Validation of a novel method for screening fungi for new antibiotics using bioluminescent bacteria

Abstract

The evolution of antibiotic resistant bacteria is a serious problem in modern medicine and there is an urgent need to discover new types of antibiotics. Historically the most reliable source for new types of antibiotics has been microbes themselves, and it is possible that microbes may make many more types of antibiotics than the limited number currently known. One promising source of new antibiotics is the International Collection of Microbes from Plants (ICMP), which contains over 15,000 species of fungi including many indigenous to NZ. Traditionally, screening microbes for new antibiotics has been a slow, labour intensive, hit and miss process that generates many false leads due to already-known compounds and compounds with non-selective toxicity. The aim of this study was to test the validity of using using bioluminescently-tagged bacteria to screen the ICMP fungi for antimicrobial activity. I focused on working with lux-tagged strains of Staphylococcus aureus, Esherichia coli, and Pseudomonas aeruginosa, three common bacterial pathogens that are becoming increasingly resistant to known antibiotics. Out of 238 different ICMP fungi screened in triplicate, I identified 18 that selectively inhibited S. aureus luminescence and 23 that non-selectively inhibited the luminescence of all of our target bacteria. Crude extracts were prepared from 10 of the active fungi using a combination of dichloromethane, ethyl acetate, and methanol. The minimum inhibitory concentration (MIC) of each crude extract was estimated by its effects on both bacterial bioluminescence and optical density over 24 hours of incubation. The minimum bactericidal concentration (MBC) of each crude extract was also determined by plate counting after 24 hours of exposure. Most of the crude extracts inhibited bacterial luminescence in a concentration dependent manner, however only 5 showed MICs or MBCs of 1 mg/ml or less. Three of the five most active crude extracts were all from different isolates of Cercospora fresen, a common plant pathogen that causes leaf spots in a variety of plant species. Two of these had an MIC and MBC of less than 16 μg/ml against S. aureus, but neither were effective against E.coli or P.aeruginosa (MIC and MBC >1mg/ml). The most active Cercospora extract was fractionated and the most active fraction was found to contain a high concentration of cercosporins, known photo-activated toxins produced by several species of the genus Cercospora. Previous studies have found that cercosporins exert non-specific toxic effects on Gram-positive and Gram-negative bacteria, as well as plant and animal cells and would therefore not make useful antibiotics. Our finding of selective toxicity against S. aureus is not consistent with this, and additional studies are needed to resolve this discrepancy. Overall, while we did not discover any new types of antibiotics during the course of this project, we have shown that our screening method works, has several advantages over traditional methods, and should facilitate the future discovery of new types of antibiotics from fungi.