Computational Methods for Identification of Novel Secondary Metabolite Biosynthetic Pathways by Genome Analysis

Introduction

Polyketides and non-ribosomal peptides constitute the largest family of small molecule natural products biosynthesized by microbes, fungi and plants as secondary metabolites (Linne et al., 2003; Schwarzer et al., 2003; Shen, 2003). These small molecule natural products not only show enormous diversity in their chemical structure, they also have a variety of biomedical and pharmaceutical applications in view of their therapeutic potentials. The elucidation of polyketide and nonribosomal peptide biosynthetic machinery by pioneering genetic and biochemical studies have revealed that these secondary metabolites are biosynthesized by multi functional megasynthases like Polyketide Synthases (PKSs) and nonribosomal peptide synthetases (NRPSs), using an assembly line mechanism which resembles fatty acid biosynthesis. The availability of complete genome sequences in an increasing number of organisms has opened up the possibility of discovering novel secondary metabolite natural products by genome mining (Van Lanen & Shen, 2006). Major advances in biosynthetic engineering during the last decade have also demonstrated the feasibility of obtaining novel engineered natural products by rational manipulation of known secondary metabolite biosynthetic pathways using biosynthetic engineering approaches (Baltz, 2006; Zhang & Wilkinson, 2007). Hence, during the last decade, the research on PKS and NRPS biosynthetic pathways in various organisms has been pursued with two major goals, namely, identification and experimental characterization of new secondary metabolites in various microbial and fungal species and production of novel rationally designed natural products by manipulation of known PKS/NRPS biosynthetic machinery using a biosynthetic engineering approach.

The remarkable conservation of secondary metabolite gene clusters across organisms has offered abundant scope for obtaining novel insights into the secondary metabolite biosynthetic code by computational analysis. Hence, development of computational methods for relating the chemical structure of the complex secondary metabolites to the amino acid sequence of their corresponding biosynthetic proteins has been an area of active research. Such computational methods (Minowa et al., 2007; Yadav et al., 2003a) have played a major role in guiding various experimental approaches involving genetics, biochemistry, proteomics and metabolomics for discovery of new secondary metabolites by genome mining and reprogramming of known biosynthetic pathways for producing novel natural products by a rational design approach.

In this article, we attempt to give a brief overview of various computational methods which have facilitated easy correlation of the chemical structure of the secondary metabolites to the amino acid sequence of the various PKS and NRPS megasynthases present in the corresponding biosynthetic gene clusters. We first provide background information on various different polyketide and non-ribosomal peptide biosynthetic paradigms. This is followed by a description of different types of computational studies that have been carried out on secondary metabolite biosynthetic clusters and the rationale behind it. The sections that follow, give a detailed account of the computational techniques, which can help in predicting the organization of various PKS/NRPS catalytic domains in a secondary metabolite gene cluster as well as their substrate specificities. These sections also mention the various software or web servers available for such analysis. In the subsequent section, we describe the computational methods for predicting the order of substrate channeling in a secondary metabolite biosynthetic cluster based on analysis of inter subunit interactions. The last section describes a few examples where bioinformatics analyses have guided experimental studies to discover new metabolites, as well as studies to generate novel metabolites by reprogramming the known biosynthetic machinery.