Use of computational modeling tools for the design and development of anti‐bacterial agents

Antimicrobial resistance (AMR) in high-priority bacterial pathogens represents a rapidly growing global health emergency, resulting from the convergence of multiple bacterial defense strategies, including membrane impermeability, activation of adaptive stress responses, and enzymatic degradation or modification of antibiotics. Overcoming these barriers requires a mechanistic understanding of resistance at atomistic resolution, which can guide the rational design of next-generation antimicrobial agents and adjuvants. In this doctoral thesis, molecular modeling, atomistic molecular dynamics (MD) simulations, and structure-based drug design (SBDD) are employed to analyze the determinants of bacterial resilience and translate them into design principles for innovative therapeutic strategies. The interaction between polymyxin antibiotics and realistic Pseudomonas aeruginosa OM models was investigated using extensive atomistic MD simulations. The analysis revealed how protonation states, charge distribution, and the spatial arrangement of colistin A cationic and hydrophobic groups modulate OM insertion and destabilization. These findings provide a mechanistic basis for the rational design of safer and more effective polymyxin analogues. Moreover, the interaction between colistin A and small molecules capable of restoring its activity in LPS-modified resistant strains was also investigated. In parallel Retinoid-based compounds were analyzed to identify structural features that promote membrane disruption in Staphylococcus aureus. Based on these findings, adarotene analogues were redesigned to increase their ability to interact with and cross the Gram-negative OM. The determinants of enzyme resistance were also examined, with particular attention to zinc metalloenzymes, specifically Helicobacter pylori carbonic anhydrase α and clinically relevant Metallo-β-lactamases (MBLs). For H. pylori α-carbonic anhydrase, molecular docking was used to evaluate the stable and selective binding mode of coumarin-based inhibitors that are potentially capable of interfering with the survival of the bacterium in acidic conditions. In the case of MBLs, molecular docking and MD simulations, accompanied by accurate parameterization of Zn(II)-ion(s) and the first coordination shell, revealed conserved zinc binding motifs and shared active site features that can be exploited for inhibitor design. These observations support the development of broad-spectrum inhibitors capable of chelating zinc and restoring the activity of β-lactam antibiotics against MBL-producing pathogens. Overall, this work translates atomistic insights on membrane and enzyme resistance into concrete design rules for next-generation antimicrobial agents and adjuvants.