MODEL
Our team conducted molecular docking experiments to reveal the binding mode between the sweet protein and the human sweet taste receptor protein TAS1R2-TAS1R3. Through the DNS assay and electronic tongue detection, we further confirmed that Thaumatin proteins can exhibit enzymatic activity and significantly enhance the sweetness of the system.
1.Acquisition of the Sweet Taste Receptor Protein Model
The cryo-electron microscopy three-dimensional structural model (with a resolution of 3.3 Å) of the human sweet taste receptor (TAS1R2-TAS1R3 heterodimer) was obtained using PDB ID: 9NOS, with the specific steps as follows:
Structure download: Download the structural file of 9NOS from the Protein Data Bank (PDB). This structure contains two subunits, TAS1R2 and TAS1R3, covering the extracellular Venus flytrap domain (VFT), cysteine-rich domain (CR), and seven-transmembrane domain (TM), which are the core functional units for the binding of the sweet taste receptor to ligands.
Pretreatment: Use AutoDock Tools software to pretreat the structure, including removing crystal water, buffer molecules, and redundant ligands; optimize the receptor structure by adding hydrogen atoms and assigning Gasteiger charges to ensure its stability and accuracy during the docking process.
2. AlphaFold Modeling of Sweet Protein Molecules
The AlphaFold tool was used to predict the three-dimensional structures of the sweet proteins TLP-A and TLP-B, with the specific steps as follows:
Sequence information:
>UYG53795.1 thaumatin family protein (plasmid) [Comamonas endophytica] TLP-A MPDQSIAPVAPVRPPGAFATGAQPTRLRVTNQCTEPIWLQYAVGSGASDQGNVLGPQQIQLQSGESHDYPIPPGRLEATRFWPKTGCDAAGLNCKIGMSSPPCPATGCMPPIESKFEATFAATDCANEGYRCMTYWNASQVDGYTLPISVFPKGPGVQPPGTPGIEQCVESRCDALDLLQCPSNERIGKRTLDLRVFAPDDRSKLVGCLSPCKAHNYPPPFGLGLPENDEHGLMLCCPEGVTPEECRQVPVLSTQYVQYINKVCPNVYAYSYDDRVGLHKCPAQTQYEVVFCGGTGSGPTATLLPQSAPKPRPRPAYPTTFGP
>WNG29905.1 thaumatin family protein [Cystobacter fuscus] TLP-B MKAYGVVLISMLSVAGCGAPPDEGETAMRGETHTEELGTESHALGCAIANDGQTTLRFINKCAVTVNFAGSNITGGALASGQEACRTIGSNTQLMRAQRYWGSRSGEALGAGKVSLAEFTFNEPFYSWKSYDWFNLSHVDAHNLPLKIIPYELGSGTTCSTMTRSCPQNLLANCPTQGQLRNAAGKVIACVSPNRDDANNPVARFFDTACAQAYSWSGDNSSMASCNAEDFDIVFCPSN
Structure prediction: Input the sequences into the AlphaFold platform, select the AlphaFold2 model (which has significant advantages in the accuracy of protein structure prediction), and set the sequence coverage to 100% to ensure structural integrity. After the prediction is completed, output the three-dimensional structure file in PDB format.
Structure validation: Check the structural model through AlphaFold's visualization tools to determine the integrity of the overall structure; use the Ramachandran plot to evaluate the rationality of the secondary structure, and screen out models with no obvious conformational abnormalities for subsequent molecular docking.
3. Molecular Docking Experiment (using Z-dock online analysis)
The Z-dock online tool (https://zdock.umassmed.edu/) was used for molecular docking between sweet protein TLP and the sweet taste receptor:
Ligand and receptor preparation: Convert the AlphaFold-predicted TLP structure (ligand) and the preprocessed 9NOS structure (receptor) into PDB format, ensuring there are no redundant atoms or unreasonable bond lengths.
Docking parameter settings: Based on the ligand-binding characteristics of the sweet taste receptor, the docking region was limited to the VFT domain of TAS1R2 (known as the key binding pocket for sweet ligands); For sampling: a rotation angle step size of 6° was used to generate 10,000 initial conformations with a clustering radius of 4.0 Å. Default docking parameters were selected, and the top 10 docking conformations were output, with Z-score (higher scores indicating stronger binding stability) as the main screening index.
4. Preliminary Analysis of Molecular Docking Results
The docking results were analyzed using visualization tools and interaction analysis methods as follows:
Optimal conformation screening: Compare the Z-scores of the Top10 conformations and select the one with the highest score as the optimal binding mode.
Binding site analysis: Use PyMOL software to observe the binding sites between sweet protein TLP and the sweet taste receptor, with a focus on verifying whether interactions occur with the VFT domain of TAS1R2.
Identification of interaction types: With reference to protein interaction force analysis methods, identify and count hydrogen bonds (donor-acceptor distance < 3.5Å), hydrophobic interactions, etc., to evaluate the stability of ligand-receptor binding.
1. Results of Sweet Sweet Taste Receptor Protein Model Acquisition
The structural file with PDB ID: 9NOS was successfully obtained. After preprocessing, the complete sequences of TAS1R2 (Chain A) and TAS1R3 (Chain B) were retained, with 12 crystal water molecules and 1 redundant PEG molecule removed. After the addition of hydrogen atoms, the VFT domain (LB1 and LB2 subdomains) of TAS1R2 showed a uniform charge distribution and a clear binding pocket structure, meeting the receptor requirements for the docking experiment.
Predicted structural features: The structure of sweet protein TLP-A predicted by AlphaFold2 consists of 12 β-sheets and 6 α-helices, presenting an overall compact globular structure. The conformational state of the flexible loop regions on the surface (e.g., residues 40-60) is reasonable, and the pLDDT scores (prediction confidence) are all > 85, indicating high structural reliability. The structure of sweet protein TLP-B predicted by AlphaFold2 comprises 12 β-sheets and 2 α-helices, also showing an overall compact globular structure. The conformations of the flexible loop regions on the surface are reasonable, with all pLDDT scores (prediction confidence) > 85, which demonstrates high structural reliability.
Structural validation: The Ramachandran plot shows that more than 95% of the residues are located in the allowed regions, which does not affect the overall structural stability and meets the requirements of the docking experiment for the ligand model.
3. Molecular Docking Results (Z-dock)
Binding site: In the optimal conformation, the sweet protein TLP mainly binds to the Venus Flytrap (VFT) domain of TAS1R2, and its binding pocket overlaps with that of artificial sweeteners such as sucralose and aspartame. This is consistent with the ligand-binding specificity of the sweet taste receptor.
4. Preliminary Analysis of Molecular Docking Results
The key interactions between the sweet taste receptor and the sweet protein were shown in Figure 1 and Figure 2. In the optimal conformation, the sweet protein TLP-A forms 5 sets of hydrogen bonds with TAS1R2. The amino acid residues in TLP-A form hydrogen bonds with S87, D406, and S408 of TAS1R2, respectively. The sweet protein TLP-B forms 3 sets of hydrogen bonds with TAS1R2. The amino acid residues in TLP-B form hydrogen bonds with W341, N368, and D364 of TAS1R2, respectively.
Figure 1 The interaction between the sweet taste receptor and TLP-A
The TAS1R2 region involved in the interaction is a key binding site for the combination of sweet ligand proteins and sweet molecules. No significant interaction between Thaumatin and the TAS1R3 subunit was observed, which supports that TAS1R2 is the main target of natural sweet proteins.
Figure 2 The interaction between the sweet taste receptor and TLP-B
A standard curve was constructed by measuring the absorbance (at 540 nm) of DNS-reacted glucose solutions with known concentrations, as shown in Figure 3. The linear equation is y = 0.9596x-0.0189 with a high correlation coefficient R2 = 0.9907, indicating a good linear relationship between absorbance and glucose concentration in the tested range.
Figure 3 Glucose standard curve
The Single-enzyme activity results showed that Thaumatin-A had the highest enzyme activity, followed by Thaumatin-B, Thaumatin-D, and Thaumatin-C.
Figure 4 Graphs of single-enzyme activity comparison
In the dual-enzyme combinations, the mixture of Thaumatin-C&D displayed the highest activity, followed by Thaumatin-A&C, Thaumatin-A&B, Thaumatin-A&D, Thaumatin-B&D, and Thaumatin-B&C.
Figure 5 Graphs of dual-enzyme activity comparison
In the triple-enzyme combinations, the mixture of Thaumatin-A&B&D displayed the highest activity, followed by Thaumatin-A&B&C, and Thaumatin-B&C&D.
Figure 6 Graphs of triple-enzyme activity comparison
Finally, the above three combinations with the best activity were selected and compared with the mixed system containing all four enzymes. As shown in Figure 7, the mixture of Thaumatin-A&B&C&D displayed the highest activity, followed by Thaumatin-C&D, Thaumatin-A&B&D, and Thaumatin-A.
Figure 7 Graphs of optimal combinations comparison
To characterize the taste properties of samples, electronic tongue detection was performed. The differences in taste characteristics among different samples were shown in Figure 8A, with each sample group forming distinct clusters, indicating significant differences in taste profiles between samples. Sweetness values for different samples were shown in Figure 8B. The 4% Sucrose group served as a control with extremely low sweetness. Thaumatin-A had the highest sweetness, followed by Thaumatin-C, Thaumatin-D, and Thaumatin-B. These results indicate that Thaumatin proteins can significantly enhance the sweetness of the system, with varying degrees of sweetness-enhancing effects among different Thaumatin variants.
Figure 8 Result of electronic tongue detection
The Molecular docking experiments employed AlphaFold modeling and protein-protein molecular docking to reveal the binding mode between the sweet protein and the human sweet taste receptor protein TAS1R2-TAS1R3. The TAS1R2 region involved in the interaction is a key binding site for sweet ligand proteins and sweet molecules, indicating that TAS1R2 is the main target of natural sweet proteins. It confirms that the sweet protein TLP can bind to the sweet taste receptor in a directional manner, thereby enabling the human body to perceive sweetness.
The DNS assay and electronic tongue detection can be used as supplementary methods for molecular docking experiments. The former detected changes in the reducing sugar content of different treated groups, which indirectly confirms the enzymatic activity of Thaumatin. The latter, on the other hand, objectively analyzes the sweetness of different samples by simulating the human taste system. These two methods indicate that Thaumatin proteins can exhibit enzymatic activity and significantly enhance the sweetness of the system, providing more diverse data support for the molecular docking experiments.
