Abstract / Summary
Chronic infection with Helicobacter pylori strains expressing the cytotoxin-associated gene A (CagA) oncoprotein is strongly implicated in gastric carcinogenesis. Membrane localisation of CagA via its phosphatidylserine-binding α 18 helix is essential for host signalling dysregulation, making this domain an attractive target for anti-virulence intervention. In the present study, a structure-based computational workflow was employed to identify Azadirachta indica -derived phytocompounds capable of targeting the membrane-binding α 18 helix of CagA. A total of 39 neem-derived compounds were subjected to site-directed (targeted) molecular docking against a grid box centred on Arg624 of the α 18 helix, and several candidates returned docking scores more negative than the reference compound DFMO. Because raw docking scores scale with ligand size, candidates were ranked using a pre-declared multi-parameter scheme combining docking score, ligand efficiency, lipophilicity-corrected efficiency, drug-likeness rule compliance and structural alerts; this scheme placed the flavonoids above the higher-scoring but markedly more lipophilic limonoids. Isorhamnetin, kaempferol and quercetin were retained as leads, their electronic structure was characterised by density functional theory (B3LYP/6-31G(d) optimisation with 6-311+G(d,p) single points, gas phase and implicit water; HOMO–LUMO gaps 3.59–3.69 eV versus 6.13 eV for the reference compound DFMO), and isorhamnetin and quercetin were advanced to 200 ns molecular dynamics simulation alongside the apo protein and the DFMO complex. Trajectory analyses, including RMSD, RMSF, radius of gyration (R g ), solvent-accessible surface area (SASA), and hydrogen bond monitoring, were complemented by descriptors computed at the helix itself. These show that α 18 flexibility is essentially unchanged across the systems (mean per-residue RMSF 0.174–0.189 nm, a spread comparable to the standard error of any one system) and that the helix is displaced as a body while its internal geometry is preserved in all three complexes. The descriptor that discriminates between the ligands is residence: isorhamnetin stays in the pocket for the whole equilibrated window, within 0.4 nm of the helix in 100% of frames and hydrogen-bonded to the protein in 88.6% of them, whereas the reference compound DFMO leaves the site. End-state binding free energy calculations gave a favourable total binding energy of − 12.17 ± 3.88 kcal/mol (block-averaged SEM 0.33) for the isorhamnetin complex, predominantly driven by van der Waals and electrostatic interactions, with per-residue decomposition localising the largest favourable contributions to the α 18 interface residues. Principal component and free energy landscape analyses, which describe the whole protein rather than the binding site, further indicated confined conformational sampling and a well-defined global energy minimum upon ligand binding. Collectively, these findings identify neem-derived flavonoids, particularly isorhamnetin, as predicted domain-specific binders of the CagA α 18 helix. All results are computational; binding, membrane-localisation blockade and anti-virulence activity remain hypotheses that require experimental testing, and the compounds are therefore described here as candidates rather than as established inhibitors.