Advancing asparaginase-based therapy for acute lymphoblastic leukemia through computational immunoengineering and molecular dynamics simulations

L-asparaginase is a critical enzyme in the treatment of acute lymphoblastic leukemia (ALL); however, its clinical application is frequently limited by immunogenicity and hypersensitivity reactions. This study utilized a comprehensive computational pipeline that integrated immunoinformatics, rational mutagenesis, molecular docking, molecular dynamics simulations, structural stability and conformational analyses, principal component analysis, dynamic cross-correlation analysis, secondary structure evaluation, and MM-GBSA free energy calculations to design and assess reduced-immunogenic variants of asparaginase from E. coli , E. chrysanthemi , and S. maxima . HLA-DRB1*07:01-restricted T-cell epitopes were identified via NetMHCIIpan, and anchor residues (P1, P4, P6, P9) were deliberately substituted to impair peptide–MHC binding while maintaining structural integrity. As a result of these findings, hydrophobic and polar uncharged residues appear to play a key role in peptide binding and may enhance immunogenicity when associated with HLA-DRB1*07:01 allele. Hydrophobic amino acid side chains fit well into the MHCII binding groove, forming strong interactions that stabilize the peptide-MHC complex. In contrast, polar or small residues generally contribute less effectively to binding interactions and were therefore considered for substitution with hydrophobic residues. The chosen mutations, T68Q in E. coli asparaginase, L174N in E. chrysanthemi asparaginase, and I64N in S. maxima asparaginase, exhibited diminished predicted immunogenicity while maintaining the integrity of the active-site architecture, as verified by structural alignment and docking analyses. 1500 ns of MD simulations (500 ns for each system) showed species-specific dynamic behaviors, with E. chrysanthemi asparaginase indicating the highest structural stability and optimal substrate binding (ΔG_bind as low as − 33.80 ± 0.85 kcal/mol), followed by S. maxima asparaginase-Asn complex, whereas E. coli ’s complex exhibited increased conformational flexibility and diminished binding affinity. PCA and DCCM analyses showed coordinated movements in S. maxima enzyme-ligand complex, intricate inter-domain coupling in E. coli ’s complex, and localized dynamics in E. chrysanthemi ’s complex. The E. chrysanthemi -Asn and E. coli -Asn complexes exhibited comparable radius of gyration, with mean values of 20.41 ± 0.15 Å and 20.51 ± 0.23 Å, respectively, whereas the S. maxima -Asn complex showed a lower mean value of 19.52 ± 0.15 Å, consistent with a more compact overall structure. The free energy landscapes produced a single dominant basin for S. maxima complex, two states separated by a 1.5-2 kcal/mol barrier for E. chrysanthemi complex and a broad shallow surface for E. coli complex. Hydrogen bond analysis confirmed the high-energy hotspot profiles, with the E. chrysanthemi- Asn complex exhibiting the highest number of simultaneous hydrogen bonds and individual occupancy values, S. maxima asparaginase displaying the highest number of distinct hydrogen-bonding contacts with Asn, and E. coli asparaginase showing fewer distinct substrate contacts with generally lower occupancy values. Our findings indicate that targeted epitope engineering can diminish predicted immunogenicity while maintaining structural stability and substrate-binding affinity. This comprehensive computational approach offers a logical foundation for the design of safer and more efficacious therapeutic asparaginase variants for the treatment of ALL, subject to experimental validation.