Abstract
Introduction: Pediatric gastrointestinal (GI) cancers with encephalopathy are complex. The role of key genes like ALDH18A1, involved in metabolism and tumorigenesis, remains unclear in these disorders. This study aims to use bioinformatics to explore a potential non-canonical structural interaction between the plant lectin concanavalin A (ConA) and the dysregulated cancer target ALDH18A1.
Methods: The amino acid sequences of the two target proteins were retrieved from the UniProt database and aligned using BLASTp. Their three-dimensional structures were then obtained via Protein Data Bank (PDB) data and the SwissModel server, and structural quality was assessed using Ramachandran plot, QMEAN, and QMEANDisCo metrics. Subsequently, protein–protein docking simulation was conducted using the ClusPro server, resulting in 30 proposed binding clusters.
Results: No significant sequence similarity was detected between ConA and ALDH18A1 (Identity<10%, non-significant E-value). Nevertheless, the structural model of ConA demonstrated high quality (Ramachandran outliers: 0.59%, QMEANDisCo: 0.87), while ALDH18A1, despite its structural complexity, showed a reliable model (Ramachandran outliers: 2.37%, QMEANDisCo: 0.72). The docking analysis revealed that several clusters—particularly clusters 3, 5, 11, and 15—had highly negative binding energies (as low as –1060.7), suggesting a thermodynamically favourable interaction with potential biological significance.
Conclusion: Despite no sequence similarity, a structural interaction between ConA and ALGH18A1 is plausible. This may influence immunity and cancer cell growth indirectly, providing a groundwork for developing targeted therapies for pediatric GI cancers. Further experimental validation is required to confirm these preliminary findings.