In-silico design and In-Vitro Validation of a Multi-Epitope Peptide Vaccine Targeting Triple-Negative Breast Cancer (TNBC)

Pengjun Zhou¹Xing Shi²Jinquan Xia³Hong Hu^3*

• ¹Guangdong Pharmaceutical University, Guangzhou, Guangdong Province, China
• ²National Genomics Data Center, Beijing Institute of Genomics, Chinese Academy of Sciences (CAS), Beijing, Beijing Municipality, China
• ³Shenzhen People's Hospital, Jinan University, Shenzhen, China

This study aimed to identify immunodominant epitopes from a panel of triple-negative breast cancer (TNBC)-associated proteins-MZF-1, Mucin-1, SOX-9, Keratin 5, Keratin 14, Twist1, and Progranulin (GP88)-to design multi-epitope peptide vaccines capable of eliciting robust anti-tumor immune responses.: A comprehensive immunoinformatics pipeline was employed. Amino acid sequences were retrieved from UniProt and analyzed to predict CTL, HTL, B-cell, and IFN-γ-inducing epitopes. Top candidates were filtered based on antigenicity, allergenicity, glycosylation, and HLA coverage. Molecular docking was conducted with HLA alleles to assess binding affinity. Five multiepitope vaccine constructs were designed using different adjuvants (GM-CSF, β-defensin, IL-2, cholera enterotoxin, and 50S ribosomal protein L7/L12), and enhanced with PADRE and HEYGAEALERA sequences. Structural modelling, refinement, disulfide engineering, and validation (via Robetta, GalaxyRefine, ProSA, and Ramachandran plots) were performed, followed by docking with TLR2 and TLR4. Immune simulation assessed cytokine responses and memory generation. In-vitro validation using MDA-MB-231 cells tested immunostimulatory activity of topranked CTL peptides. Results: Thirteen CD8+ CTL, thirteen CD4+ HTL, and seven B-cell epitopes were selected based on favourable immunogenic properties and high HLA promiscuity. Constructs V1 (GM-CSF-linked) and V5 (β-defensin-linked) exhibited superior TLR2/4 docking affinity. Immune simulation showed V2 and V5 induced strong cytokine release and memory cell responses. In vitro assays demonstrated enhanced expression of MZF-1, SOX-9, and Twist1, confirming epitope-driven immune activation. Conclusion: This study successfully identified potent immunogenic epitopes from TNBCassociated proteins and constructed promising multi-epitope vaccines. Constructs V1 and V5 demonstrated superior immunogenicity and TLR binding, while V2 and V5 induced strong immune responses in silico. These findings provide a foundation for developing effective peptide vaccines against TNBC.