Antibiotic resistance is widely recognized as one of the most pressing global health challenges, resulting from the rapid and uncontrolled spread of multidrug-resistant bacterial and fungal pathogens. The progressive loss of effectiveness of conventional antibiotics highlights the urgent need for alternative therapeutic strategies. In this scenario, antimicrobial peptides (AMPs) have attracted increasing attention as promising candidates for the treatment of infectious diseases, due to their broad-spectrum activity, diverse mechanisms of action, and reduced likelihood of inducing resistance compared to conventional antibiotics.

This study investigates the potential of novel synthetic AMPs inspired by natural peptide scaffolds and developed through advanced rational design approaches. Natural peptides from a Mediterranean medical plant Charybdis pancration (Steinh.) Speta were used as starting templates for the design of multiple modified peptide sequences employing in silico tools based on artificial intelligence and deep-learning models. These computational strategies enabled the optimization of key physicochemical properties associated with antimicrobial and antibiofilm activity.

The designed peptides were subsequently evaluated using a combination of in silico analyses and in vitro tests to validate their predicted MICs against relevant Gram-negative pathogens (Acinetobacter baumannii and Pseudomonas aeruginosa) and Gram-positive pathogens (antibiotic-resistant Staphylococcus aureus). In particular, the in silico analysis predicted MIC values against Pseudomonas aeruginosa ranging from 48.8 to 10.9 µg/mL.

Overall, this work highlights the value of integrating natural peptide sources with artificial intelligence-based design methodologies for the discovery of new antimicrobial candidates. Such an approach may help to expand the current antimicrobial pipeline and provide innovative solutions to combat the growing threat of antibiotic resistance.