Structural aspects of non-ribosomal peptide biosynthesis

Small peptides have powerful biological activities ranging from antibiotic to immune suppression. These peptides are synthesized by non-ribosomal peptide synthetases (NRPS). Structural understanding of NRPS took a huge leap forward in 2002; this information has led to several detailed biochemical studies and further structural studies. NRPS are complex molecular machines composed of multiple modules and each module contains several autonomously folded catalytic domains. Structural studies have largely focused on individual domains, isolated from the context of the multienzyme. Biochemical studies have looked at individual domains, isolated whole modules and intact NRPS, and the combined data begin to allow us to visualize the process of peptide assembly by NRPS.

Introduction

Amino acids are structurally diverse and reactive molecules. The proteinogenic α-amino acids are limited to only 22 different structures. To increase the diversity and utility of these molecules, Nature has polymerised them via a peptide bond. The number of possible molecules for a peptide of length n amino acids is 22ⁿ. Large polypeptides are made using the ribosome and the post-translation machinery that are familiar to anyone with a rudimentary knowledge of biology. The vast majority of peptide bond formation is catalysed by ribosomes and, until recently, the catalysis of peptide bond formation by non-ribosomal peptide synthetases (NRPS) has been largely overlooked. A list of molecules made by NRPS is beyond the scope of this article and the reader is referred to a recent review for a comprehensive account of these molecules [1]. However, it is instructive to highlight some of the best-known examples to illustrate the importance of NRPS systems. The last-resort antibiotic vancomycin and its analogues (Figure 1) have exquisitely complex structures made by NRPS and associated enzymes [2]. Indeed, almost all peptide-based antibiotics are made by NRPS. Chelation of iron by bacteria is vital for their survival and is often a virulence determinant in pathogens. NRPS synthesize macrocycles such as enterobactin (Figure 1), which have an extraordinarily high iron affinity [3]. Cyclosporin (Figure 1), an immune suppressor that made organ transplants possible, and the potent anti-tumour compound bleomycin (Figure 1) are both made by NRPS 4., 5..

A cursory examination of the structures in Figure 1 highlights the contrast between familiar ribosomal peptides and non-ribosomal peptides. The molecules made by NRPS are often cyclic, have a high density of non-proteinogenic amino acids and often contain amino acids connected by bonds other than peptide or disulfide bonds. Given the complexity of the molecules, one might assume that each was made by a specific pathway and that very little in the way of general rules could be established. NRPS are now known to be very large proteins and, despite the obvious complexity of the products, consist of a series of repeating enzymes fused together. Such fusion of repeating enzymes in a single polypeptide closely parallels the protein machines responsible for polyketide biosynthesis [6]. For NRPS, one amino acid building block is normally incorporated into the peptide product by one module, thus products with ten amino acids would be expected to be constructed by an NRPS with ten modules stitched together. This is often termed the colinearity rule. Each module is normally specific to a particular amino acid substrate. It is important to note that there are several exceptions to the colinearity rule, particularly for NRPS that assemble siderophores 3., 7., 8.. The normally colinear sequence of modules in both NRPS and the peptide product offers the potential for great versatility in the bioengineering of these systems.

This review discusses the structures and mechanisms of action of the isolated components of the NRPS. We highlight new biochemical data that improve our understanding of these molecular machines.