With the advent of rapid automated in silico identification of biosynthetic gene clusters (BGCs), genomics presents vast opportunities to accelerate natural product (NP) discovery. However, prolific NP producers, Streptomyces, are exceptionally GC-rich (>80%) and highly repetitive within BGCs. These pose challenges in sequencing and high-quality genome assembly which are currently circumvented via intensive sequencing. Here, we outline a more cost-effective workflow using multiplex Illumina and Oxford Nanopore sequencing with hybrid long-short read assembly algorithms to generate high quality genomes. Our protocol involves subjecting long read-derived assemblies to up to 4 rounds of polishing with short reads to yield accurate BGC predictions. We successfully sequenced and assembled 8 GC-rich Streptomyces genomes whose lengths range from 7.1 to 12.1 Mb with a median N50 of 8.2 Mb. Taxonomic analysis revealed previous misrepresentation among these strains and allowed us to propose a potentially new species, Streptomyces sydneybrenneri. Further comprehensive characterization of their biosynthetic, pan-genomic and antibiotic resistance features especially for molecules derived from type I polyketide synthase (PKS) BGCs reflected their potential as alternative NP hosts. Thus, the genome assemblies and insights presented here are envisioned to serve as gateway for the scientific community to expand their avenues in NP discovery.

Fig. 1. Effect of polishing long read assemblies on genome quality. (A) Percentage lineage-specific completeness and contamination levels in polished assemblies as assessed by BUSCO. (B) Effect of polishing on completeness and contamination. (C) Effect of polishing on completeness in polished assemblies as assessed by CheckM. (D) Effect of polishing on number of BGC identified in assemblies. R1, R2, R3 and R4 in (B–D) correspond to the number of rounds of polishing the long read assemblies with short reads using Pilon (v1.24).

Fig. 2. Molecular networking cluster of Desferrioxamine E, and its putative biosynthetic gene cluster (BGC) in Streptomyces sp. A44034.(A) Desferrioxamine E and its analogs as determined by their MS/MS spectra and molecular networking (B) Revolution of annotated BGCs with increasing polishing rounds. Overview of clusters and their relative positions on the genome, along with multi-gene alignment as analysed using antiSMASH (v6.0) are shown. Blue arrows are used to annotate location of desferrioxamine BGC. Gene cluster table of the putative BGC is available in Table S5 of the paper. Pilon 1–4 refers to the assembly after 1–4 rounds of polishing, respectively.

Fig. 3. Pan-genomic analysis of the Streptomyces strains. The circular strip represents the presence/absence of gene clusters in each of the strains. PKSI: type I PKS; rpoB, gyrB, atpD, recA and trpB are housekeeping genes of Streptomyces. Bar plots at the end of each strip represents additional layers summarizing pan-genomic statistics of strains. ‘Num gene clusters’ represents the total number of gene clusters in strains, ‘Singleton gene clusters’ are orphan clusters which do not have homologs in other strains. Dendrogram is constructed using gene cluster presence/absence and Ward clustering. Tree over the layers is constructed based on Euclidean distance between the corresponding values for the strains using hierarchical clustering. Numbers corresponding to different layers are provided in detail in Table S6.

Fig. 4. Mining BGCs and resistance mechanisms (A) Enrichment of BGC classes in different Streptomyces strains. The size of the circle is proportional to number of BGCs harboured by the corresponding strain. The tree is constructed based on whole-genome phylogeny (bootstrap = 100). The clade overrepresenting model Streptomyces is coloured blue, while the strains in this study mostly fall under the green clade. More details are provided in Table S7. (B) Presence of known-resistance mechanisms and duplicated core genes inside BGCs of different classes as identified by ARTS analysis (v2.0). NRPS: Non ribosomal peptide synthase, RiPPs: Ribosomally synthesized and post-translationally modified peptides, PKSI: Type I PKS, PKSother: PKS other than type I PKS, PKS-NRP_hybrids: Hybrid PKS-NRPS, Others: Cluster that does not fit into any of categories shown.

Fig. 5. Biochemical network-guided metabolic pathway enrichment. The heatmap showcases enrichment of different KEGG pathway maps related to NP biosynthesis in the 8 strains, 4 prototypical Streptomyces and the widely used heterologous host, E. coli. The color scale corresponds to number of metabolites in each KEGG pathway, red being ‘high’ and blue corresponding to ‘low’ values. Labels ‘1’ and ‘2’ at the top represent two distinct clades showing significant differential enrichment in NP biosynthetic pathways (p-value = 1.2055e-04, chi-squared = 14.78, Kruskal-Wallis test). The ‘map’ IDs denote KEGG pathways. The dendrogram was generated using hierarchical clustering. Detailed numbers can be found in Table S16.

Cost-effective hybrid long-short read assembly delineates alternative GC-rich Streptomyces hosts for natural product discovery