. Jaitak, V., Gupta, A., Kaul, V., Ahuja, P.S., 2008. Validated high-performance thinlayer chromatography method for Steviol glycosides in Stevia rebaudiana. J. Pharm. Biomed. Anal. 47, 790–794. Jaitak, V., Kaul, V.K., Kumar, N., Singh, B., Savergave, L., Jogdand, V., Nene, S., 2009b. Simple and efficient enzymatic transglycosylation of stevioside by bcyclodextrin glucanotransferase from Bacillus firmus. Biotechnol. Lett. 31, 1415–1420. Jaitak, V., Kaul, K., Kaul, V.K., Singh, V., Singh, B., 2011. Stevia rebaudiana: a natural substitute for sugar. In: Genetic Resources, Chromosome Engineering, and Crop Improvement. CRC Press, pp. 885–910. Kellenberger, E., Rodrigo, J., Muller, P., Rognan, D., 2004. Comparative evaluation of eight docking tools for docking and virtual screening accuracy. Proteins 57, 225–242. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Levine, W., Dyer, A.R., Shekelle, R.B., Schoenberger, J.A., Stamler, J., 1990. Post-load plasma glucose and cancer mortality in middle-aged men and women: 12-year follow-up findings of the Chicago Heart Association Detection Project in Industry. Am. J. Epidemiol. 131, 254–262. Li, X., Staszewski, L., Xu, H., Durick, K., Zoller, M., Adler, E., 2002. Human receptors for sweet and umami taste. Proc. Natl. Acad. Sci. U.S.A. 99, 4692–4696. Lovell, S.C., Davis, I.W., Arendall, W.B., de Bakker, P.I., Word, J.M., Prisant, M.G., Richardson, J.S., Richardson, D.C., 2003. Structure validation by Ca geometry: /, w and Cb deviation. Proteins 50, 437–450.
9
Masuda, K., Koizumi, A., Nakajima, K.-I., Tanaka, T., Abe, K., Misaka, T., Ishiguro, M., 2012. Characterization of the modes of binding between human sweet taste receptor and low-molecular-weight sweet compounds. PLoS ONE 7, e35380. Michel, K.B., Solomon, C.G., Hu, F.B., Rosner, B.A., Hankinson, S.E., Colditz, G.A., Manson, J.E., 2003. Type 2 diabetes and subsequent incidence of breast cancer in the Nurses’™ Health Study. Diabetes Care 26, 1752–1758. Nie, Y., Vigues, S., Hobbs, J.R., Conn, G.L., Munger, S.D., 2005. Distinct contributions of T1R2 and T1R3 taste receptor subunits to the detection of sweet stimuli. Curr. Biol. 15, 1948–1952. Ogden, C.L., Carroll, M.D., Curtin, L.R., McDowell, M.A., Tabak, C.J., Flegal, K.M., 2006. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 295, 1549–1555. Paital, B., Kumar, S., Farmer, R., Chainy, G.B.N., 2013. In silico prediction of 3D structure of Mn superoxide dismutase of Scylla serrata and its binding properties with inhibitors. Interdiscip. Sci. Comput. Life Sci. 5, 69–76. Prakash, I., DuBois, G., Clos, J., Wilkens, K., Fosdick, L., 2008. Development of rebiana, a natural, non-caloric sweetener. Food Chem. Toxicol. 46, S75–S82. Patel, A.V., Rodriguez, C., Bernstein, L., Chao, A., Thun, M.J., Calle, E.E., 2005. Obesity, recreational physical activity and risk of pancreatic cancer in a large U.S. cohort. Cancer Epidemiol. Biomarkers Prev. 14, 459–466. Persson, P.G., Ahlbom, A., Hellers, G., 1992. Diet and inflammatory bowel disease: a case-control study. Epidemiology 3, 47–52. Pin, J.-P., Galvez, T., Prézeau, L., 2003. Evolution, structure, and activation mechanism of family 3/C G-protein-coupled receptors. Pharmacol. Ther. 98, 325–354. Pizzo, G., Giuliana, G., Milici, M.E., Giangreco, R., 2000. Effect of dietary carbohydrates on the in vitro epithelial adhesion of Candida albicans, Candida tropicalis, and Candida krusei. New Microbiol. 23, 63–71. Purkayastha, S., 2012. Glucosylated SG composition as a flavor modifier. U.S. Patent Application 14/005,852. Puri, M., Sharma, D., Tiwari, A.K., 2011. Downstream processing of stevioside and its potential applications. Biotechnol. Adv. 29, 781–791. Sardesai, V.M., Waldshan, T.H., 1991. Natural and synthetic intense sweeteners. J. Nutr. Biochem. 2, 236–244. Schiffman, S.S., Gatlin, L.A., Sattely-Miller, E.A., Graham, B.G., Heiman, S.A., Stagner, W.C., Erickson, R.P., 1994. The effect of sweeteners on bitter taste in young and elderly subjects. Brain Res. Bull. 35, 189–204. Sheiham, A., 2001. Dietary effects on dental disease. Public Health Nutr. 4, 569–591. Walters, D.E., Hellekant, G., 2006. Interactions of the sweet protein brazzein with the sweet taste receptor. J. Agr. Food Chem. 54, 10129–10133. Weiderpass, E., Gridley, G., Nyren, O., Ekbom, A., Persson, I., Adami, H.O., 1997. Diabetes mellitus and risk of large bowel cancer. J. Natl Cancer Inst. 89, 660– 661. Xu, H., Staszewski, L., Tang, H., Adler, E., Zoller, M., Li, X., 2004. Different functional roles of T1R subunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. U.S.A. 101, 14258–14263. Zhou, Z., Felts, A.K., Friesner, R.A., Levy, R.M., 2007. Comparative performance of several flexible docking programs and scoring functions: enrichment studies for a diverse set of pharmaceutically relevant targets. J. Chem. Inf. Model. 47, 1599– 1608.
Please cite this article in press as: Mayank, , Jaitak, V. Interaction model of steviol glycosides from Stevia rebaudiana (Bertoni) with sweet taste receptors: A computational approach. Phytochemistry (2015), http://dx.doi.org/10.1016/j.phytochem.2015.05.006
