Mining multidrug-resistant desert soil bacteria for biocontrol activity and biologically-active compounds

Eddie Cytryn,Mark R. Liles,Omer Frenkel

DOI: https://doi.org/10.32747/2014.7598174.bard

2014-01-01

Abstract:Control of agro-associated pathogens is becoming increasingly difficult due to increased resistance and mounting restrictions on chemical pesticides and antibiotics. Likewise, in veterinary and human environments, there is increasing resistance of pathogens to currently available antibiotics requiring discovery of novel antibiotic compounds. These drawbacks necessitate discovery and application of microorganisms that can be used as biocontrol agents (BCAs) and the isolation of novel biologically-active compounds. This highly-synergistic one year project implemented an innovative pipeline aimed at detecting BCAs and associated biologically-active compounds, which included: (A) isolation of multidrug-resistant desert soil bacteria and root-associated bacteria from medicinal plants; (B) invitro screening of bacterial isolates against known plant, animal and human pathogens; (C) nextgeneration sequencing of isolates that displayed antagonistic activity against at least one of the model pathogens and (D) in-planta screening of promising BCAs in a model bean-Sclerotiumrolfsii system. The BCA genome data were examined for presence of: i) secondary metabolite encoding genes potentially linked to the anti-pathogenic activity of the isolates; and ii) rhizosphere competence-associated genes, associated with the capacity of microorganisms to successfully inhabit plant roots, and a prerequisite for the success of a soil amended BCA. Altogether, 56 phylogenetically-diverse isolates with bioactivity against bacterial, oomycete and fungal plant pathogens were identified. These strains were sent to Auburn University where bioassays against a panel of animal and human pathogens (including multi-drug resistant pathogenic strains such as A. baumannii 3806) were conducted. Nineteen isolates that showed substantial antagonistic activity against at least one of the screened pathogens were sequenced, assembled and subjected to bioinformatics analyses aimed at identifying secondary metabolite-encoding and rhizosphere competence-associated genes. The genome size of the bacteria ranged from 3.77 to 9.85 Mbp. All of the genomes were characterized by a plethora of secondary metabolite encoding genes including non-ribosomal peptide synthase, polyketidesynthases, lantipeptides, bacteriocins, terpenes and siderophores. While some of these genes were highly similar to documented genes, many were unique and therefore may encode for novel antagonistic compounds. Comparative genomic analysis of root-associated isolates with similar strains not isolated from root environments revealed genes encoding for several rhizospherecompetence- associated traits including urea utilization, chitin degradation, plant cell polymerdegradation, biofilm formation, mechanisms for iron, phosphorus and sulfur acquisition and antibiotic resistance. Our labs are currently writing a continuation of this feasibility study that proposes a unique pipeline for the detection of BCAs and biopesticides that can be used against phytopathogens. It will combine i) metabolomic screening of strains from our collection that contain unique secondary metabolite-encoding genes, in order to isolate novel antimicrobial compounds; ii) model plant-based experiments to assess the antagonistic capacities of selected BCAs toward selected phytopathogens; and iii) an innovative next-generation-sequencing based method to monitor the relative abundance and distribution of selected BCAs in field experiments in order to assess their persistence in natural agro-environments. We believe that this integrated approach will enable development of novel strains and compounds that can be used in large-scale operations.