Wu Yashuai, Yin Ruiyang, Guo Liyun, Song Yumei, He Xiuli, Huang Mingtao, Ren Yi, Zhong Xian, Zhao Dongrui, Li Jinchen, Liu Mengyao, Sun Jinyuan, Huang Mingquan, Sun Baoguo
School of Food Science and Engineering, South China University of Technology, Guangzhou 510640, China.
China Food Flavor and Nutrition Health Innovation Center, Beijing Technology and Business University, Beijing 100048, China.
Foods. 2025 Aug 6;14(15):2743. doi: 10.3390/foods14152743.
This study was designed to systematically identify novel umami peptides in lager beer, clarify their molecular interactions with the T1R1/T1R3 receptor, and determine their specific effects on multidimensional sensory attributes. The peptides were characterized by LC-MS/MS combined with de novo sequencing, and 906 valid sequences were obtained. Machine-learning models (UMPred-FRL, Tastepeptides-Meta, and Umami-MRNN) predicted 76 potential umami peptides. These candidates were docked to T1R1/T1R3 with the CDOCKER protocol, producing 57 successful complexes. Six representative peptides-KSTEL, DELIK, DIGISSK, IEKYSGA, DEVR, and PVPL-were selected for 100 ns molecular-dynamics simulations and MM/GBSA binding-energy calculations. All six peptides stably occupied the narrow cleft at the T1R1/T1R3 interface. Their binding free energies ranked as DEVR (-44.09 ± 5.47 kcal mol) < KSTEL (-43.21 ± 3.45) < IEKYSGA (-39.60 ± 4.37) ≈ PVPL (-39.53 ± 2.52) < DELIK (-36.14 ± 3.11) < DIGISSK (-26.45 ± 4.52). Corresponding taste thresholds were 0.121, 0.217, 0.326, 0.406, 0.589, and 0.696 mmol L (DEVR < KSTEL < IEKYSGA < DELIK < PVPL < DIGISSK). TDA-based sensory validation with single-factor additions showed that KSTEL, DELIK, DEVR, and PVPL increased umami scores by ≈21%, ≈22%, ≈17%, and ≈11%, respectively, while DIGISSK and IEKYSGA produced marginal changes (≤2%). The short-chain peptides thus bound with high affinity to T1R1/T1R3 and improved core taste and mouthfeel but tended to amplify certain off-flavors, and the long-chain peptides caused detrimental impacts. Future formulation optimization should balance flavor enhancement and off-flavor suppression, providing a theoretical basis for targeted brewing of umami-oriented lager beer.
本研究旨在系统鉴定拉格啤酒中的新型鲜味肽，阐明它们与T1R1/T1R3受体的分子相互作用，并确定它们对多维感官属性的具体影响。通过液相色谱-串联质谱（LC-MS/MS）结合从头测序对这些肽进行表征，获得了906条有效序列。机器学习模型（UMPred-FRL、Tastepeptides-Meta和Umami-MRNN）预测了76种潜在的鲜味肽。使用CDOCKER协议将这些候选肽与T1R1/T1R3对接，生成了57个成功的复合物。选择了六种代表性肽——KSTEL、DELIK、DIGISSK、IEKYSGA、DEVR和PVPL进行100纳秒的分子动力学模拟和MM/GBSA结合能计算。所有六种肽都稳定地占据了T1R1/T1R3界面处的狭窄裂隙。它们的结合自由能排序为：DEVR（-44.09±5.47千卡/摩尔）<KSTEL（-43.21±3.45）<IEKYSGA（-39.60±4.37）≈PVPL（-39.53±2.52）<DELIK（-36.14±3.11）<DIGISSK（-26.45±4.52）。相应的味觉阈值分别为0.121、0.217、0.326、0.406、0.589和0.696毫摩尔/升（DEVR<KSTEL<IEKYSGA<DELIK<PVPL<DIGISSK）。基于单因素添加的基于TDA的感官验证表明，KSTEL、DELIK、DEVR和PVPL分别使鲜味得分提高了约21%、约22%、约17%和约11%，而DIGISSK和IEKYSGA产生的变化很小（≤2%）。因此，短链肽与T1R1/T1R3具有高亲和力结合，改善了核心风味和口感，但倾向于放大某些异味，而长链肽则产生有害影响。未来的配方优化应平衡风味增强和异味抑制，为定向酿造鲜味导向的拉格啤酒提供理论依据。
