The genomic landscape of ribosomal peptides containing thiazole and oxazole heterocycles

Abstract

Background: Ribosomally synthesized and post-translationally modified peptides (RiPPs) are a burgeoning class of natural products with diverse activity that share a similar origin and common features in their biosynthetic pathways. The precursor peptides of these natural products are ribosomally produced, upon which a combination of modification enzymes installs diverse functional groups. This genetically encoded peptide-based strategy allows for rapid diversification of these natural products by mutation in the precursor genes merged with unique combinations of modification enzymes. Thiazole/oxazole-modified microcins (TOMMs) are a class of RiPPs defined by the presence of heterocycles derived from cysteine, serine, and threonine residues in the precursor peptide. TOMMs encompass a number of different families, including but not limited to the linear azol(in)e-containing peptides (streptolysin S, microcin B17, and plantazolicin), cyanobactins, thiopeptides, and bottromycins. Although many TOMMs have been explored, the increased availability of genome sequences has illuminated several unexplored TOMM producers.

Methods: All YcaO domain-containing proteins (D protein) and the surrounding genomic regions were were obtained from the European Molecular Biology Laboratory (EMBL) and the European Bioinformatics Institute (EBI). MultiGeneBlast was used to group gene clusters contain a D protein. A number of techniques were used to identify TOMM biosynthetic gene clusters from the D protein containing gene clusters. Precursor peptides from these gene clusters were also identified. Both sequence similarity and phylogenetic analysis were used to classify the 20 diverse TOMM clusters identified.

Results: Given the remarkable structural and functional diversity displayed by known TOMMs, a comprehensive bioinformatic study to catalog and classify the entire RiPP class was undertaken. Here we report the bioinformatic characterization of nearly 1,500 TOMM gene clusters from genomes in the European Molecular Biology Laboratory (EMBL) and the European Bioinformatics Institute (EBI) sequence repository. Genome mining suggests a complex diversification of modification enzymes and precursor peptides to create more than 20 distinct families of TOMMs, nine of which have not heretofore been described. Many of the identified TOMM families have an abundance of diverse precursor peptide sequences as well as unfamiliar combinations of modification enzymes, signifying a potential wealth of novel natural products on known and unknown biosynthetic scaffolds. Phylogenetic analysis suggests a widespread distribution of TOMMs across multiple phyla; however, producers of similar TOMMs are generally found in the same phylum with few exceptions.

Conclusions: The comprehensive genome mining study described herein has uncovered a myriad of unique TOMM biosynthetic clusters and provides an atlas to guide future discovery efforts. These biosynthetic gene clusters are predicted to produce diverse final products, and the identification of additional combinations of modification enzymes could expand the potential of combinatorial natural product biosynthesis.

Fig. 1

Schematic of bioinformatics analysis. a TOMM biosynthesis begins with the ribosomal synthesis of a precursor peptide. The characteristic thiazoline/oxazoline heterocycles of a TOMM are installed by the C and D protein complex colored green and blue, respectively. Other tailoring enzymes (red and teal) often install additional modifications on the maturing product before the proteolytic cleavage (orange) of the leader peptide. b To identify TOMMs, all proteins containing a YcaO domain were identified using InterPro (IPR003776). The genomic regions surrounding the YcaO domains were retrieved, analyzed, and grouped by their cumulative BLAST bit score and synteny (Step 1). TOMM clusters were then separated from non-TOMM gene clusters determined by the inclusion of a C protein, precursor peptide, or bottromycin-like D protein (Step 2). (c) BLAST and synteny values from MultiGeneBlast were used to group TOMM clusters (Step 1). (d) A gene cluster was classified as a TOMM if it contained a C protein, precursor peptide, or was similar to bottromycin (Step 2)

Fig. 2

Sequence similarity network of TOMM D-proteins. Each node represents a unique D protein (YcaO, from InterPro family IPR003776), while an edge indicates that two proteins have a BLAST expectation value < 10⁻⁵⁴. All nodes belonging to TOMM families with at least one characterized gene cluster (structure of final product not necessary) are colored as noted in the legend. Black isofunctional groups indicate that no member of the group has been characterized

Fig. 3

Representative gene clusters from each TOMM subclass. Open reading frame diagrams are shown for a representative organism of each TOMM family. Uncharacterized gene clusters represent subclasses of TOMMs from which no gene clusters that have explored. Characterized clusters represent subclasses from which at least one gene cluster has been explored

Fig. 4

Sequence logos from bioinformatically identified TOMM precursor peptides. Sequence logos were created (WebLogo) using the C-terminal region of identified precursor peptides (cleavage sites were estimated based on length and the presence of a glycine or alanine residue as seen in other TOMM precursor peptides). Cys are labeled in red, Ser in blue, Thr in green, and other amino acids in black

Fig. 5

The prevalence and distribution of enzymes involved in TOMM biosynthesis. A sequence similarity network was constructed with all proteins in the TOMM biosynthetic gene clusters visualized at a BLAST expectation value of 10⁻³⁰. All proteins with 100 % identity were removed and are represented as larger nodes on the network (size is dependent on the number of redundant proteins). Groups are number for reference within the manuscript

Fig. 6

Phylogenetic analysis of TOMM producers. A maximum likelihood tree was constructed using 16S sequences from all organisms that contain a TOMM gene cluster. Coloring indicates which class of TOMM that particular organism contains, per the legend. The phyla of the producing organisms are labeled around the tree. Most classes of TOMMs appear to be produced within the same phylum; however, some classes are found in multiple phyla