Purification, sequence identification, and taste-presenting mechanism analysis of four novel taste-active peptides enzymatically prepared from shiitake mushroom (Lentinula edodes)

Abstract

Shiitake mushroom-derived taste-active peptides were prepared by enzymatic hydrolysis and purified step-by-step using D001 macroporous adsorption resin, Sephadex G-100 gel filtration, and reversed-phase high-performance liquid chromatography. The initial hydrolysate contained 16.12 mg/mL peptides with saltiness and umami scores of 6. The active fraction was screened by electronic tongue and sensory evaluation, and four novel peptides (PDDER, QPEEQDER, SPDDER, and WEPGPQ) were identified by liquid chromatography tandem mass spectrometry-based de novo sequencing, all exhibiting strong binding to the heterodimeric umami receptor T1R1/T1R3 via hydrogen bonds and electrostatic interactions. Verification with synthesized peptides confirmed their favorable taste-active properties, excellent aqueous solubility, non-toxicity and non-allergenicity, demonstrating their potential for further development as umami-enhancing ingredients. Among them, SPDDER performed best, with outstanding electronic tongue response values (umami 9.68, salty 8.54) and sensory scores (umami 9.5, salty 8.5). Thereby providing a methodological and theoretical basis for producing taste-active peptides from macro-fungi.

Graphical abstract

Highlights

• •Obtained the four novel peptides with good taste-presenting effect.
• •Four peptides bound strongly and stably to heterodimeric umami receptor T1R1/T1R3.
• •Verified their favorable taste-active properties of synthetic peptides by e-tongue.
• •Revealed great potential for application in functional seasoning products.

1. Introduction

Shiitake mushroom (Lentinula edodes) is a popular traditional edible fungi in China, renowned for its unique taste and rich nutritional profile (Xu et al., 2023). It has a high protein content and is rich in taste-active amino acids such as glutamic acid and aspartic acid, making it an ideal source for the discovery of natural taste-active peptides (Du et al., 2024). In recent years, driven by growing consumer demand for natural and healthy seasonings, the isolation and identification of peptides with well-defined taste activity from edible fungi has become a major aim of food flavor and additive research (Du et al., 2026; García-Casal et al., 2016; Liang et al., 2021; Zhang et al., 2025).

Umami and saltiness, two fundamental taste attributes, play a key role in harmonizing and enhancing the overall flavor profile of food products (Sun et al., 2022). Umami peptides directly contribute to umami perception, and also enhance the overall flavor complexity and prolong the taste persistence of foods (Zhang et al., 2019). Salty peptides can replace or partially replace salt, thereby facilitating the development of healthier low-sodium foods (Chen et al., 2023; Le et al., 2022). Considerable progress has been made in research on umami peptides from edible fungi, both domestically and internationally. For example, multiple umami peptides were identified in dried L. edodes and their substantial contribution to overall flavor was confirmed (Kong et al., 2019). Six umami peptides with pronounced umami-enhancing properties were isolated from white beech mushroom (Hypsizygus marmoreus) (Li, 2021). However, previous studies have predominantly focused on common edible fungal species, while systematic investigations on taste-active peptides in shiitake mushroom remain limited (Cheng et al., 2026). Research on the isolation, identification, and functional analysis of salty peptides from this versatile source is particularly scarce (Chen et al., 2021; Xiao et al., 2024; Zheng et al., 2021).

Conventional screening and sensory evaluation of taste-active peptides can directly reflect human gustatory perception, but it is susceptible to multiple confounding factors including inter-individual variability among panelists, environmental conditions, physiological states, and psychological fatigue, all of which can compromise the reproducibility and objectivity of the results (Chen et al., 2022; Fu et al., 2024; Lu et al., 2025; Maurya, 2024). To overcome these limitations, electronic tongue (e-tongue) technology has emerged (Schlossareck & Ross, 2019; Tahara & Toko, 2013). In recent years, this technology has been extensively applied to taste evaluation of various food products including alcoholic beverages (Garcia-Hernandez et al., 2025), tea (Xu et al., 2019), shiitake mushroom (Kong et al., 2019) and other edible fungi (Feng et al., 2023; Phat et al., 2016), and aquatic products (Jiang et al., 2018), demonstrating favorable stability and reproducibility. Integrating e-tongue with human sensory evaluation to construct an “instrument-sensory” collaborative evaluation system can compensate for the subjectivity and instability inherent in traditional methods, and also achieve quantitative characterization of taste profiles through instrumental data, making the screening and evaluation of taste-active compounds more accurate and efficient (Calvini & Pigani, 2022).

Based on the crude peptide solution prepared under the previously optimized enzymatic hydrolysis process conditions, this study aims to further systematically screen and identify novel umami and salty peptides by liquid chromatography-tandem mass spectrometry (LC-MS/MS), performed e-tongue sensory evaluation, and verified taste-active properties through peptide synthesis. This work provided a systematic methodology for resource exploration and flavor analysis of taste-active peptides from shiitake mushroom, and a theoretical basis for the development and utilization of natural taste-active peptides derived from edible fungi.

2. Materials and methods

2.1. Materials

Shiitake mushroom powder was prepared from commercially available fresh L. edodes samples. After washing and slicing, samples were air-dried at 60 °C to constant weight, pulverized, and passed through an 80-mesh sieve to obtain a dry powder with uniform color and consistent psize.

2.2. Methods

2.2.1. Extraction of shiitake mushroom protein

Following a published method (Li, Chen, et al., 2022), shiitake mushroom protein was extracted by ultrasound-assisted alkaline extraction and acid precipitation using an SB-5200DTD ultrasonic cleaner (Ningbo Xinzhi Biotechnology Co., Ltd., Ningbo, China). The resulting protein precipitate was dissolved in an appropriate amount of distilled water, allowed to cool to room temperature, and freeze-dried using an LGJ-10E freeze dryer (Beijing Sihuan Furui Keyi Technology Development Co., Ltd., Beijing, China) to obtain protein powder. The average protein content of the protein powder was determined to be 13.0 g/100 g. All samples were stored at −20 °C in sealed, light-protected containers until further analysis.

2.2.2. Enzymatic preparation of taste-active peptides

Based on the results of preliminary protease screening and single-factor experiments, Flavourzyme (food grade, Nanning Pangbo Biological Engineering Co., Ltd., Nanning, China) was selected as the hydrolytic enzyme for preparation of taste-active peptides. Subsequently, response surface methodology (RSM) was employed to optimize the key hydrolysis parameters, yielding optimal conditions of enzyme dosage of 2060 U/g, temperature of 49 °C, pH 7.0, and hydrolysis duration of 94 min. Under these optimized conditions, it was demonstrated through three repeated validation experiments that the yield of peptides from shiitake mushroom reached 80.6%. Meanwhile, the model F-value with p < 0.0001 confirmed the high significance of the quadratic model. The correlation coefficient R² was 0.9988, and the adjusted R² was 0.9975, indicating excellent agreement between the observed and predicted values. The lack-of-fit test gave a p-value of 0.6282, confirming that the model adequately fitted the experimental data. These results collectively demonstrated the reliability and predictive capability of the model. Furthermore, sensory evaluation verified a positive correlation between peptide content and both umami (R² = 0.894，P < 0.001) and saltiness (R² = 0.959，P < 0.001) sensory scores.

2.2.3. Trained panel sensory analysis

A sensory panel consisting of 10 healthy assessors (ages: 22–28 years, mean: 25.2 ± 1.8 years, with a one-to-one male-to-female ratio) was recruited through open recruitment and screening (Sipos et al., 2021). All panelists had no history of smoking, no history of taste or olfactory dysfunction, no allergic reactions to the experimental components, and demonstrated sensitivity to and interest in discriminating basic tastes (Chamoun et al., 2019). Prior to evaluation, panelists underwent 1 week of systematic training to establish an internal intensity scale using gradient concentrations of reference standard solutions (Fry Vennerød et al., 2018) prepared with ultrapure water. These were bitterness (quinine sulfate: 0.035 g/L), sourness (citric acid: 0.60 g/L), saltiness (sodium chloride: 1.20 g/L), umami (monosodium glutamate: 1.50 g/L), and sweetness (sucrose: 8.00 g/L) (Melis & Tomassini Barbarossa, 2017). During training, these concentrations were anchored as medium intensity (∼5 points) on a 10-point scale. The test samples consisted of the enzymatic peptide hydrolysate at a concentration of 1 mg/mL. After standing, 5 mL aliquots were dispensed into cups coded with three-digit random numbers and presented in a random order (Lee et al., 2008). Sensory evaluations were conducted in a quiet, environmentally controlled room (22 ± 2 °C, relative humidity 60–70%, neutral lighting), following the general principles of ISO 8589. Samples were coded with three-digit random numbers and presented to panelists in a randomized complete block design to eliminate order and carry-over effects. Panelists were blinded to sample identities throughout the evaluation, as all sample codes were known only to the laboratory technician who prepared them. During evaluation, each panelist took ∼3 mL of sample into the mouth, swished it for 10 s, expectorated, and rinsed immediately with ultrapure water (Douglas et al., 2018), with a 10-min rest interval between samples. A 10-point structured numerical intensity scale (0 = none, 10 = extremely strong) (Torrico et al., 2023) was used to independently rate sweetness, sourness, saltiness, umami, and bitterness in real time. Each sample was evaluated in triplicate by all 10 panelists and mean scores were used for statistical analysis.

2.2.4. Determination of polypeptide content

Protein concentrations of samples were determined using a Bicinchoninic Acid (BCA) Protein Assay Kit (PC0020; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) (Cortés-Ríos et al., 2020). The working solution was prepared by mixing the BCA reagent and Cu²⁺ reagent at a volume ratio of 50:1. A standard curve was generated using bovine serum albumin (BSA; Beijing Solarbio Science & Technology Co., Ltd.) as the standard. Absorbance was measured at 562 nm using a SpectraMax ABS Plus full-wavelength microplate reader (Molecular Devices, Shanghai, China), and protein concentration was subsequently calculated.

2.2.5. E-tongue detection

E-tongue measurement was performed as described previously (Jiang et al., 2022) with appropriate modifications. The SA402B e-tongue system (INSENT Inc., Atsugi-shi, Kanagawa, Japan), equipped with nine built-in biosensors, enables specific responses to sourness, bitterness, astringency, bitter aftertaste, astringent aftertaste, umami, richness (umami aftertaste), saltiness, and sweetness.

2.2.6. Macroporous resin and gel filtration chromatography purification

Based on the results of preliminary resin screening experiments, D001 macroporous adsorption resin (Beijing Solarbio Science & Technology Co., Ltd.) was first used for preliminary purification. The resulting peptide fractions were further purified using Sephadex G-100 dextran gel (Shanghai Yuanye Bio-Technology Co., Ltd., Shanghai, China). The column was eluted with distilled water at a controlled flow rate of 1.0 mL/min, and a Col-SBS100 automatic fraction collector (Shanghai Qingpu Huxi Instrument Factory, Shanghai, China) was used to collect 3.0 mL per tube, yielding 40 fractions. The elution profile was monitored by UV detection at 214 nm, and two major elution peaks were detected, designated as A1 (tube 13) and A2 (tube 27). Aliquots from each collected fraction were subjected to sensory evaluation. Based on saltiness and umami scores, fractions with the highest scores were pooled, freeze-dried, and stored for subsequent reversed-phase high-performance liquid chromatography (RP-HPLC) analysis using a 2695/2996 instrument (Waters Corporation, Milford, MA, USA).

2.2.7. RP-HPLC purification

Fraction A2, which exhibited the highest sensory scores after gel filtration chromatography fractionation, was further separated by RP-HPLC. Two peaks, designated A2–1 and A2–2, were collected, and their taste-active properties were evaluated by e-tongue. Chromatographic separation was performed on an Amethyst C18—H column (4.6 × 250 mm, 5 μm, 120 Å). Mobile phase A consisted of 2% acetonitrile +98% water +0.05% trifluoroacetic acid (TFA), and mobile phase B consisted of 90% acetonitrile +10% water +0.05% TFA. The elution gradient was as follows: 0–4 min, 10% B; 30 min, 70% B; 35–40 min, 90% B. The detection wavelength was set at 214 nm, the flow rate was 0.8 mL/min, the sample concentration was 2 mg/mL, the injection volume was 10 μL, and the column temperature was maintained at 25 °C.

2.2.8. De novo peptide sequencing

To identify the amino acid sequences of taste-active peptides, peak A2–2 from RP-HPLC was selected for sequencing. LC-MS/MS analysis was performed on an Easy-nLC 1200 system (Thermo Fisher Scientific, Waltham, MA,USA), and analytical work was performed by Beijing Biotech-Pack Scientific Co., Ltd. (Beijing, China). Sample pretreatment included reduction and alkylation modification steps, followed by desalting and purification. The pretreated peptide sample was separated by HPLC then introduced into the MS instrument for detection, with data recorded in the RAW file format. MS data were analyzed using the PEAKS de novo sequencing algorithm, which identifies and ranks peptides in complex samples based on peptide match scores, thereby enabling the accurate determination of each peptide sequence within the mixture. This method utilizes fragment ion spectra generated by tandem mass spectrometry combined with a de novo sequencing algorithm, allowing direct peptide sequence readout without reliance on pre-existing databases. It is particularly well-suited for structural elucidation of novel bioactive peptides.

2.2.9. Prediction of taste activity and physicochemical properties of peptides based on online peptide databases

Sequence novelty was verified using NCBI BLASTp (Zaru et al., 2023) (https://blast.ncbi.nlm.nih.gov, accessed April 8, 2026) to confirm that sequences had not been previously reported. Toxicity and allergenicity of these peptide sequences were subsequently predicted using ToxinPred (https://webs.iiitd.edu.in/raghava/toxinpred/index.html, accessed April 8, 2026) and AllerCatPro 2.0 (https://allercatpro.bii.a-star.edu.sg/, accessed April 8, 2026), respectively (Nguyen et al., 2022; Sharma et al., 2022). Isoelectric point, net charge, hydrophobicity, and aqueous solubility were calculated using PepDraw (https://www.pepdraw.com/, accessed April 8, 2026) and Innovagen (http://www.innovagen.com/proteomics-tools, accessed April 8, 2026) (Li et al., 2023). Finally, umami potential was predicted using Umami-MRNN (Qi et al., 2023) (http://umami-mrnn.herokuapp.com/, accessed April 8, 2026), Umami_YYDS (Cui et al., 2023) (http://www.tastepeptides-meta.com/, accessed April 8, 2026), iUmami-SCM (Charoenkwan et al., 2020) (http://camt.pythonanywhere.com/iUmami-SCM, accessed April 8, 2026), and UmamiPredict (https://cosylab.iiitd.edu.in/umami/, accessed April 8, 2026) (Singh et al., 2025). Alignment searches were performed against the BIOPEP-UWM (https://biochemia.uwm.edu.pl/en/biopep-uwm-2/, accessed April 8, 2026) and Peptide DB (https://www.peptidedb.com/, accessed April 8, 2026) databases (Cui et al., 2022; Iwaniak et al., 2024) to screen for novel peptide sequences.

2.2.10. Molecular docking of taste-active peptides

Amino acid sequences of T1R1/T1R3 were retrieved from the UniProtKB database. Homology modeling was performed using SWISS-MODEL, and the resulting initial model was imported into Discovery Studio 2019. Model optimization was carried out through the Minimization protocol, and the quality of the refined structure was assessed by Ramachandran plot analysis. Small-molecule structures were obtained from the PubChem website and subjected to energy minimization using Discovery Studio 2019, then saved as pdb files. Autodock 4.0 was used to prepare the pdbqt files for proteins and small molecules. Docking center coordinates were set to x = 49.247, y = 42.943, z = 7.769, and docking was performed using Autodock Vina 1.2.6 to calculate binding energies. Three-dimensional interaction diagrams were generated with PyMOL 3.1, and two-dimensional diagrams were constructed using Discovery Studio 2019.

2.2.11. Synthesis and activity verification of taste-active peptides

Umami peptides identified and investigated in this study were prepared by solid-phase synthesis conducted by Wuhan Dangang Biotechnology Co., Ltd. (Wuhan, China). The purity of peptides was ≥95%. Peptide solutions were prepared at a concentration of 1 mg/mL, and their taste-active properties were evaluated using a combined approach of sensory evaluation and e-tongue analysis.

2.2.12. Data processing

All experiments were conducted in triplicate as independent replicate trials. Data were analyzed using IBM SPSS Statistics 27 software, with a p-value <0.05 considered statistically significant. RSM analysis was performed using Design-Expert 13 and figures were plotted using Microsoft Excel and Origin 2021.

3. Results and analysis

3.1. Separation and purification of taste-active peptides

3.1.1. Gel filtration chromatography purification

The shiitake mushroom taste-active peptide fraction obtained after preliminary purification by macroporous resin was further purified using Sephadex G-100 dextran gel. As shown in Fig. 1a, based on the absorbance at 214 nm, an elution profile was constructed for the first 40 collected fractions, revealing two major characteristic absorption peaks, designated A1 (corresponding to tube 13) and A2 (corresponding to tube 27). The sensory evaluation results (Fig. 1b) showed that the saltiness and umami scores of the A1 fraction were 5 and 4, respectively, whereas those of the A2 fraction both exceeded 6, significantly outperforming the A1 fraction. Based on the combined elution profile and sensory evaluation results, the A2 fraction was determined to contain more desirable taste-active properties. Accordingly, the eluate corresponding to A2 (tube 27) was collected, freeze-dried, and stored for subsequent further purification by RP-HPLC.

Fig. 1.

(a) Purification of mushroom peptide by gel chromatography. (b) Sensory evaluation of salty and umami taste of the two components after gel purification. Note: Error bars represent the standard deviation of triplicate independent trials (n = 3). Different uppercase letters indicate significant differences in saltiness among groups (P < 0.05), and different lowercase letters indicate significant differences in umami among groups (P < 0.05).

3.1.2. RP-HPLC separation of taste-active peptides

To further purify the active fraction obtained by gel filtration chromatography, fraction A2, which exhibited the highest sensory scores for both saltiness and umami, was subjected to RP-HPLC. The chromatographic analysis results (Fig. 2a) showed that fraction A2 was effectively separated into two characteristic sub-fractions, designated A2–1 and A2–2. Each sub-fraction was collected, concentrated, and freeze-dried, and their taste-active properties were evaluated using e-tongue. The results indicated differences in taste-active properties among the sub-fractions (Fig. 2b), a phenomenon that may be closely associated with their molecular polarity and structural characteristics.

Fig. 2.

(a) Separation of mushroom peptides by RP-HPLC. (b) E-tongue analysis of mushroom peptides after RP-HPLC separation. Note: Data represent mean values (sensory: n = 10 panelists, triplicate measurements; e-tongue: n = 3 independent measurements). The same applies to subsequent radar charts.

3.2. Sensory evaluation and E-tongue verification

The crude enzymatic hydrolysate of shiitake mushroom protein, after preliminary fractionation by D001 macroporous adsorption resin and subsequent fine purification by Sephadex G-100 dextran gel, exhibited significantly improved taste-active properties. As shown in Fig. 3a, the e-tongue results indicated that purification effectively suppressed bitterness signals while fully preserving the two key taste attributes of umami and saltiness. As illustrated in Fig. 3b, the quantitative sensory scores from the panel further corroborated this trend; compared with the unpurified sample, the umami and saltiness scores of the purified peptide mixture were increased from 6 to 8 and from 6 to 7, respectively, while the slight sourness decreased from 1 to 0, bitterness was suppressed, and sweetness exhibited a modest increase. The combined e-tongue analysis and sensory evaluation demonstrated that, after serial purification by D001 resin and Sephadex G-100 gel, the sample maintained umami and salty taste profiles, and the perceived intensity of these two tastes was elevated to varying degrees, while undesirable taste responses, particularly bitterness, were effectively decreased.

Fig. 3.

(a) E-tongue radar map before and after purification. (b) Sensory evaluation radar map before and after purification.

The improvement in the flavor profile described above could be attributed to the targeted enrichment of taste-active components and the effective removal of non-target impurities during the purification process. The D001 cation exchange resin selectively adsorbs positively charged non-target components based on charge differences, whereas the Sephadex G-100 gel filtration chromatography achieves fractionation of different molecular weight ranges via molecular sieving. Through this combined purification strategy, residual pigments, polysaccharides, macromolecular proteins, and other tasteless or off-flavor precursor substances in the enzymatic hydrolysate were removed to varying degrees, thereby reducing their interference with taste receptors. Meanwhile, low-molecular-weight peptides with umami- and saltiness-enhancing activities were enriched and concentrated in the target fractions, leading to an enhanced taste contribution of the active peptides per unit concentration. These results confirmed that combined serial purification by D001 resin and Sephadex G-100 gel selectively enriched umami and salty low-molecular-weight peptides derived from shiitake mushroom, while removing bitterness-related components, thereby providing a high-quality sample basis for subsequent identification of active peptide sequences and structure-activity relationship studies.

3.3. Peptide sequence identification

The A2–2 fraction obtained from RP-HPLC purification was subjected to sequence analysis by LC-MS/MS, from which 123 peptides with a de novo sequencing score ≥ 95 were identified. Following the screening workflow described in Section 2.2.9, seven candidate peptides that met the preset criteria, that had not been previously reported in the literature, and that exhibited the most favorable predicted umami activity were selected for molecular docking analysis. Based on a comprehensive evaluation of binding energy, hydrogen bonding interactions, hydrophobicity, non-toxicity, and non-allergenicity, four novel peptides with potential umami activity (PDDER、QPEEQDER、SPDDER and WEPGPQ) were ultimately screened. The limited size of the active pocket of the heterodimeric T1R1/T1R3 umami receptor meant that macromolecular proteins or large polypeptides had difficulty entering the binding cavity, whereas taste-active peptide sequences, typically composed of 2–8 amino acid residues, may have accessed the pocket (Zhang et al., 2024). Accordingly, the length and structural characteristics of these four peptides aligned with the typical properties of umami peptides, it suggested that they had potential umami taste capacity. The MS/MS fragmentation spectra of the four peptides are presented in Fig. 4. Database searches using NCBI BLASTp and other resources confirmed that these four peptides had not been previously reported. They are all rich in acidic acid residues such as Asp (D) and Glu (E), which provide an important structural basis for the taste activity of umami peptides (Li, Wang, et al., 2022). However, compared with known umami peptides from various fungal sources, including Boletus edulis and Agaricus bisporus as reported in the literature (Wu et al., 2025), the sequences of our four peptides are significantly different from theirs.

Fig. 4.

MS/MS fragmentation spectra of four novel taste-active peptides from shiitake mushroom.

3.4. Homology modeling of heterodimeric umami receptor T1R1/T1R3

Homology modeling is a key approach for elucidating the binding mechanism between umami peptides and the T1R1/T1R3 receptor. As shown in Fig. 5a, the heterodimeric extracellular Venus flytrap domain (VFT) model constructed in this study exhibited a typical bilobed architecture. The binding cavity of the T1R1 subunit on the left was relatively small, whereas that of the T1R3 subunit on the right was larger. The two lobes were connected by a flexible hinge, and the cavity jointly formed between them constituted the primary region for the specific binding of low-molecular-weight umami peptides. A Ramachandran plot can confirm the reliability of homology model structures. As shown in Fig. 5b, 92.0% of the residues were located in the most favored regions, with 7.9% in additionally allowed regions, and 0.1% in generously allowed regions, with no residue in disallowed regions, meaning 99.9% of the residues were within the reasonable range. The proportion of residues in the most favored regions far exceeded the 90% critical evaluation criterion, indicating a stereochemically reasonable conformation and a reliable structure that could be used for subsequent molecular docking.

Fig. 5.

(a) The 3D structure of the T1R1/T1R3 protein. (b) Ramachandran plot.

3.5. Molecular docking

Molecular docking, widely applied to predict ligand-receptor binding modes, was used to analyze the interactions of the four peptides (PDDER、QPEEQDER、SPDDER and WEPGPQ) with the T1R1/T1R3 umami receptor. The binding energies of the four peptides with the receptor were − 9.1, −9.6, −9.1, and − 11.2 kcal/mol, respectively, indicating favorable binding affinity for all, with the lowest binding energy for WEPGPQ, indicating the strongest affinity. As shown in Fig. 6 and Table 1, all four peptides bound to the VFT domain of T1R1/T1R3, which served as the core binding site for ligand recognition by the umami receptor. Specifically, PDDER formed hydrogen bonds with Glu45, Thr305, Val277, Gln389, Ser147, Ser146, Arg247, Ser306, and His278; QPEEQDER formed hydrogen bonds with Phe65, Asp307, Ser306, Leu308, Asn68, Thr305, Glu301, Gln389, Glu105, Ser104, Ser147, and Gln181; SPDDER formed hydrogen bonds with Ser306, Glu148, Ser147, Ser146, Ala46, Ser104, and Cys103; and WEPGPQ formed hydrogen bonds with Glu105, Arg247, Ser146, Gly168, Ser170, Ser147, Ala302, and His145. Hydrogen bonds were the predominant noncovalent forces mediating peptide binding to T1R1/T1R3, and the number of hydrogen bonds was generally positively correlated with binding strength. The hydrogen bond lengths observed in the four peptide-receptor complexes ranged from 1.8 to 2.9 Å. Because shorter hydrogen bonds corresponded to lower potential energy, this range of short bonds suggested high binding stability for the protein-ligand complexes.

Fig. 6.

(A-H) Molecular docking of the four novel peptides with the T1R1/T1R3 umami taste receptor.

Table 1.

Molecular docking of four novel peptides with the T1R1/T1R3 umami receptor.

Peptide Sequence	Conventional Hydrogen Bond	Van der Waals Interaction	Carbon Hydrogen Bond	Other Interactions	Binding Energy (kcal/mol)
PDDER	Glu45,Thr305,Val277, Gln389,Ser147,Ser146, Arg247,Ser306,His278	Gly69,His387,Asn68,Ser66, Pro42,Gly44,Glu301,Try218, Gly168,Ala169,Ser170,Glu148, Asp216,Leu245,Glu48,Arg180, Leu51,Ser104,Pro106	His145, Ala302, Ser276	Trp72,Glu105	−9.1
QPEEQDER	Phe65,Asp307,Ser306, Leu308,Asn68,Thr305, Glu301,Gln389,Glu105, Ser104,Ser147,Gln181	Ser67,Trp303,Gly69,His387, Pro42,Ser66,Trp72,Arg64, Glu45,Val277,Leu51,Val107, Tyr218,Arg180,Pro106,Glu48, Leu149,Ser146,Ser170,Leu149	His388, Ala302, Asp216, Ser276	His145,Glu148, His278,Arg247, Leu245,Pro246	−9.6
SPDDER	Ser306,Glu148,Ser147, Ser146,Ala46,Ser104, Cys103	Asp307,His388,Trp72,Asn68, Leu308,Trp303,Thr305,Val277, His145,Tyr218,Leu149,Glu45, Pro246,Arg180	Ala302, Pro106, Ser276	Gln389,Glu301, Asp216,Glu105, Glu48,His278, His387,Leu245, Arg247	−9.1
WEPGPQ	Glu105,Arg247,Ser146, Gly168,Ser170,Ser147, Ala302,His145	Leu245,Pro246,Glu148,Leu149, Ser276,Leu51,Gln181,Ser104, Ser66,Gly69,Asn68,Trp72, His387,Ala169,Gln389,Ser306	Glu48	Arg180,Tyr218, Pro106,His278, Pro42,Val277	−11.2

However, although WEPGPQ did not form the highest number of hydrogen bonds, it exhibited the lowest binding energy. Further molecular docking analysis revealed that a greater variety of non‑hydrogen bonding interactions occurred between WEPGPQ and the receptor, including electrostatic interactions with residues Arg180, Tyr218, Pro106, His278, Pro42, and Val277, as well as π-cation interactions and other auxiliary stabilizing forces. The combined strength of these forces, particularly charge–charge and π–cation interactions, compensated for the lower hydrogen-bond count, resulting in the lowest binding energy among the four peptides. This observation was consistent with previous reports that hydrogen bonds enhance binding stability while van der Waals forces optimize binding conformation to reduce binding energy (Gao et al., 2025). These synergistic noncovalent interactions collectively enhanced the binding affinity of WEPGPQ for the receptor, suggesting that it may have possessed the strongest umami activation potential among the four peptides. QPEEQDER demonstrated the next strongest affinity, whereas PDDER and SPDDER exhibited comparable affinities, and both may also have possessed notable umami activation capacity.

3.6. Taste-active properties of synthesized peptides

The physicochemical properties and safety profiles of the four novel peptides were predicted, and the results are presented in Table S1. All peptides were predicted to be non-toxic and non-allergenic, confirming their safety basis for application as food-derived taste-active peptides. In terms of physicochemical properties, PDDER, QPEEQDER, and SPDDER were strongly acidic (pI ≤ 4.0) and carried a high negative net charge (−3 to −4) at pH 7.0, with charge densities of −0.60, −0.50, and − 0.50, respectively. This pronounced electronegative character was conducive to electrostatic attraction with positively charged residues (e.g., arginine and lysine) that were abundant within the binding pocket of the T1R1/T1R3 umami receptor, providing a structural basis that explained their umami taste activity. In contrast, WEPGPQ had a lower net charge (−1) but relatively higher hydrophobicity with a grand average of hydropathicity (GRAVY) value of −1.44, which may have enhanced its receptor-binding capacity through hydrophobic interactions, consistent with its lowest binding energy (−11.2 kcal/mol) among the four peptides. Moreover, all peptides exhibited good to excellent aqueous solubility (high or relatively high), demonstrating their potential for further development as umami-enhancing ingredients.

The e-tongue analysis results for the four synthesized peptides are shown in Fig. 7a. All four peptides elicited distinctly positive responses on the umami sensor, with response values of 9.46, 8.87, 9.68, and 8.44 for PDDER, QPEEQDER, SPDDER (the highest umami response), and WEPGPQ, respectively. PDDER and SPDDER also displayed prominent responses on the saltiness sensor, with values of 8.67 and 8.54, respectively, whereas QPEEQDER and WEPGPQ showed relatively lower saltiness responses (5.75 and 3.33). Moreover, the responses of all four peptides on the sourness, bitterness, astringency, bitter aftertaste, and astringent aftertaste sensors were either negative or close to zero, indicating the absence of undesirable taste attributes.

Fig. 7.

(a) Evaluation of four synthetic peptides e-tongue. (b) Evaluationof four synthetic peptides sensory.

The sensory evaluation results presented in Fig. 7b further corroborated the e-tongue findings. Peptide SPDDER achieved an umami score of 9.5 and a saltiness score of 8.5, showing strong consistency with the e-tongue data. This indicated that synthesized peptide SPDDER possessed potent umami activity, but also exerted a favorable saltiness-releasing effect. In combination with the molecular docking results (binding energy of SPDDER = −9.1 kcal/mol), it was inferred that its strong umami activity may have attributed to the abundance of aspartic acid (D) and glutamic acid (E) residues in its sequence. These negatively charged acidic residues were capable of forming strong electrostatic attractions with positively charged residues in the T1R1/T1R3 receptor, while also participating in the formation of hydrogen bond networks. The saltiness activity, in turn, may have been attributable to the arginine (R) residue in the sequence, whose guanidinium group could interact with salt taste receptors such as the epithelium sodium channel (ENaC) (Li et al., 2025). The fact that SPDDER simultaneously exhibited high umami and saltiness scores suggested its potential as a novel “umami-salty” taste-active peptide, providing potential value in food applications requiring salt reduction and umami enhancement.

4. Discussion

Bioactive peptides have attracted widespread attention in the scientific community for their application in functional foods and natural seasonings, owing to their low molecular weight, high safety profile, and ease of absorption and digestion by the human body (Kong et al., 2019; Li, 2021). Traditional chemically synthesized additives may raise potential health concerns and provoke consumer resistance, which is why using food-derived taste-active peptides as substitutes for monosodium glutamate or as auxiliary components for salt reduction and umami enhancement, is a major focus in the flavor improvement field (Zhang et al., 2019).

Through the widespread adoption of industrialized cultivation technology and the continuous expansion of production capacity, the edible fungi industry is confronting the dual challenges of insufficient deep processing and low added value. Shiitake mushroom, an edible fungus with a crude protein content far exceeding that of common vegetables, is rich in taste-active amino acids such as glutamic acid and aspartic acid, making it an ideal raw material for preparation of natural umami peptides (Xu et al., 2023). Compared with chemically synthesized taste-active peptides agents, shiitake mushroom-derived taste-active peptides provide the advantages of high safety and no potential neurotoxicity risk. Furthermore, compared with other common food-borne sources of umami peptides (e.g., soybean meal and milk proteins), shiitake mushroom protein exhibits extremely low allergenicity (Du et al., 2024), rendering it more suitable for the development of clean-label products targeting a broad consumer base.

Regarding the preparation process, this study has previously optimized the enzymatic hydrolysis parameters of Flavourzyme using RSM, which effectively released active peptide fragments from shiitake mushroom protein. Under optimal conditions, the preparation yield of taste-active peptides reached 80.6%, demonstrating the feasibility of industrial-scale production. Although dual-enzyme or multi-enzyme hydrolysis generally yields a greater diversity of small peptide fragments, in practical production, a single-enzyme hydrolysis system using Flavourzyme is likely to minimize cost and maximize process reproducibility. Notably, the reaction conditions of proteases must be strictly controlled during the preparation process. An excessively high degree of hydrolysis may lead to further degradation of core peptide fragments with established taste activity into free amino acids or dipeptides, consequently reducing taste-presenting intensity or altering taste-active properties. Therefore, precise regulation of hydrolysis duration, temperature, and pH through RSM is the key to achieving enrichment of active peptide fragments rather than excessive hydrolysis.

Molecular docking can predict the binding mode between a ligand and its receptor. The active pocket of the T1R1/T1R3 umami receptor has limited space and is generally unable to accommodate high-molecular-weight proteins or polypeptides, which is why taste-active peptide sequences are typically short (Zhang et al., 2024). Studies have demonstrated that potential umami peptides are mostly composed of 2–8 amino acids (Zhang et al., 2024), and the abundance of acidic amino acid residues (Asp and Glu) in their sequences serves as an important structural basis for umami activity (Li et al., 2022b). The four novel peptides (PDDER、QPEEQDER、SPDDER and WEPGPQ) identified in this study were highly consistent with the above characteristics in terms of sequence composition. All four were short-chain oligopeptides consisting of 5–8 amino acids, with molecular weights <1 kDa, allowing them to readily enter the VFT binding domain of the T1R1/T1R3 receptor. The molecular docking visualization results confirmed that these four novel peptides possessed the potential to function as potent ligands for the umami receptor. Toxicity and allergenicity prediction analyses performed using ToxinPred and AllerCatPro 2.0 indicated that all four peptide sequences were non-toxic and non-allergenic (Nguyen et al., 2022; Sharma et al., 2022), meeting the basic requirements for subsequent development as umami-enhancing ingredients. Furthermore, sensory evaluation and e-tongue analysis of synthesized peptides demonstrated that all four possessed distinct umami activity and a degree of salty character. Among them, peptide SPDDER performed best in terms of umami and salty taste, with e-tongue response values of 9.68 and 8.54, and sensory scores of 9.5 and 8.5, respectively.

In summary, the sequence length and structural characteristics of these four peptides conformed to the structure-activity relationship rules of umami peptides reported in previous studies. All four peptides were confirmed to possess significant umami properties based on in silico prediction using peptide databases and subsequent sensory and e-tongue validation. This study provided new molecular targets and a theoretical basis for the high-value utilization of shiitake mushroom, and the development of natural compound seasonings.

Of course, it must be acknowledged that this study still has some limitations: firstly, the safety assessment (toxicity and allergenicity) conclusions are on in silico tool predictions, and experimental validation is still required before food application. Secondly, the electronic tongue evaluation and validation of the synthetic peptides need to be conducted positive controls, such as a comparison with sodium glutamate and sodium chloride. Furthermore, due to the general lack of research on salty peptides and the limited understanding of salty receptors, salty mechanisms of the four obtained peptides in this study have not been explored in depth.

5. Conclusion

Through three-step separation and purification coupled with MS identification, four novel taste-active peptides (PDDER、QPEEQDER、SPDDER and WEPGPQ) were obtained from shiitake mushroom protein hydrolysate. Molecular docking results demonstrated that all four peptides formed hydrogen bonding interactions with amino acid residues within the VFT domain of umami receptor T1R1/T1R3, with favorable receptor binding affinity. Sensory evaluation and e-tongue analysis of the four synthesized peptides confirmed that they possessed pronounced umami activity, and also contributed to salty taste. Notably, peptide SPDDER displayed outstanding umami and salty taste performance, with e-tongue response values of 9.68 and 8.54, and sensory scores of 9.5 and 8.5, respectively. Thereby, the shiitake mushroom-derived peptides purified and prepared in this study had potential for development as novel salt-reducing and umami-enhancing seasonings.

CRediT authorship contribution statement

Jiaqi Li: Writing – original draft, Validation, Methodology, Data curation. Xia Ma: Software. Xiaoyu Ye: Resources, Formal analysis. Wenwei Zhang: Data curation. Yuling Qu: Supervision. Fengyun Zhao: Visualization. Jianmin Yun: Writing – review & editing, Project administration, Funding acquisition, Conceptualization.

Ethical statement

The sensory evaluation were not required by national laws to obtain the ethical approval. All participants provided informed consent before participating in the study. The anonymity and confidentiality of the participants were guaranteed, and participation was completely voluntary. Ethical permission about our study is not a requirement of my institution.

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.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.