Understanding biosynthetic protein–protein interactions

Investigations into natural product biosynthesis continue to yield a wealth of novel chemical structures and biological functions. The central role of natural products in medicine is unquestioned – for example, over half of new anticancer drug approvals (excluding vaccines) in 1981–2016 were for natural products or derivatives thereof, and this proportion was even higher for antibacterials.¹ Natural product biosynthesis also provides access to high value compounds of significance to agriculture, animal health, and the flavour and fragrance industries. An explosion in the development and implementation of new synthetic biology tools to enhance titres or engineer desirable analogues has made a comprehensive understanding of key biosynthetic processes more important than ever. Whilst there has been an exponential increase in the discovery of biosynthetic gene clusters, resulting from rapid advances in DNA sequencing technology, a substantial majority of these are cryptic, and development of generally applicable methods for accessing the metabolic products of such clusters remains a key challenge. An alternative approach for accessing structural novelty exploits the inherent diversity in the biosynthetic assembly lines that produce many important natural product classes. The enzymatic logic of these assembly lines can often be explained in terms of basic principles, and offers considerable potential for exploitation to access new chemical entities. However, it is clear from the challenges of biosynthetic engineering that our understanding of the intricacies of these molecular machines is often deficient. This two-part themed issue within Natural Product Reports aims to summarise the current state of the art and our current understanding of the protein–protein interactions (PPIs) at the core of several key biosynthetic systems.

Within this themed issue, emphasis has been placed on the three major groups of megasynth(et)ases: fatty acid synthases (FASs), polyketide synthases (PKSs) and non-ribosomal peptide synthetases (NRPSs). These giant multi-domain proteins employ a complex molecular choreography that is enabled by carrier proteins that tether and present the various intermediates of biosynthesis to a variety of catalytic domains responsible for chain assembly. The reliance on substrate tethering to carrier proteins makes understanding the protein–protein interactions critical to interpreting the choreography of such megasynth(et)ases. The development of chemical tethering approaches to capture and study key states has proven invaluable for this. Gulick and Aldrich summarise the design and implementation of such approaches (DOI: 10.1039/c8np00044a), demonstrating how they have provided access to complexes and transient intermediates that otherwise would not have been observable.

The reliance on carrier proteins in these systems is further complicated by the diversity of function within each class of megasynth(et)ase. A key example of this is the evolution of two distinct classes of modular PKSs. In cis-AT PKSs the acyltransferase (AT) domains responsible for loading various (alkyl)malonyl extender units onto carrier protein domains are incorporated into each module of the assembly line, whereas in trans-AT PKSs a single trans-acting AT typically supplies malonyl extender units to all of the carrier protein domains. Insights into PPIs in the cis-AT PKS class are provided by Smith et al. (DOI: 10.1039/c8np00058a), with Challis et al. detailing the current understanding of interactions in trans-AT systems (DOI: 10.1039/c8np00066b).

In addition to modular PKSs, iterative systems are known that use a single carrier protein and set of enzymatic domains to carry out multiple rounds of chain extension and modification. Iterative PKSs can be classed as either type I (where the carrier protein and catalytic domains are incorporated into a single multienzyme) or type II (a standalone carrier protein and several distinct catalytic subunits). In the analogous FAS systems, which also catalyse iterative chain assembly, the same division into type I and type II architectures is observed. The PPIs central to the function of type I PKSs and FASs are discussed by Herbst, Townsend and Maier (DOI: 10.1039/c8np00039e), while Burkart et al. focus on the type II systems (DOI: 10.1039/c8np00040a). The remarkable complexity of PKS assembly lines is reflected by the structural complexity of their metabolic products, which often makes chemical synthesis of these molecules on commercially-relevant scales challenging. Engineering of such assembly lines to produce novel polyketide derivatives in high titres is thus a high priority for the field. Current approaches for this are evaluated by Klaus and Grininger (DOI: 10.1039/c8np00030a).

In addition to the various different types of FAS and PKS assembly lines, NRPSs also employ carrier protein-based enzymatic logic. In contrast to PKSs and FASs, which possess several covalent attachment sites (such as the carrier protein, AT active site and ketosynthase active site) for biosynthetic intermediates, NRPS carrier proteins act as the sole site for covalent attachment of intermediates throughout the peptide chain assembly. The oligomerisation states of these systems also differ, with type I PKS and FAS subunits typically forming homodimers, whereas NRPS subunits are believed to function as monomers. Concatenating these systems would therefore appear to be difficult, but in fact numerous hybrid PKS–NRPS systems are known. The state of understanding of hybrid assembly lines and the PPIs underpinning their function is discussed by Eguchi et al. (DOI: 10.1039/c8np00022k). Shifting the focus to pure NRPS systems, the central role played by the carrier protein in mediating PPIs is discussed by Izoré and Cryle (DOI: 10.1039/c8np00038g), while approaches to NRPS engineering are discussed by Ackerley et al. (DOI: 10.1039/c8np00036k), who emphasise the importance of maintaining key PPI interfaces for the creation of productive engineered assembly lines.

Expanding the scope beyond the megasynth(et)ase paradigm, Laursen et al. focus on the specialised metabolism of plants, whose ability to generate extreme phytochemical diversity not only aids their survival, but offers additional possibilities for synthetic biology approaches to exploit the underlying biosynthetic machineries (DOI: 10.1039/c8np00037a).

Taken together, the articles in this two-part themed issue serve to highlight how the fascinating and diverse chemical structures produced by specialised metabolism are underpinned by a complex and essential network of PPIs. Our ability to interrogate and understand these interactions will continue to facilitate attempts to control and re-task biosynthetic systems to produce novel molecules, with the potential to help overcome some of the significant challenges currently facing society. As such, we anticipate that detailed analysis of the PPIs in natural product biosynthesis will continue to provide fertile ground for research and discovery for many years to come.

References

1. D. J. Newman and G. M. Cragg, J. Nat. Prod., 2016, 79, 629–661.

---