Sequence-based classification of type II polyketide synthase biosynthetic gene clusters for antiSMASH

Rasmus Villebro,Kai Blin,Tilmann Weber,Simon Shaw

doi:10.1007/s10295-018-02131-9

Abstract

The software antiSMASH examines microbial genome data to identify and analyze biosynthetic gene clusters for a wide range of natural products. So far, type II polyketide synthase (PKS) gene clusters could only be identified, but no detailed predictions for type II PKS gene clusters could be provided. In this study, an antiSMASH module for analyzing type II PKS gene clusters has been developed. The module detects genes/proteins in the type II PKS gene cluster involved with polyketide biosynthesis and is able to make predictions about the aromatic polyketide product. Predictions include the putative starter unit, the number of malonyl elongations during polyketide biosynthesis, the putative class and the molecular weight of the product. Furthermore, putative cyclization patterns are predicted. The accuracy of the predictions generated with the new PKSII antiSMASH module was evaluated using a leave-one-out cross validation. The prediction module is available in antiSMASH version 5 at https://antismash.secondarymetabolites.org .

Highlights

Natural products are of huge interest due to their diverse bioactivities, and many are harnessed as medical drugs
For non-ribosomal peptide synthetases (NRPS) and modular type I polyketide synthases (PKS) pathways, which rely on the principle of an assembly-line biosynthesis, the domain composition of the PKS and NRPS megasynthase enzymes is used to predict putative PKS/NRPS precursors purely based on DNA/protein sequence information
This study extends antiSMASH to improve the predictive capabilities of the software concerning type II PKS gene clusters

Summary

Introduction

Natural products are of huge interest due to their diverse bioactivities, and many are harnessed as medical drugs. New secondary metabolites are continuously discovered in bacteria, fungi and plants and still constitute the most important drug discovery resource, especially for antibiotics [23]. This article is part of the Special Issue “Natural Product Discovery and Development in the Genomic Era 2019". The antibiotics and secondary metabolite analysis shell (antiSMASH) is one of the leading and most widely used secondary metabolite biosynthetic gene cluster (BGC) detection tools first introduced in 2011 and continuously developed and expanded since [2, 4, 20, 32]. With antiSMASH’s rule-based approach, the software can identify gene clusters responsible for the biosynthesis of 47 different classes of secondary metabolites including thiopeptides, lanthipeptides, lassopeptides, non-ribosomal peptide synthetases (NRPS), polyketide synthases (PKS) and many more. For NRPS and modular type I PKS pathways, which rely on the principle of an assembly-line biosynthesis, the domain composition of the PKS and NRPS megasynthase enzymes is used to predict putative PKS/NRPS precursors purely based on DNA/protein sequence information

Methods

Results

Conclusion