Dipeptidyl peptidase 4 (DPP4) inhibitors can effectively inhibit the activity of DPP4, increasing the concentrations of glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), which allows for them to effectively contribute to the reduction of blood sugar levels. Leu-Pro-Ala-Val-Thr-Ile-Arg (LPAVTIR) and Leu-Pro-Pro-Glu-His-Asp-Trp-Arg (LPPEHDWR) were the two peptides with the strongest inhibitory activity against DPP4 selected from silkworm pupa proteins. In this study, four systems were established: Apo (ligand-free DPP4), IPI (IPI-bound DPP4), LPAVTIR (LPAVTIR-bound DPP4), LPPEHDWR (LPPEHDWR-bound DPP4), and Gaussian accelerated molecular dynamic (GaMD) simulation was conducted to investigate the mechanism of action of two inhibitory peptides binding to DPP4. Our study revealed that the LPAVTIR peptide possessed a more stable structure and exhibited a tighter binding to the Ser630 active site in DPP4, thus exhibiting a favorable competitive inhibition effect. In contrast, the LPPEHDWR peptide caused the horizontal α-helix (residues 201–215) composed of Glu205 and Glu206 residues in DPP4 to disappear. The spatial arrangement of active sites Ser630 relative to Glu205 and Glu206 was disrupted, resulting in enzyme inactivation. Moreover, the size of the substrate channel and cavity volume was significantly reduced after the binding of the inhibitory peptide to the protein, which was an important factor in the inhibition of the enzyme activity. A similar effect was also found from IPI (our positive control). By stabilizing the active site of DPP4, the IPI peptide induced the disappearance of the horizontal α-helix and a notable reduction in the active cavity volume. In conclusion, our study provided a solid theoretical foundation for the inhibitory mechanisms of IPI, LPAVTIR, and LPPEHDWR on DPP4, offering valuable insights for advancing the development of drug targets for type 2 diabetes.

The vaccinia virus (VV) is extensively utilized as a vaccine vector in the treatment of various infectious diseases, cardiovascular diseases, immunodeficiencies, and cancers. The vaccinia virus Tiantan strain (VVTT) has been instrumental as an irreplaceable vaccine strain in the eradication of smallpox in China; however, it still presents significant adverse toxic effects. After the WHO recommended that routine smallpox vaccination be discontinued, the Chinese government stopped the national smallpox vaccination program in 1981. The outbreak of monkeypox in 2022 has focused people’s attention on the Orthopoxvirus. However, there are limited reports on the safety and toxic side effects of VVTT. In this study, we employed a combination of transcriptomic analysis and machine learning-based feature selection to identify key genes implicated in the VVTT infection process. We utilized four machine learning algorithms, including random forest (RF), minimum redundancy maximum relevance (MRMR), eXtreme Gradient Boosting (XGB), and least absolute shrinkage and selection operator cross-validation (LASSOCV), for feature selection. Among these, XGB was found to be the most effective and was used for further screening, resulting in an optimal model with an ROC curve of 0.98. Our analysis revealed the involvement of pathways such as spinocerebellar ataxia and the p53 signaling pathway. Additionally, we identified three critical targets during VVTT infection—ARC, JUNB, and EGR2—and further validated these targets using qPCR. Our research elucidates the mechanism by which VVTT infects cells, enhancing our understanding of the smallpox vaccine. This knowledge not only facilitates the development of new and more effective vaccines but also contributes to a deeper comprehension of viral pathogenesis. By advancing our understanding of the molecular mechanisms underlying VVTT infection, this study lays the foundation for the further development of VVTT. Such insights are crucial for strengthening global health security and ensuring a resilient response to future pandemics.

Oxidative stress, caused by excessive free radicals, leads to cellular damage and various diseases. Antioxidant peptides from natural proteins offer potential in alleviating this stress. In this study, antioxidant peptides were identified from deer antler proteins using in silico enzymatic hydrolysis and machine learning. Peptides with high prediction scores and non-toxic profiles were selected for evaluation. The antioxidant activities of top candidates, PHPAPTL and VPHGL, were confirmed by radical scavenging assays and their protective effects in HepG2 cells. Molecular dynamics simulations revealed stable binding of these peptides to Keap1, enhancing system stability and reducing residue fluctuations at the ligand-binding interface. Key interactions involved Arg415, Arg483, Arg380, and Ser555. Secondary structure analysis showed peptide binding induced local conformational transitions, notably increasing parallel β-sheet formation near active sites. These findings provide mechanistic insight into their antioxidant effects and support their potential application in functional food development.

Chrysanthemum (Chrysanthemum morifolium Ramat.) has been recognized as both a food and medicinal substance in China since 2002 and possesses antioxidant, anti-inflammatory, antibacterial, and immunomodulatory activities. Previous studies suggest that Chrysanthemum may alleviate skin lesions resembling atopic dermatitis (AD); however, its underlying mechanisms remain unclear. In this study, we integrated network pharmacology and machine learning to systematically explore the potential mechanisms of Chrysanthemum in AD treatment. Four algorithms—Random Forest (RF), Lasso regression with cross-validation (LassoCV), Elastic Net (EN), and Extreme Gradient Boosting (XGB)—were compared, among which the XGB model achieved the best performance (accuracy = 0.9393). Further analysis identified 15 optimal features, two core targets (PTGS2 and MMP9), and one critical pathway (NF-κB signaling). To experimentally validate these findings, HaCaT keratinocytes were co-stimulated with TNF-α and IFN-γ to establish an in vitro inflammatory model, and co-treatment with three major flavonoids from Chrysanthemum—Acacetin, Diosmetin, and Chryseriol—significantly suppressed cytokine-induced COX-2 overexpression and reduced NF-κB p65 phosphorylation, confirming their inhibitory effects on NF-κB activation. These results were consistent with molecular docking and dynamics simulations, which demonstrated that these flavonoids, along with celecoxib, could stably bind to COX-2, thereby enhancing system stability and reducing residue fluctuations at the binding interface, revealing the molecular basis by which Chrysanthemum alleviates AD and supporting its modernization and therapeutic potential.

Neurotoxicity from drugs and environmental pollutants poses serious risks to brain function, yet existing computational models mainly target general neurotoxicity and lack specialized tools for brain-specific assessment. This study aimed to develop and validate a high-performance, brain-focused neurotoxicity prediction framework to improve drug safety evaluation and toxicity screening. We systematically analyzed molecular features, clustering patterns, and target predictions of brain-toxic compounds. Multiple feature representations were compared, including traditional molecular fingerprints, knowledge-guided pre-trained graph Transformer (KPGT) embeddings, and transformer-based MolFormer embeddings, combined with machine learning classifiers. Model performance was evaluated using multiple metrics, and SHAP analysis was conducted to identify influential molecular substructures. Toxic molecules showed physicochemical properties favoring central nervous system (CNS) penetration, including lower molecular weight, lower LogP, fewer hydrogen bond donors/acceptors, fewer rotatable bonds, and lower polar surface area (PSA). The KPGT-MLP model achieved the best balanced performance, with an accuracy (ACC) of 0.8928 and an ROC-AUC of 0.9459, clearly outperforming traditional fingerprint-based models, MolFormer-based models, and general prediction tools such as DI-NeuroT and ADMETlab 3.0. Overall, this study establishes a robust framework for brain-specific neurotoxicity prediction, with the KPGT-MLP model demonstrating strong accuracy and robustness. The proposed approach provides an effective strategy for early neurotoxicity screening and risk assessment, offering valuable insights for safer drug design and advancing computational toxicology and drug discovery.

RNA polymerases are essential enzymes that catalyze DNA transcription into RNA, vital for protein synthesis, gene expression regulation, and cellular responses. Non-template-dependent RNA polymerases, which synthesize RNA without a template, are valuable in biological research due to their flexibility in producing RNA without predefined sequences. However, their substrate polymerization mechanisms are not well understood. This study examines Poly (A) polymerase (PAP), a nucleotide transferase superfamily member, to explore its substrate selectivity using computational methods. Previous research shows PAP’s polymerization efficiency for nucleoside triphosphates (NTPs) ranks ATP > GTP > CTP > UTP, though the reasons remain unclear. Using 500 ns Gaussian accelerated molecular dynamics simulations, stability analysis, secondary structure analysis, MM-PBSA calculations, and Markov state modeling, we investigate PAP’s differential polymerization efficiencies. Results show that ATP binding enhances PAP’s structural flexibility and increases solvent-accessible surface area, likely strengthening protein–substrate or protein–solvent interactions and affinity. In contrast, polymerization of other NTPs leads to a more open conformation of PAP’s two domains, facilitating substrate dissociation from the active site. Additionally, ATP binding induces a conformational shift in residues 225–230 of the active site from a loop to an α-helix, enhancing regional rigidity and protein stability. Both ATP and GTP form additional π–π stacking interactions with PAP, further stabilizing the protein structure. This theoretical study of PAP polymerase’s substrate selectivity mechanisms aims to clarify the molecular basis of substrate recognition and selectivity in its catalytic reactions. These findings offer valuable insights for the targeted engineering and optimization of polymerases and provide robust theoretical support for developing novel polymerases for applications in drug discovery and related fields.