Computational Identification of Aggregation-Prone Regions in Brain-Derived Neurotrophic Factor (BDNF)

Background

Brain-Derived Neurotrophic Factor (BDNF) is a promising therapeutic protein for neurological disorders; however, its clinical application is limited by aggregation susceptibility, which can compromise stability, efficacy, and safety. Identification of aggregation-prone structural determinants is therefore essential for rational stabilization strategies.

Materials and Methods

The homodimeric structure of human BDNF (PDB ID: 1BND) was analyzed using the BioLuminate module of Schrödinger for residue-level structural interrogation and energetic decomposition. Aggregation propensity was evaluated using AggScore, Aggrescan, and Zyggregator algorithms. Residue-wise hydrophobic energetic contributions, stabilizing and destabilizing interaction energies, Solvent-Accessible Surface Area (SASA), and patch surface areas were quantified for 293 residues across both chains. Descriptive statistics, correlation analysis, and chain-wise comparisons were performed to identify dominant aggregation hotspots.

Results

Aggregation scores demonstrated marked heterogeneity (mean 2.78±3.26; maximum 17.13), revealing a strongly right-skewed distribution dominated by isolated high-intensity hotspots. Chain A exhibited a prominent aggregation-prone cluster within residues 60-63, centered on GLY62, characterized by elevated aggregation score and increased solvent accessibility. Chain B displayed a broader but lower-intensity aggregation pattern, with consistent signatures observed in the C-terminal region (95-110). Global statistical analysis showed no significant linear correlation between hydrophobic energetic contribution and aggregation score, indicating that aggregation susceptibility is governed by multifactorial structural determinants rather than hydrophobicity alone.

Conclusion

Aggregation propensity in BDNF appears to be concentrated within discrete structural microdomains. The identified hotspot provides a structural basis for rational protein engineering and formulation strategies to enhance therapeutic stability. Experimental validation is required to confirm these computational findings.