Genome mining of ant-associated bacteria to identify novel secondary metabolites

Abstract

Antibiotic discovery has stagnated since the end of the 1960s, despite an increasing level of antimicrobial resistance in the clinic. However, recently there has been a renewed interest in searching for novel antibiotics, particularly in under-explored niches. Symbiotic relationships between bacteria and eukaryotes, for example plants and insects, can prove to be a fruitful source of novel secondary metabolites. Many ant species form mutualistic relationships with antibiotic-producing bacteria to protect their colonies against parasitic invasion. However, the antibiotic-producing potential of bacteria associated with different ant species has rarely been explored in detail. We isolated 20 bacteria from the colonies of ant species in the genera Acromrymex, Cyphomyrmex and Mycocepurus, which are all fungus-farming ants, as well as Allomerus and Tetraponera genera which predate on other insects. High quality genome sequences were generated for each bacterial strain using PacBio sequencing. AntiSMASH analysis showed that many of the strains had the potential to produce many secondary metabolites, including a modified candicidin-like compound and several other clusters with low percentage homology to known antimicrobial compounds. Under standard lab conditions less than 20 % of strains show bioactivity against bacteria Bacillus subtillis and fungi Candida albicans however their genomes suggest a much higher potential. Pleiotropic methods such as growth media variations and co-culture with other microorganisms are now being used to switch on cryptic clusters in these novel isolates. Genome mining of symbiotic strains could make a valuable contribution to antibiotic discovery since such strains are constantly having to evolve in response to their host.