Actinomycetes, such as mycobacteria and streptomycetes, synthesize α-glucan with α-1,4 linkages and α-1,6 branching to help evade immune responses and to store carbon. α-Glucan is thought to resemble glycogen except for having shorter constituent linear chains. However, the fine structure of α-glucan and how it can be defined by the maltosyl transferase GlgE and branching enzyme GlgB were not known. Using a combination of enzymolysis and mass spectrometry, we compared the properties of α-glucan isolated from actinomycetes with polymer synthesized in vitro by GlgE and GlgB. We now propose the following assembly mechanism. Polymer synthesis starts with GlgE and its donor substrate, α-maltose 1-phosphate, yielding a linear oligomer with a degree of polymerization (∼16) sufficient for GlgB to introduce a branch. Branching involves strictly intrachain transfer to generate a C chain (the only constituent chain to retain its reducing end), which now bears an A chain (a nonreducing end terminal branch that does not itself bear a branch). GlgE preferentially extends A chains allowing GlgB to act iteratively to generate new A chains emanating from B chains (nonterminal branches that themselves bear a branch). Although extension and branching occur primarily with A chains, the other chain types are sometimes extended and branched such that some B chains (and possibly C chains) bear more than one branch. This occurs less frequently in α-glucans than in classical glycogens. The very similar properties of cytosolic and capsular α-glucans from Mycobacterium tuberculosis imply GlgE and GlgB are sufficient to synthesize them both.

Streptomyces bacteria are ubiquitous in soils and are well-known for producing secondary metabolites, including antimicrobials. Increasingly, they are being isolated from plant roots and several studies have shown they are specifically recruited to the rhizosphere and the endosphere of the model plant Arabidopsis thaliana. Here we test the hypothesis that Streptomyces bacteria have a beneficial effect on A. thaliana growth and could potentially be used as plant probiotics. To do this, we selectively isolated streptomycetes from surface washed A. thaliana roots and generated high quality genome sequences for five strains which we named L2, M2, M3, N1 and N2. Re-infection of A. thaliana plants with L2, M2 and M3 significantly increased plant biomass individually and in combination whereas N1 and N2 had a negative effect on plant growth, likely due to their production of polyene natural products which can bind to phytosterols and reduce plant growth. N2 exhibits broad spectrum antimicrobial activity and makes filipin-like polyenes, including 14-hydroxyisochainin which inhibits the Take-all fungus, Gaeumannomyces graminis var. tritici. N2 antifungal activity as a whole was upregulated ∼2-fold in response to indole-3-acetic acid (IAA) suggesting a possible role during competition in the rhizosphere. Furthermore, coating wheat seeds with N2 spores protected wheat seedlings against Take-all disease. We conclude that at least some soil dwelling streptomycetes confer growth promoting benefits on A. thaliana while others might be exploited to protect crops against disease. Importance. We must reduce reliance on agrochemicals and there is increasing interest in using bacterial strains to promote plant growth and protect against disease. Our study follows up reports that Arabidopsis thaliana specifically recruits Streptomyces bacteria to its roots. We test the hypothesis that they offer benefits to their A. thaliana hosts and that strains isolated from these plants might be used as probiotics. We isolated Streptomyces strains from A. thaliana roots and genome sequenced five phylogenetically distinct strains. Genome mining and bioassays indicated that all five have plant growth promoting properties, including production of IAA, siderophores and ACC deaminase. Three strains significantly increased A. thaliana growth in vitro and in combination in soil. Another produces potent filipin-like antifungals and protected germinating wheat seeds against the fungal pathogen Gaeumannomyces graminis var. tritici (wheat Take-all fungus). We conclude that introducing Streptomyces strains into the root microbiome provides significant benefits to plants.

Fungus-growing attine ants are under constant threat from fungal pathogens such as the specialized mycoparasite Escovopsis , which uses combined physical and chemical attack strategies to prey on the fungal gardens of the ants. In defence, some species assemble protective microbiomes on their exoskeletons that contain antimicrobial-producing Actinobacteria. Underlying this network of mutualistic and antagonistic interactions are an array of chemical signals. Escovopsis weberi produces the shearinine terpene-indole alkaloids, which affect ant behaviour, diketopiperazines to combat defensive bacteria, and other small molecules that inhibit the fungal cultivar. Pseudonocardia and Streptomyces mutualist bacteria produce depsipeptide and polyene macrolide antifungals active against Escovopsis spp. The ant nest metabolome is further complicated by competition between defensive bacteria, which produce antibacterials active against even closely related species.

To meet the ever-growing demand of antibiotic discovery, new chemical matter and antibiotic targets are urgently needed. Many potent natural product antibiotics which were previously discarded can also provide lead molecules and drug targets. One such example is the structurally unique β-lactone obafluorin, produced by Pseudomonas fluorescens ATCC 39502. Obafluorin is active against both Grampositive and -negative pathogens, however the biological target was unknown. We now report that obafluorin targets threonyl-tRNA-synthetase and we identify a homologue, ObaO, which confers self-immunity to the obafluorin producer. Disruption of obaO in P. fluorescens ATCC 39502 results in obafluorin sensitivity, whereas expression in sensitive E. coli strains confers resistance. Enzyme assays demonstrate that E. coli threonyl-tRNA synthetase is fully inhibited by obafluorin, whereas ObaO is only partly susceptible, exhibiting a very unusual partial inhibition mechanism.Altogether, our data highlight the utility of a self-immunity guided approach for the identification of an antibiotic target de novo and will ultimately enable the generation of improved obafluorin variants..