Lessons from assembling a microbial natural product and pre-fractionated extract library in an academic laboratory.

Microbial natural products are specialized metabolites that are sources of many bioactive compounds including antibiotics, antifungals, antiparasitics, anticancer agents, and probes of biology. The assembly of libraries of producers of natural products has traditionally been the province of the pharmaceutical industry. This sector has gathered significant historical collections of bacteria and fungi to identify new drug leads with outstanding outcomes-upwards of 60% of drug scaffolds originate from such libraries. Despite this success, the repeated rediscovery of known compounds and the resultant diminishing chemical novelty contributed to a pivot from this source of bioactive compounds toward more tractable synthetic compounds in the drug industry. The advent of advanced mass spectrometry tools, along with rapid whole genome sequencing and in silico identification of biosynthetic gene clusters that encode the machinery necessary for the synthesis of specialized metabolites, offers the opportunity to revisit microbial natural product libraries with renewed vigor. Assembling a suitable library of microbes and extracts for screening requires the investment of resources and the development of methods that have customarily been the proprietary purview of large pharmaceutical companies. Here, we report a perspective on our efforts to assemble a library of natural product-producing microbes and the establishment of methods to extract and fractionate bioactive compounds using resources available to most academic labs. We validate the library and approach through a series of screens for antimicrobial and cytotoxic agents. This work serves as a blueprint for establishing libraries of microbial natural product producers and bioactive extract fractions suitable for screens of bioactive compounds. ONE-SENTENCE SUMMARY: Natural products are key to discovery of novel antimicrobial agents: Here, we describe our experience and lessons learned in constructing a microbial natural product and pre-fractionated extract library.

Structural and molecular rationale for the diversification of resistance mediated by the Antibiotic_NAT family.

The environmental microbiome harbors a vast repertoire of antibiotic resistance genes (ARGs) which can serve as evolutionary predecessors for ARGs found in pathogenic bacteria, or can be directly mobilized to pathogens in the presence of selection pressures. Thus, ARGs from benign environmental bacteria are an important resource for understanding clinically relevant resistance. Here, we conduct a comprehensive functional analysis of the Antibiotic_NAT family of aminoglycoside acetyltransferases. We determined a pan-family antibiogram of 21 Antibiotic_NAT enzymes, including 8 derived from clinical isolates and 13 from environmental metagenomic samples. We find that environment-derived representatives confer high-level, broad-spectrum resistance, including against the atypical aminoglycoside apramycin, and that a metagenome-derived gene likely is ancestral to an aac(3) gene found in clinical isolates. Through crystallographic analysis, we rationalize the molecular basis for diversification of substrate specificity across the family. This work provides critical data on the molecular mechanism underpinning resistance to established and emergent aminoglycoside antibiotics and broadens our understanding of ARGs in the environment.

Identifying novel ß-lactamase substrate activity through in silico prediction of antimicrobial resistance.

Diagnosing antimicrobial resistance (AMR) in the clinic is based on empirical evidence and current gold standard laboratory phenotypic methods. Genotypic methods have the potential advantages of being faster and cheaper, and having improved mechanistic resolution over phenotypic methods. We generated and applied rule-based and logistic regression models to predict the AMR phenotype from Escherichia coli and Pseudomonas aeruginosa multidrug-resistant clinical isolate genomes. By inspecting and evaluating these models, we identified previously unknown ß-lactamase substrate activities. In total, 22 unknown ß-lactamase substrate activities were experimentally validated using targeted gene expression studies. Our results demonstrate that generating and analysing predictive models can help guide researchers to the mechanisms driving resistance and improve annotation of AMR genes and phenotypic prediction, and suggest that we cannot solely rely on curated knowledge to predict resistance phenotypes.