<b>Identification and 3D modeling of bioactive peptides from Lactobacillus brevis RAMULAB49 protein hydrolysate with in silico ERK1 phosphorylation inhibition activity targeting diabetic nephropathy</b>

Diabetic nephropathy (DN) poses a significant health challenge, necessitating novel therapeutic approaches. In this study, we isolated proteins from cell-free supernatant (CFS) from the culture of the lactic acid bacteria Lactobacillus brevis RAMULAB49 strain. The proteins were subjected to simulated in vitro gastrointestinal digestion using gut enzymes – pepsin, pancreatin, and trypsin. The hydrolysates were filtered using 3kDa threshold ultra-centrifugal filters and were desalted using C18 disks. This was followed by nLC-ESI MS/MS tandem mass spectrometry-based identification of peptides, leading in the identification of a of 258 unique peptides across three enzyme combinations. The resultant sequences were made into peptide library construction based on their, bioactivity scores, allergenicity, toxicity, and antidiabetic potential, a total of 10 peptides was constructed and modeled in 3D. On the other hand, 266 DN associated genes were identified using a network pharmacology approach. The resultant protein-protein (PPI) network was analysed using the gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment approaches, resulting in identification of critical pathways, ERK1, PI3K-Akt, EGRF and TNF signaling as significantly involved in DN, where, ERK1 emerging as a key node due to its involvement in cell proliferation, inflammation, and fibrosis associated with DN. Top two 3D-modelled bioactive peptides were selected for interaction study with the target protein ERK1. Peptide TNEDPYTIDVES showed a strong binding energy of −9.9 kcal/mol, at the ATP-binding site and dynamics simulations confirmed the structural stability of this complex over 100 ns, showing consistent hydrogen bond interactions and RMSD values below 2.5 Å. These findings suggest that TNEDPYTIDVES may act as a competitive ERK1 inhibitor by occupying the adenine-mimicking ATP-binding cleft, thereby interfering with phosphorylation activity. This integrative approach highlights L. brevis RAMULAB49 strain derived peptides as promising candidates for the development of peptide-based therapeutics target and could pave the way for new drug development treating diabetic nephropathy.糖尿病肾病（Diabetic nephropathy, DN）已成为一项亟待解决的重大健康难题，亟需开发新型治疗手段。本研究从短乳杆菌（Lactobacillus brevis）RAMULAB49菌株的培养液无细胞上清液（cell-free supernatant, CFS）中分离得到蛋白，随后采用胃蛋白酶、胰酶与胰蛋白酶三种胃肠道酶对该蛋白进行体外模拟胃肠道消化。消化产物经3kDa截留分子量的超速离心过滤膜过滤，并通过C18固相萃取小柱进行脱盐处理。后续基于纳流液相色谱-电喷雾电离串联质谱（nLC-ESI MS/MS）对肽段进行鉴定，最终在三种酶消化组合中鉴定得到共计258种独特肽段。基于各肽段的生物活性评分、致敏性、毒性及抗糖尿病活性，对所得序列进行肽库构建，最终筛选得到10条肽段并完成三维结构建模。 另一方面，本研究通过网络药理学方法鉴定得到266个与DN相关的基因。对所得蛋白质-蛋白质相互作用（protein-protein interaction, PPI）网络采用基因本体（gene ontology, GO）与京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）通路富集分析进行解析，最终确定ERK1、PI3K-Akt、EGRF及TNF信号通路为参与DN发生发展的关键通路，其中ERK1因参与DN相关的细胞增殖、炎症反应及纤维化过程而成为核心调控节点。 选取2条经三维结构建模的生物活性肽段，与靶蛋白ERK1开展相互作用研究。其中肽段TNEDPYTIDVES在ERK1的ATP结合位点展现出-9.9 kcal/mol的强结合能，分子动力学模拟证实该复合物在100 ns模拟时长内结构稳定，始终保持稳定的氢键相互作用，且均方根偏差（root mean square deviation, RMSD）值低于2.5 Å。上述结果表明，TNEDPYTIDVES可通过占据模拟腺嘌呤的ATP结合裂隙发挥竞争性ERK1抑制剂的作用，进而干扰其磷酸化活性。本整合研究策略证实，短乳杆菌（Lactobacillus brevis, L. brevis）RAMULAB49菌株来源的肽段有望成为肽类治疗药物的优质候选物，为糖尿病肾病的新型药物研发开辟新路径。