Beta-hairpin antimicrobial peptides as antifungal agents

About This Grant

PROJECT SUMMARY/ABSTRACT The overall goal of this proposal is to explore a small class of antimicrobial peptides with cyclic β-hairpin structure (β-AMPs) as antifungal agents through the development of high-throughput in silico screens and synthetic biology approaches to rapidly identify and optimize lead candidates and expand this small class’s known antifungal sequence space. The lack of treatment options for fungal infections, along with the increasing frequency of resistance in Candida species and Aspergillus fumigatus, is a major concern for global health, highlighting a critical need for innovation in antifungal discovery and development. Most β-AMP sequences demonstrate strong antifungal activity and their small size, macrocyclic structure, stability to proteases, and plethora of biological functions make them a relatively untapped well for antifungal drug development. Our team, made up of Drs. Randall, Wilke, and White have a wealth of experience in synthetic β-AMP discovery, computational biology, and clinical mycology respectively. Aim 1 will collect focused antifungal activity datasets to train machine learning algorithms to predict the antifungal potential of over 5.7 million variants of one promising β-AMP, Protegrin-1 (PG-1), against Candida albicans, Candida auris, and Aspergillus fumigatus. We will then cross reference these predictions with already established human toxicity predictions to develop a mutational profile promoting broadly acting, non-toxic, PG-1 antifungal activity and identify new clinical leads. Aim 2 will adapt a high-throughput functional antibacterial screening technology developed by Drs. Randall and Wilke to function in S. cerevisiae. We will use this technique to screen three independent peptide libraries consisting of over half a million synthetic sequences biased toward β-AMP sequence features and structure. This will likely identify hundreds of novel synthetic β-AMPs with antifungal activity. We can then validate the results of our screen in vitro by synthesizing the top 96 antifungal peptides candidates from each library and examining their activity directly via minimum inhibitory concentration against Candida albicans, Candida auris, and Aspergillus fumigatus and further characterize their structure and human toxicity. This will greatly expand the sequence space of known antifungal β-AMPs and identify new lead candidates for clinical optimization. Once developed, this proposal’s approach can be applied more broadly to screen additional β-AMPs in silico for variants with improved clinical potential and identify novel synthetic antifungal peptides from additional structural classes.