Miruna Serian: Unsupervised learning elucidates the interplay between conformational flexibility and aggregation in synergistic antimicrobial peptides

Despite this, the development of new antibiotics has declined, and the World Health Organization (WHO) describes the antibacterial clinical and preclinical pipeline as stagnant and far from meeting global needs (1). Therefore, there is an urgent need for increased efforts to develop alternatives to current antimicrobial agents. Consequently, the generation of novel medications to control and treat infections caused by multidrug-resistant pathogens has become a pressing priority for the scientific community. Antimicrobial peptides have quickly gained traction as promising drug candidates because of their potency against both Gram-negative and Gram-positive bacteria (2). Antimicrobial peptides are evolutionary conserved components of the immune system, found in almost all life forms, from prokaryotes to humans (3), and have been shown to have distinct roles.

While in higher life forms they are produced to protect the host against infection, bacteria can also produce AMPs to kill other bacteria competing for the same environment (4). Despite their potential, several challenges hinder the widespread use of antimicrobial peptides as antibiotics alternatives. These include concerns about host toxicity (5), the emergence of bacterial resistance (6) and high production costs (7). To address these limitations, combining different antimicrobial peptides has emerged as a promising strategy. Similar to combination therapy with traditional drugs, combining antimicrobial peptides can lead to synergistic effects, potentially reducing the necessary dosage, minimizing side effects, and lowering the risk of resistance development (8).

Despite the benefits of synergistic combinations of AMPs, the mechanisms of their synergy are not yet fully understood. Computational and experimental studies have proposed several
mechanisms, including pore formation. For instance, combining PGLa and MAG2, two peptides produced in the skin of Xenopus laevis can lead to the formation of a toroidal pore structure, that can in turn lead to more membrane disruption (9). In other cases, such as the interaction between the two AMPs produced by bumblebees, abaecin and the pore forming AMP hymenoptaecin, distinct mechanisms are observed; hymenoptaecin forms membrane pores, destabilizing bacterial membranes and allowing abaecin entry into bacterial cells (10). Abaecin can also synergise with pore-forming peptides from other organisms (11). Additionally, synergistic behaviour may arise from complementary mechanisms, such as the in the case of coleoptericin and defensin. Coleoptericin acts to improve host survival while defensin can reduce bacterial load (12). Overall, pore formation or peptide aggregation remains among the most suggested mechanisms of action for synergistic peptides.

Read and download full publication here