Identification of Oncorhynchus mykiss nebulin-derived peptides as bitter taste receptor TAS2R14 blockers by in silico screening and molecular docking

Zhipeng Yu a, Yingxue Wang a, Wenzhu Zhao a,*, Jianrong Li a,*, David Shuian b, Jingbo Liu c a College of Food Science and Engineering, Bohai University, Jinzhou 121013, PR China b Institute of Drug Discovery Technology, Ningbo University, Ningbo 315211, PR China c Lab of Nutrition and Functional Food, Jilin University, Changchun 130062, PR China

A R T I C L E I N F O

Keywords: Oncorhynchus mykiss TAS2R14 Bitter taste receptor blockers Molecular docking Electronic tongue

A B S T R A C T

Human bitter taste receptor TAS2R14 (T2R14) can widely perceive bitterness, which has always been an issue for people to overcome. This study was aimed at identifying bioactive peptides obtained from Oncorhynchus mykiss nebulin hydrolysates as bitter taste receptor blockers by physicochemical property prediction, molecular dock- ing, and in vitro determination of bitterness intensity using electronic tongue. EXploration of the interaction mechanism of these peptides with T2R14 by molecular docking models indicated that peptides ADM and ADW
had high affinities for T2R14 to block the binding of bitter substances into the receptor. Addition of ADM and ADW to quinine caused reduction in bitterness intensity, with IC50 values of 420.32 ± 6.26 μM and 403.29 ±
4.10 μM, respectively. Hydrogen bond interaction and hydrophobic interaction were responsible for manifesting
the high affinities of these peptides for the receptor. Residues Thr86, Asp168, and Phe247 may be the key amino acids within the binding site.

1. Introduction

Humans are able to perceive five basic tastes, namely sour, sweet, bitter, salty, and umami (Yarmolinsky, Zuker, & Ryba, 2009). Bitterness, an unpleasing taste sensation, is not only derived from compounds produced naturally by plants, such as bitter gourd, goldthread, and chicory, but is also derived from compounds produced during food proteolysis and drug processing (FitzGerald & O’Cuinn, 2006; Kohl, Behrens, Dunkel, Hofmann, & Meyerhof, 2013). People often avoid foods that are bitter but highly nutritious or drugs that are efficacious for treating their respective conditions, as bitterness is ordinarily a warning of potential toXicity (Drewnowski & Gomez-Carneros, 2000). Accord- ingly, suppressing bitterness is of great commercial value for the food and pharmaceutical industries.
Human bitter taste perception is mediated by at least 25 bitter taste receptors (TAS2Rs) that belong to the G protein-coupled receptor family (Jaggupilli, Howard, Upadhyaya, Bhullar, & Chelikani, 2016). Twenty- five TAS2Rs that recognize thousands of bitter substances have been identified. Previous studies have demonstrated that the detection of bitter substances correlated with the molecular tuned ranges of TAS2Rs
(Meyerhof et al., 2010). TAS2Rs have single ligand-binding pockets that recognize many structurally different bitter compounds while main- taining high selectivity (Born, Levit, Niv, Meyerhof, & Behrens, 2013; Brockhoff, Behrens, Niv, & Meyerhof, 2010). Interestingly, the com- bined activities of three widely receptive bitter taste receptors, TAS2R10 (T2R10), TAS2R14 (T2R14), and TAS2R46 (T2R46), may be adequate for detecting about half of all bitter compounds recognized by humans (Meyerhof, et al., 2010). T2R14 is one of the most broadly tuned bitter taste receptors (Behrens, Brockhoff, Kuhn, Bufe, Winnig, & Meyerhof, 2004; Roland, Vincken, Gouka, van Buren, Gruppen, & Smit, 2011). T2R14 displays a wide range of tuning towards various structurally diverse bitter compounds, including azathioprine, benzoin, diphenhy- dramine, and potent neurotoXins like (-)-α-thujone and picrotoXinin (Behrens, Brockhoff, Kuhn, Bufe, Winnig, & Meyerhof, 2004; Meyerhof, et al., 2010). Therefore, T2R14 is an effective target for suppressing bitterness.
One proven method for suppressing bitterness is to add sweeteners or fragrances to confuse the nerve impulses generated by bitter molecules to the brain. However, this approach is not feasible for bioactive peptides that enable bitterness suppression can solve this Syst`emes Biovia, San Diego, CA, USA). Therefore, the peptides with drawback. At present, food-derived bioactive peptides have attracted extensive attention in the fields of food science, chemistry, and biology because they are safe and nontoXic, have diverse functions, and can be easily modified (Yu, Kan et al., 2020). For example, two beef protein- derived peptides, AGDDAPRAVF and ETSARHL, were identified as good water solubility, biological activity, and screened out.

2. Materials and methods

2.1. Materials and reagents
Quinine was purchased from Shanghai Yi’en Chemical Technology Co., Ltd. (Shanghai, China). Tartaric acid, ethyl alcohol, hydrochloric acid, potassium chloride, and potassium hydroXide were obtained from Youpu Chemical Reagent Co., Ltd. (Tianjin, China). All reagents and chemicals used were of analytical grade.

2.2. In silico enzymolysis of Oncorhynchus mykiss nebulin
Oncorhynchus mykiss nebulin (NCBI accession: XP_021453115) was retrieved from the National Center for Biotechnology Information (NCBI), available at https://www.ncbi.nlm.nih.gov/ (accessed July 25th, 2020). Nebulin was in silico hydrolyzed by three commercially available enzymes, pepsin (EC 3.4.23.1), trypsin (EC 3.4.21.4), and chymotrypsin (EC 3.4.21.1), using the EXPASy PeptideCutter (http://w eb.expasy.org/peptide_cutter/). Pepsin preferentially cleaves at residues Phe, Tyr, Trp, and Leu in position P1 or P1′. Trypsin preferentially cleaves at residues Arg and Lys in position P1 with higher rates for Arg, especially at high pH. Chymotrypsin preferentially cleaves with high specificity at residues Trp, Tyr, and Phe in position P1, and it also cleaves with a low specificity at residues Leu, Met, and His in position P1 (Keil, 1987). The hydrolyzed peptides were then used for the following prediction.

2.3. Water solubility, biological activity, and toxicity prediction of peptides
The water solubilities of the hydrolyzed peptides were predicted using the peptide property calculator provided online at: http://www. innovagen.com/proteomics-tools (accessed July 28th, 2020) (Lafarga, O’Connor, & Hayes, 2015). The peptide sequences were entered into the online program, then the results showed whether the peptides had good water solubility. PeptideRanker (http://distilldeep.ucd.ie/Pepti deRanker/ (accessed August 3th, 2020)) was used to predict the bio- logical activities of the hydrolyzed peptides. The peptides were considered biologically active when their scores exceeded 0.5 (Mooney, Haslam, Pollastri, & Shields, 2012). In addition, the toXicities, including Ames mutagenicity, developmental toXicity potential, and skin sensiti- zation, of the selected peptides were predicted using the ToXicity Pre- diction tool in the Discovery Studio (DS) 2017 R2 software (Dassault The BitterDB Protein Data Bank (http://bitterdb.agri.huji.ac.il/dbbi tter.php) was used to download the X-ray crystal structure of T2R14 (BitterDB ID: 14, accessed August 12th, 2020) with 6883 amino acids (Di Pizio & Niv, 2015). The T2R14 was prepared by adding the hydrogen atoms. The active site was defined from receptor cavities using the Receptor-Ligand Interactions module in the DS 2017 R2 software. Among the 7 sites in the crystal structure of T2R14, site 1 was defined as the active site because its docking result was most consistent with pre- vious study (Karaman et al., 2016). The remaining 6 sites were deleted. The structures of the selected peptides were constructed in the Macro- molecules module. The Minimize Ligands and Prepare Ligands tools of the Small Molecules module were used to optimize the energetic con- formations of the peptides with the CHARMm force field. In the CDOCKER protocol of the DS 2017 R2 software, molecular docking simulations were conducted using the following coordinates: X = 1.195, y = 2.034, z = 12.00, and a radius of 17.4 Å.

2.4. Molecular docking of peptides with T2R14 non-toXicity were bitter taste receptor T2R4 blockers (Zhang, Alashi, Singh, Liu, Chelikani, & Aluko, 2018). Oncorhynchus mykiss is a species of fish with an abun- dant amount of protein from which to obtain bioactive peptides, and its rich nutritional value needs to be further explored. Furthermore, in silico enzymolysis, physicochemical property prediction, and molecular docking simulations are powerful tools that can aid in the rapid and efficient identification of bitter taste receptor blockers from a large number of peptides derived from Oncorhynchus mykiss nebulin (Seifert & Lang, 2008).
The purpose of this study was to identify bioactive peptides as effi- cacious bitter taste receptor blockers via in silico enzymolysis, physico- chemical property prediction, and molecular docking simulations. Subsequently, in vitro validation of potential peptides for suppressing bitterness was performed using an electronic tongue assay. Furthermore, the interaction mechanism of promising peptides with T2R14 was analyzed by molecular docking models.

2.5. Peptides synthesis
The AAPPTEC Apex 396 peptide synthesizer was utilized to experi- mentally synthesize the selected peptides. The molecular weights and purities of the peptides were validated by HPLC coupled with electro- spray ionization-quadrupole-mass spectrometry (Nanjing Yuanpeptide Biotech Co., Ltd., Nanjing, Jiangsu, China). The peptides were stored at —20℃ under desiccation until further use.

2.6. Measurement of the bitterness intensity via an electronic tongue
The method was according to the previous study with some modi- fications (Yu, Kang, et al., 2020). The SA402B electronic tongue (INSENT, Kanagawa, Japan), equipped with five test sensors (CA0, C00, AE1, CT0, AAE, and GL1) and two reference electrodes, was used to measure the in vitro bitterness intensity reduction activity of the syn- thesized peptides at different concentrations (0.075, 0.1125, 0.15, 0.1875, and 0.25 mg/mL). The reference solution consisted of 0.3 mM tartaric acid and 30 mM KCl. Quinine (1 mM), which is known to acti- vate many T2Rs, was selected as the blank control, and the peptide LEGSLE was selected as the positive control (Xu et al., 2019). The selected peptides were dissolved in 30 mL water, and the resulting so- lution was added to 50 mL quinine solution (1 mM), which was stirred uniformly. The treatment of the positive control was the same as above. After setting the parameters and installing the electrodes, the mea- surement occurred as follows: Firstly, the sensors were cleaned in pos- itive and negative solutions for 90 s, followed by the other two reference solutions for 120 s; Secondly, the sensors were balanced in conditioning solution for 30 s; Next, each sample was measured for 30 s; Lastly, the sensors were washed twice for 3 s and steeped in reference solution for 30 s to measure the aftertaste value. Each sample was measured three times repeatedly. The bitterness intensity was evaluated by the sensor potentiometric difference between the conventional reference electrodes and the bitter sensor (Phat, Moon, & Lee, 2016).

3. Results and discussions

3.1. Virtual enzymolysis and physicochemical property prediction of peptides
Oncorhynchus mykiss nebulin was in silico hydrolyzed to 687 peptides by pepsin, trypsin, and chymotrypsin. Good water solubility, biological activity, and non-toXicity are considered important physicochemical properties that affect the application of bioactive peptides in the food industry. The screening results of water solubility showed that 483 peptides had good water solubilities, while the other peptides had poor water solubilities. Among these 483 peptides, only 31 bioactive peptides with corresponding peptide scores exceeding the 0.5 threshold in Pep- tideRanker were screened for further toXicity prediction (Table 1). Following the prediction of the toXicities, including Ames mutagenicity, developmental toXicity potential, and skin sensitization, 22 peptides GKP, ADM, LPPDAPEL, PASDNPVL, DW, SM, DM, DAM, GSG, NDPQF, CSPVDM, PDSM, ADW, DGM, PADAPEF, PADSPQF, SDW, SPVDM, MPDSM, IPPDM, SIPDRPEF, and PDIM were considered non-toXic, non- irritant, and non-mutagenic.

3.2. Molecular docking
The CDOOCKER program, a flexible and accurate docking tool in the DS 2017 R2 software, was applied to identify bioactive peptides exhibiting high binding affinity for the T2R14 receptor, and explore action mechanism of the peptides as T2R14 blockers. The higher the ‘–CDOCKER_Energy’ score, the higher the binding affinity of the peptide ‘–CDOCKER_Energy’ than the peptides LEGSLE (66.21 kcal/mol) and PIGNIN (42.66 kcal/mol). In previous study, the peptides LEGSLE, and PIGNIN were reported as the most likely T2R14 blockers in hen protein hydrolysates (Xu, et al., 2019). It could be concluded that the binding affinities of the peptides SPVDM, CSPVDM, ADW, and ADM to T2R14 were in descending order, and were better than the peptides LEGSLE and PIGNIN. Therefore, these four bioactive peptides were selected for in vitro validation of quinine- inhibitory activity, and the peptide LEGSLE was considered positive control.

3.3. In vitro validation of the bitterness intensity reduction activity via electronic tongue
The peptides SPVDM, CSPVDM, ADM, ADW, and positive control LEGSLE were experimentally synthesized, and the bitter inhibition rates of these peptides to quinine were determined by the SA402B electronic tongue. When the concentration of the peptides was 0.15 mg/mL, the inhibition rates of ADM, ADW, LEGSLE, SPVDM, and CSPVDM were 51.10%, 43.81%, 36.63%, 35.83%, and 34.13%, respectively. These results indicated that ADM and ADW exhibited more effective quinine- inhibitory activity than LEGSLE, SPVDM, and CSPVDM, while SPVDM and CSPVDM exhibited less effective quinine-inhibitory activity than the positive control LEGSLE. Therefore, ADM and ADW possessed bitterness intensity reduction activity as the most effective bitter taste receptor blockers, with IC50 values of 420.32 ± 6.26 μM and 403.29 ± 4.10 μM, respectively. It was previously reported that the IC50 values of beef protein-derived peptides ETSARHL, AGDDAPRAVF, and AAMY as T2R4 blockers were 118 ± 40 μM, 85 ± 20 μM, and 119 ± 35 μM, respectively
Fig. 1. The docking interactions of ADM with T2R14. (A) The crystal structure of T2R14 and its binding site with ADM. (B) The 2D diagram of the docking of ADM with T2R14. (C) The 3D diagram of the docking of ADM with T2R14. (D) The 3D hydrogen bond surface plot at the binding site. (Zhang, Alashi, Singh, Liu, Chelikani, & Aluko, 2018). And the IC50 values of γ-aminobutryic acid (GABA) and Nα,Nα-bis(carboXymethyl)-L- lysine (BCML) as T2R4 blockers were 3.2 0.3 μM and 59 18 nM, respectively (Pydi, Sobotkiewicz, Billakanti, Bhullar, Loewen, & Cheli-kani, 2014). Previous studies have demonstrated that the IC50 values of bitter taste receptor blockers range orders of magnitude from μM to nM. Thus, the peptides ADM and ADW were identified as effective bitter taste receptor blockers. It has been reported that beef protein-derived pep- tides ETSARHL and AGDDAPRAVF blocked T2R4. In addition, hen protein-derived peptides GDDAPR, LELNQ, LEGSLE, and PIGNIN blocked T2R4, T2R7, and T2R14 (Xu, et al., 2019; Zhang, Alashi, Singh, Liu, Chelikani, & Aluko, 2018). Those peptides can block specific T2Rs to reduce bitterness intensity of bitter ingredients.

3.4. Action mechanism of bioactive peptides as bitter taste receptor blockers
The peptides screened above were found to block the interactions between bitter substances and the receptor by binding to T2R14 owing to their higher affinities. The docking diagrams of the ligands ADM,
Fig. 1. (continued). ADW, SPVDM, and CSPVDM with T2R14 were performed to analyze the interactions and key amino acids implicated in the binding of the bitter taste receptor blockers to the receptor.
The X-ray crystal structure of T2R14 and its binding site with ADM are shown in Fig. 1(A). Fig. 1(B) and 1(C) present the 2D and 3D dia- grams of the ADM-T2R14 complex, respectively. Residues Thr86 and
Fig. 2. The docking interactions of ADW with T2R14. (A) The crystal structure of T2R14 and its binding site with ADW. (B) The 2D diagram of the docking of ADW with T2R14. (C) The 3D diagram of the docking of ADW with T2R14. (D) The 3D hydrogen bond surface plot at the binding site.
Asp168 generated conventional hydrogen bond interactions with ADM, with distances of 1.94 Å and 1.83 Å, respectively. Residue Asp168 generated carbon hydrogen bond interaction with ADM, with a distance of 2.56 Å. Residue Phe247 generated Pi-Alkyl interaction with ADM, with a distance of 4.88 Å. The X-ray crystal structure of T2R14 and its binding site with ADW are shown in Fig. 2(A). Fig. 2(B) and 2(C) correspond to the 2D and 3D diagrams of the ADW-T2R14 complex, respectively. Residues Thr86 and Asp168 generated conventional hydrogen bond interactions with ADW, with distances of 1.98 Å and 2.40 Å, respectively. Residue Asp168 generated carbon hydrogen bond interaction with ADW, with a distance of 3.00 Å. Residue Trp89 generated Pi-Pi Stacked interactions with ADW, with distances of 4.54 Å and 5.27 Å, respectively. Residue Phe243 generated Pi-Pi T-shaped interaction with ADW, with a distance of 5.09 Å. Residue Phe247 generated Pi-Pi T-shaped interactions with ADW, with distances of 4.80 Å and 5.05 Å, respectively.
Considering the docking results of these four peptides in the T2R14 “” Indicates hydrogen bond interaction, “” indicates carbon hydrogen bond interaction, “” indicates hydrophobic interaction, and “” indicates electro- static interaction. active site, hydrogen bond interaction and hydrophobic interaction played the most important role in the formation of the ligand-receptor complexes (shown in Table 3). Fig. 1(D) and 2(D) present the 3D hydrogen bond surface plots of ADM and ADW at the binding site, respectively. There were 3, 3, 4, and 6 hydrogen bonds and 1, 5, 7, and 8 hydrophobic bonds in the ADM-T2R14, ADW-T2R14, SPVDM-T2R14, and CSPVDM-T2R14 complexes, respectively. Residues Thr86 and Phe247 participated in the interactions of all these complexes. Residue Asp168 simultaneously generated hydrogen bond interaction and car- bon hydrogen bond interaction in three complexes except for the SPVDM-T2R14 complex, which was likely why SPVDM had the highest docking energy value but had lower in vitro activity than ADM and ADW. It could be concluded that residues Thr86, Asp168, and Phe247 might be the key amino acids that facilitated the binding of the bitter taste re- ceptor blockers to T2R14. Compared to ADM and ADW, more amino acid residues including Trp66, Trp89, Ser167, Ser169, Phe175, Leu178, Phe186, Phe243, and Ile262, binding to SPVDM and CSPVDM, caused higher docking energy values of SPVDM and CSPVDM, but these resi- dues had little effect on the practical activity.

4. Conclusions
In the present study, four peptides SPVDM, CSPVDM, ADM, and ADW with good water solubility, bioactivity, and non-toXicity were screened from Oncorhynchus mykiss nebulin through virtual enzymol- ysis, physicochemical property prediction, and molecular docking. In vitro validation of the bitterness intensity reduction via electronic tongue identified the peptides ADM and ADW as effective bitter taste receptor T2R14 blockers. The IC50 values of ADM and ADW were 420.32 6.26 μM and 403.29 4.10 μM, respectively.
These two peptides were found to block the interactions between bitter substances and the T2R14 receptor by binding to bitter taste receptor due to their higher affinities. Residues Thr86, Asp168, and Phe247 may be the key amino acids responsible for the binding of the bitter taste receptor blockers to T2R14. This study revealed that molecular docking was an efficient approach for screening bitter taste receptor blockers that can be used for reducing bitterness of foods and drugs. The pharmacological role of bitter taste receptor blockers remains to be explored.
Zhipeng Yu: Conceptualization, Methodology, Writing – review & editing, Supervision. : . Yingxue Wang: Data curation, Formal analysis, Writing – original draft. Wenzhu Zhao: Software, Validation, Project administration. Jianrong Li: Validation, Software. David Shuian: Su- pervision. Jingbo Liu: Investigation.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement
This paper was supported by National Key R&D Program of China (2019YFD0901702).

References
Behrens, M., Brockhoff, A., Kuhn, C., Bufe, B., Winnig, M., & Meyerhof, W. (2004). The human taste receptor hTAS2R14 responds to a variety of different bitter compounds.
Biochemical and Biophysical Research Communications, 319(2), 479–485.
Born, S., Levit, A., Niv, M. Y., Meyerhof, W., & Behrens, M. (2013). The human bitter taste receptor TAS2R10 is tailored to accommodate numerous diverse ligands. The
Journal of Neuroscience, 33(1), 201–213.
Brockhoff, A., Behrens, M., Niv, M. Y., & Meyerhof, W. (2010). Structural requirements of bitter taste receptor activation. Proceedings of the National Academy of Sciences of the
United States of America, 107(24), 11110–11115.
Di Pizio, A., & Niv, M. Y. (2015). Promiscuity and selectivity of bitter molecules and their
receptors. Bioorganic & Medicinal Chemistry, 23(14), 4082–4091.
Drewnowski, A., & Gomez-Carneros, C. (2000). Bitter taste, phytonutrients, and the consumer: A review. American Journal of Clinical Nutrition, 72(6), 1424–1435.
FitzGerald, R. J., & O’Cuinn, G. (2006). Enzymatic debittering of food protein hydrolysates. Biotechnology Advances, 24(2), 234–237.
Jaggupilli, A., Howard, R., Upadhyaya, J. D., Bhullar, R. P., & Chelikani, P. (2016). Bitter taste receptors: Novel insights into the biochemistry and pharmacology. The
International Journal of Biochemistry & Cell Biology, 77(Pt B), 184–196.
Karaman, R., Nowak, S., Di Pizio, A., Kitaneh, H., Abu-Jaish, A., Meyerhof, W., …
Behrens, M. (2016). Probing the binding pocket of the broadly tuned human bitter
taste receptor TAS2R14 by chemical modification of cognate agonists. Chemical Biology & Drug Design, 88(1), 66–75.
Keil, B. (1987). Proteolysis Data Bank: Specificity of alpha-chymotrypsin from computation of protein cleavages. Protein Sequences & Data Analysis, 1(1), 13–20.
Kohl, S., Behrens, M., Dunkel, A., Hofmann, T., & Meyerhof, W. (2013). Amino acids and peptides activate at least five members of the human bitter taste receptor family.
Journal of Agricultural and Food Chemistry, 61(1), 53–60.
Lafarga, T., O’Connor, P., & Hayes, M. (2015). In silico methods to identify meat-derived prolyl endopeptidase inhibitors. Food Chemistry, 175, 337–343.
Meyerhof, W., Batram, C., Kuhn, C., Brockhoff, A., Chudoba, E., Bufe, et al. (2010). The molecular receptive ranges of human TAS2R bitter taste receptors. Chemical Senses, 35(2), 157-170.
Mooney, C., Haslam, N. J., Pollastri, G., Shields, D. C., & Kurgan, L. (2012). Towards the improved discovery and design of functional peptides: Common features of diverse classes permit generalized prediction of bioactivity. PLoS One, 7(10), e45012.
Phat, C., Moon, B., & Lee, C. (2016). Evaluation of umami taste in mushroom extracts Unesbulin by chemical analysis, sensory evaluation, and an electronic tongue system. Food
Chemistry, 192, 1068–1077.
Pydi, S. P., Sobotkiewicz, T., Billakanti, R., Bhullar, R. P., Loewen, M. C., & Chelikani, P.
(2014). Amino acid derivatives as bitter taste receptor (T2R) blockers. Journal of Biological Chemistry, 289(36), 25054–25066.
Roland, W. S. U., Vincken, J.-P., Gouka, R. J., van Buren, L., Gruppen, H., & Smit, G. (2011). Soy isoflavones and other isoflavonoids activate the human bitter taste receptors hTAS2R14 and hTAS2R39. Journal of Agricultural and Food Chemistry, 59
(21), 11764–11771.
Seifert, M., & Lang, M. (2008). Essential Factors for Successful Virtual Screening. Mini-
Reviews in Medicinal Chemistry, 8(1), 63–72.
Sohi, H., Sultana, Y., & Khar, R. K. (2004). Taste masking technologies in oral pharmaceuticals: Recent developments and approaches. Drug Development and
Industrial Pharmacy, 30(5), 429–448.
Xu, Q., Singh, N., Hong, H., Yan, X., Yu, W., Jiang, X.u., et al. (2019). Hen protein-
derived peptides as the blockers of human bitter taste receptors T2R4, T2R7 and T2R14. Food Chemistry, 283, 621–627.
Yarmolinsky, D. A., Zuker, C. S., & Ryba, N. J. P. (2009). Common sense about taste:
From mammals to insects. Cell, 139(2), 234–244.
Yu, Z., Kan, R., Ji, H., Wu, S., Zhao, W., Shuian, D., et al. (2020). Identification of tuna protein-derived peptides as potent SARS-CoV-2 inhibitors via molecular docking and molecular dynamic simulation. Food Chemistry, 128366.
Yu, Z., Kang, L., Zhao, W., Wu, S., Ding, L., Zheng, F., et al. (2020). Identification of novel umami peptides from myosin via homology modeling and molecular docking. Food Chemistry, 128728.
Zhang, C., Alashi, A. M., Singh, N., Liu, K., Chelikani, P., & Aluko, R. E. (2018). Beef protein-derived peptides as bitter taste receptor T2R4 blockers. Journal of Agricultural and Food Chemistry, 66(19), 4902–4912.