Beyond structural divergence: Multiscale computational immunogenicity modelling of botulinum neurotoxin E and cross-reactivity in botulinum neurotoxin A primed hosts.

Botulinum neurotoxin type E (BoNT/E) is clinically attractive for its rapid onset and short duration of effect but carries uncertain immunogenic risks. Using sequence analysis, structural modelling, and immunoinformatics and multiscale computational workflow platform (AesthetiSIM™), we compared BoNT/E with type A (BoNT/A) across multiple molecular scales. Pairwise alignments revealed modest global identity (34 %) but ≥70 % conservation in immunodominant epitopes within the SNAP-25 cleavage groove and SV2-binding loops, with structural superposition yielding backbone RMSD values of 1.9-2.5 Å. BoNT/E epitopes exhibited greater solvent accessibility (+7-9 % SASA) and more positive electrostatic potentials (+0.42 vs + 0.11 kT/e), favouring HLA class II groove compatibility. Epitope predictions across 27 alleles showed 73 % enrichment of strong binders (IC50 ≤ 50 nM) for BoNT/E, with higher entropy values (0.74-0.83 vs 0.58-0.64), indicating broader repertoire coverage. Docking and molecular dynamics confirmed stronger stability of BoNT/E-HLA complexes (ΔG -44 to -49 kcal/mol, 5-6 hydrogen bonds) compared with BoNT/A (ΔG -30 to -34 kcal/mol, 2-3 bonds). Recall modelling demonstrated that BoNT/A priming amplified BoNT/E binding affinity by 13-16 %, accelerating T-cell activation thresholds. Systems immunology simulations projected neutralizing antibody risk at 9.2 % for BoNT/E versus 7.2 % for BoNT/A at 10 years, with combined A + E regimens reaching 13.7 %. Population coverage analyses predicted BoNT/E epitopes would engage 94 % of individuals globally, increasing to 98 % with A + E. These molecular findings identify BoNT/E as an epitope-dense, highly accessible, and cross-reactive antigen with amplified immunogenicity under simulation, necessitating further exploration with longer clinical trials.