Antibiotic resistance has become a serious global problem, threatening countless lives of human beings. Together with the recent decline of novel antibiotics introduced to the market, it is the time for us to place more efforts to study antibiotic resistance mechanisms and to discover new antibiotics. Natural products of microbial origin and their derivatives have been the main source of antibiotic drugs. Especially, nonribosomal peptides (NRPs), synthesized by large, modular, multifunctional enzymes known as nonribosomal peptide synthetases (NRPSs), are among the most widespread and structurally diverse antibiotics in nature. In addition to proteinogenic amino acids, NRPSs can also incorporate nonproteinogenic amino acids or carboxylic acids, acyl tails and cyclic structures, d...[ Read more ]

Antibiotic resistance has become a serious global problem, threatening countless lives of human beings. Together with the recent decline of novel antibiotics introduced to the market, it is the time for us to place more efforts to study antibiotic resistance mechanisms and to discover new antibiotics. Natural products of microbial origin and their derivatives have been the main source of antibiotic drugs. Especially, nonribosomal peptides (NRPs), synthesized by large, modular, multifunctional enzymes known as nonribosomal peptide synthetases (NRPSs), are among the most widespread and structurally diverse antibiotics in nature. In addition to proteinogenic amino acids, NRPSs can also incorporate nonproteinogenic amino acids or carboxylic acids, acyl tails and cyclic structures, dramatically expanding their bioactive chemical space.

Traditional screening platform for antibiotic natural products is extremely expensive and labor-intensive. The introduction of genome mining into natural product discovery has accelerated the process of bioactive compound discovery, yet still insufficient. Fortunately, with the dramatically increasing information of bacterial genomes in public domains, now we can perform large scale genome mining, which is impossible in the past. Although plenty of genome mining methods have been developed, most of them do not provide a direct view into structural information of natural products. Molecular networking can somehow fill this gap, but it requires the extraction of natural products from the producers, which is low-efficient and may not be applicable if gene clusters are silent.

In my thesis we established a genome mining platform that helped us analyze NRPs structurally. By applying this platform to 5585 complete bacteria genomes spanning the entire domain of bacteria, we first demonstrated a mechanism of resistance towards NRP antibiotics that was based on hydrolytic cleavage by D-stereospecific peptidases, providing a potential early indicator of antibiotic resistance to NRP antibiotics.

Moreover, we applied this genome mining platform to cationic NRPs to investigate their biosynthesis capacity and facilitate their genome-guided discovery for Gram-negative antibiotics. Cationic peptides have the cationic amino acid residues that not only facilitate their penetration through the highly impermeable outer membrane of Gram-negative bacteria but also enable their interaction with multiple anionic intracellular targets. Applying this platform to cationic NRPs from 7,395 bacteria genomes, we identified two novel compounds — brevicidine and laterocidine that showed strong bactericidal activities against Gram-negative pathogens, high efficacy in animal model, and low risk of resistance. Furthermore, using the predicted primary peptide sequence, we discovered enterocidine as a potential adjuvant for antibiotics that potentiate conventional Gram-positive antibiotics in Gram-negative pathogens. These findings will probably accelerate the discovery and development of antibiotics against Gram-negative bacteria.