Catholic University of Brasilia and MIT researchers have developed a streamlined approach to come up with more potent and new versions of antimicrobial peptides to fight off bacteria that have become resistant to existing antibiotics. One potential candidate for such drugs has already been yielded using the new strategy that depends on a computer algorithm imitating evolution’s natural process. It has been found to successfully kill bacteria in mice. The computational approach is labeled as much more time and cost-effective by Areces Foundation Fellow and MIT postdoc, Cesar de la Fuente-Nunez. Computers could be used as a tool to discover new antimicrobial peptide sequences, said de la Fuente-Nunez.
Guavanin 2 Tested to Fight Gram-negative Bacteria Better than Pg-AMP1
The computer algorithm used by de la Fuente-Nunez and his colleagues incorporated principles similar to Darwin’s natural selection theory. It could start with a random peptide sequence, produce scores of variants, and examine them for the required traits specified by the researchers. The study commenced with antimicrobial peptide Pg-AMP1 present in guava plant seeds. Then the algorithm was asked to generate peptide sequences having a couple of features that could aid peptides to infiltrate bacterial membranes using two features: a certain hydrophobicity level and a propensity to form alpha helices.
100 most promising candidates for peptide sequences out of tens of thousands evaluated by the algorithm were synthesized to be tested against lab-grown bacteria. Containing 20 amino acids, guavanin 2 was noted as a top performer by the researchers. Though it contains only a single glycine molecule, it is found to be rich in arginine, unlike the richness of amino acid glycine observed in the original Pg-AMP1 peptide. This shows that guavanin 2 could be a more potent candidate against bacteria such as Gram-negative which is responsible for UTIs, pneumonia, and HAIs.