Engineering MHC Class I Molecules to Enhance Tumor Antigen Visibility for Improved Cancer Immunotherapy
Eunbin Kang
Presentation High School, San Jose, United States
Publication date: July 19, 2026
Presentation High School, San Jose, United States
Publication date: July 19, 2026
DOI: http://doi.org/10.34614/JIYRC2026I40
ABSTRACT
Cancer immunotherapies rely on effective antigen presentation through MHC Class I molecules, yet many tumor-associated peptides bind weakly to MHC, allowing cancer cells to escape immune surveillance. To address this limitation, we investigated whether engineered HLA-A02:01 variants could enhance peptide binding and improve antigen display. Using in silico structural modeling of the HLA-A02:01–β2-microglobulin–peptide complex (PDB 6Q3K), we mapped peptide-contact residues and performed alanine scanning to identify positions suitable for modification. DiffDock-based docking simulations guided the design of two mutant variants predicted to alter peptide stabilization in the binding groove. To validate these designs experimentally, we cloned, expressed, and purified wild-type HLA-A02:01, Mutant 1 (W167A), Mutant 2 (Y7,99,159,171A), and β2-microglobulin in E. coli. Although the proteins localized primarily to inclusion bodies, optimized colony selection and urea-based solubilization enabled high-yield purification. SDS-PAGE confirmed purity, and BCA assays indicated concentrations of 16–19 mg/mL, sufficient for biochemical testing. Functional assays were performed using a fluorescence-based peptide-binding system with pp65 peptide as the ligand. Wild-type HLA-A02:01 exhibited strong fluorescence, confirming correct refolding and high binding affinity. In contrast, both engineered mutants displayed significantly reduced fluorescence, indicating weakened peptide binding. While the mutations did not enhance affinity, the results demonstrate that MHC Class I binding properties are tunable through targeted structural engineering. This proof-of-concept establishes a foundation for future synthetic biology efforts aimed at improving antigen presentation. By combining computational design with iterative wet-lab optimization, our project highlights a scalable approach toward developing next-generation tools to strengthen immune recognition of cancer.
Cancer immunotherapies rely on effective antigen presentation through MHC Class I molecules, yet many tumor-associated peptides bind weakly to MHC, allowing cancer cells to escape immune surveillance. To address this limitation, we investigated whether engineered HLA-A02:01 variants could enhance peptide binding and improve antigen display. Using in silico structural modeling of the HLA-A02:01–β2-microglobulin–peptide complex (PDB 6Q3K), we mapped peptide-contact residues and performed alanine scanning to identify positions suitable for modification. DiffDock-based docking simulations guided the design of two mutant variants predicted to alter peptide stabilization in the binding groove. To validate these designs experimentally, we cloned, expressed, and purified wild-type HLA-A02:01, Mutant 1 (W167A), Mutant 2 (Y7,99,159,171A), and β2-microglobulin in E. coli. Although the proteins localized primarily to inclusion bodies, optimized colony selection and urea-based solubilization enabled high-yield purification. SDS-PAGE confirmed purity, and BCA assays indicated concentrations of 16–19 mg/mL, sufficient for biochemical testing. Functional assays were performed using a fluorescence-based peptide-binding system with pp65 peptide as the ligand. Wild-type HLA-A02:01 exhibited strong fluorescence, confirming correct refolding and high binding affinity. In contrast, both engineered mutants displayed significantly reduced fluorescence, indicating weakened peptide binding. While the mutations did not enhance affinity, the results demonstrate that MHC Class I binding properties are tunable through targeted structural engineering. This proof-of-concept establishes a foundation for future synthetic biology efforts aimed at improving antigen presentation. By combining computational design with iterative wet-lab optimization, our project highlights a scalable approach toward developing next-generation tools to strengthen immune recognition of cancer.
