Actinomycetota bioprospecting from ore-forming environments

Abstract

Natural products from Actinomycetota have served as inspiration for many clinically relevant therapeutics. Despite early triumphs in natural product discovery, the rate of unearthing new compounds has decreased, necessitating inventive approaches. One promising strategy is to explore environments where survival is challenging. These harsh environments are hypothesized to lead to bacteria developing chemical adaptations (e.g. natural products) to enable their survival. This investigation focuses on ore-forming environments, particularly fluoride mines, which typically have extreme pH, salinity and nutrient scarcity. Herein, we have utilized metagenomics, metabolomics and evolutionary genome mining to dissect the biodiversity and metabolism in these harsh environments. This work has unveiled the promising biosynthetic potential of these bacteria and has demonstrated their ability to produce bioactive secondary metabolites. This research constitutes a pioneering endeavour in bioprospection within fluoride mining regions, providing insights into uncharted microbial ecosystems and their previously unexplored natural products.

Keywords: Actinomycetota; fluoride mines; genome mining; natural products; ore-forming environments.

Fig. 1.. Microbiome composition of soils from PMT and Topaz Mountain mines. (a) Relative abundance of bacterial phyla from each soil. Note that only the top five most abundant phyla are indicated. Details of all phyla can be found in Fig. S1 and File S1. (b) Relative abundance of phyla that were at least twofold different between PMT (dark blue) and Topaz Mountain (light blue) soils.

Fig. 2.. Actinomycetota abundance in the Topaz Mountain and PMT soils. (a) Relative abundance of Actinomycetota genera from each soil. Only the top five genera are indicated. Genus abundance <1 % (abund) includes any unknown taxa. See File S1 for more details. (b) Comparison of relative abundance of genera from PMT (dark blue) and Topaz Mountain (light blue) other than Streptomyces and those at <1 %. Colours of genus names reflect colours of bars in (a).

Fig. 3.. Deciphering metabolic pathways from metagenomic data. (a) Actinomycetota metabolic landscape. The graph displays the relative abundance of orthologues from Topaz Mountain (inner circle) and PMT (outer circle), sorted into 13 pathway modules and two signature modules (gene set and module set). M=metabolism; B=biosynthesis; PKS=polyketide synthase. (b) Orthologues with differing abundances. Relative abundance of orthologues from (a) with at least twofold difference in percentage relative abundance. Colours for orthologues indicate the pathway module they belong to, as seen in (a). (c) Comparative analysis of functional orthologues. The totality of orthologues identified for each site were divided into two groups, actinomycetota and bacteria (bacteria not belonging to the phylum Actinomycetota), and subsequently categorized into ‘core’, ‘soft-core’, ‘shell’ and ‘cloud’ clusters. (d) Main exclusive actinomycetota pathways. The graph shows the proportion of orthologes for the principal actinomycetota pathways found exclusively in Topaz Mountain and PMT. Certain pathways unique to specific sites are also identified. *Enediyne biosynthesis: kedarcidin, C1027 chromophores; **Xenobiotic metabolism: benzoate degradation; ***Membrane transport/uptake: xylobiose, erythritol, glutamate, galactitol and l-ascorbate; and Membrane trasport/export: lantibiotics; ****Enediyne biosynthesis: polyene macrolides; †Type I PKS: calicheamicin and maduropeptin; ††Type II PKS: ansamycins; PKS sugar biosynthesis: jadomycin, nogalamycin and landomycins.

Fig. 4.. Big-SCAPE network with 0.3 cut-off BGC similarity value. BGC clusters and singletons from six indicated strains. Clusters with BGCs similar to known BGCs in the MiBIG database are indicated (red is BGC from MiBIG). Shapes indicate the type of BGC. Colour indicates the source of the BGC.

Fig. 5.. MolNET-enhanced GNPS network. (a) A chemical classified network obtained with the MolNetEnhancer tool. Each colour represents a unique chemical superfamily, while internal labels within the network specify the chemical subclass. Size of node indicates relative ion abundance in the network. Following the same colour code, the bars at lower left denote the relative abundance of each superfamily in each strain and as a whole. (b) Most significant sub-networks from (a) presented individually. They are colour-coded with a 15-colour scheme to represent their bacterial origin. Edges in the graph show the direction from the parent to the fragment ion.