Structural and Mechanistic Investigations of Antibiotic Production in Bacteria

Lead Research Organisation: University of Bristol
Department Name: Chemistry


There is an indisputable problem of antibiotic resistance and an urgent need for the discovery and development of new antibiotics that are cost effective to produce. In February 2017, the World Health Organisation (WHO) published its first ever list of antibiotic resistant priority pathogens that pose serious risks to human health. This list includes high priority targets such as the well-publicised methicillin resistant Staphylococcus aureus (MRSA) and there is clear recognition that urgent action and research into new antibiotics is required. However against this alarming headline, a drop in investment in research and development into new antibiotics has led to a dramatic fall in the number of new drugs being discovered and a reduction in knowledge and expertise that is capable of delivering them.

Natural products and their derivatives have, and will continue to be, an important source of these antibiotics that are critical for human and animal health. Polyketides are a family of natural products which include high value compounds including antibiotics. They are derived from a wide range of sources including bacteria and fungi. Over the last fifty years some understanding of the remarkably complex and diverse ways in which antibiotics are synthesised in Nature has been gained. Whilst bacteria and fungi might be viewed as simple organisms, they arguably outperform the world's best synthetic chemists in terms of their elegance and efficiency and it is this power we wish to harness. By fully understanding nature's biosynthetic machinery we can engineer pathways to deliver new bioactive compounds.

It turns out that many antibiotics are made by a series of chemical reactions catalysed by megaprotein assemblies that act as nano-scale factories inside the microbe. Simple organic molecules are activated and loaded at one end, joined together and then released as completed (usually elaborate) products at the other end. The nano-factories join the simple building blocks on an assembly line of individual modules, akin to a group of robots performing operations in vehicle manufacture. The chemical structure of each antibiotic is thus determined by the enzymes present at each stage of the assembly line, the blueprint being the biosynthetic gene cluster. We understand some rules for building these factories and can rearrange the order of modules to produce new compounds, but sometimes this just breaks the assembly line, or produces an unexpected compound. This reveals we don't truly understand how they have evolved to fit together and we do not know all of the chemical steps required.

Our aim is to investigate two different "factories" that produce anti-MRSA antibiotics. The first produces mupirocin, which is used commercially but is restricted to topical use and in nasal sprays because of its instability. The second is thiomarinol, structurally related to mupirocin, but has features that might lead to wider applications. A substantial number of important biosynthetic steps are not fully understood in both of these systems. Combining the expertise of chemists, biochemists, structural biologists and molecular modellers we will elucidate these steps which could lead to a more stable version of mupirocin. We will also use these systems to answer quite general questions so we can build new pathways to novel compounds in a rational way.

Technical Summary

Bacterial Type 1 trans-AT polyketide synthases (T1 PKS) are well-suited for engineering to produce new compounds by virtue of their non-linear processing and chemical diversity via the recruitment of many different in-trans acting enzymic components. We will use in vitro studies of the mupirocin (mup) (a mixture of pseudomonic acids (PA-A etc.)) and thiomarinol (tml) PKSs to reveal the nature of key biosynthetic transformations that have, to date, remained obscure but we believe will enable rational engineering of the pathways.

Using in vivo studies we will investigate the timing and mechanism of the following key biosynthetic steps:

6-Hydroxylation believed to be catalysed in trans by MupA, a putative flavin mononucleotide (FMNH2) oxygenase in combination with a non-elongating junction between two Type I PKS modules (mmpD and mmpA).

Ring formation in mupirocin and thiomarinol involving the Rieske oxidoreductases MupW and MupT and TmlW/TmlT respectively. This selective transformation is intriguing and the substrate scope and enzyme mechanism will be investigated using a library of synthetic analogues to investigate the potential of developing new biocatalysts of wide value.

Formation of the fatty acid side-chains and timing of attachment to the polyketide fragment via an ester linkage. Unusual starter units are involved and their biosynthesis will be elucidated using purified proteins (MupS, MupQ and MacpD and the tml equivalents). Fatty acid chain-extension (mmpF) and esterification (MupB and MupL) will be investigated using synthetic substrates and purified ACPs.

Mupirocin is a commercial antibiotic but the major component, PA-A, is unstable rearranging to inactive products due to the 10,11-epoxide. We aim to engineer the mup and tml biosynthetic pathways to direct production to generate the more stable PA-C (with a 10,11-alkene) as the major product by systematic investigation of which steps in the mup pathway constrain the production of PA-C

Planned Impact

Industrialists and the commercial sector.
Two NIH workers David Newman and Gordon M Cragg recently (J. Nat. Prod. 2016, 79, 629-661) reported on the enormous contribution of natural products as sources of new drugs spanning the last 34 years up to 2014. Of these, polyketides are worth billions of pounds to the World economy each year and our work will contribute to development of this field. Our primary goal is to use an interdisciplinary approach to understand the fundamental steps in polyketide assembly using the biosynthesis of mupirocin and thiomarinol as model systems, with the added advantage that mupirocin is a commercialised antibiotic. Since mupirocin and numerous other natural products are produced commercially by fermentation, the knowledge gained could enable engineering of these pathways and the biotechnology directly applied to manufacturing. The greatest impact is likely to be between the academic research community and those involved in commercial R&D and the applicants have strong links with commercial partners in this arena, for example GSK (see letter of support) and Syngenta. We have IP agreements in place, and there are defined routes to translate the outcomes from this study into saleable products, tools and technologies.

Sixty eight natural product or natural product derived anti-bacterial agents were approved by the FDA between 1981-99, but from 2000 to 2014 this number fell to just 14 as industrial sectors dropped investment in this area. The subsequent loss of knowledge and expertise in this field needs to be addressed urgently as natural products will continue to play a vital role in human and animal health. We must train a new generation of scientists in the analytical, chemical and biological skills associated with this field to address to challenges of the next decades. This project will provide high-level training for two PDRAs as part of a highly interdisciplinary team and implicitly individuals involved in the project will receive an internationally competitive training in experimental and theoretical methods at the chemistry/biology interface. The PDRAs will also benefit from presenting their findings at national and international conferences and from engagement in synthetic biology initiatives at Bristol (BrisSynBio), the new Bristol BioDesign Institute and the joint Bristol/Oxford/Warwick EPSRC/BBSRC synthetic biology CDT.

The general public through knowledge sharing and outreach activities.
There are potential significant impacts of this research program via knowledge sharing with the public. In light of this we will actively participate in public engagement and outreach activities during the tenure of the grant. As detailed in the Pathways to Impact we will to create a number of outreach workshops for school children as part of the world leading Bristol ChemLabs program to promote understanding of antibiotics and antibiotic resistance.

The general public through improvement in human health.
One of the aims of this research program is to develop basic science that underpins our ability to manipulate biosynthetic pathways and metabolite flux to generate novel molecules with antibacterial activities. Producing such molecules will have significant impact on public health, prolonging and improving life, for example, by combating antibiotic resistant "superbugs". We will achieve this impact by (i) demonstrating the utility of new antimicrobial compounds that emerge during the study and we have engaged with Public Health England and Dr Jim Spencer for this purpose (see letters of support) (ii) protecting resulting intellectual property, and (iii) engaging with our industrial partners to translate outcomes out of the laboratory to the clinic.
Title Chemical Shift Files for mupirocin mAcpD 
Description BMRB code 34451: NMR chemical shifts associated with PDB code 6TG5 and deposited in BioMagResBank 
Type Of Material Database/Collection of data 
Year Produced 2020 
Provided To Others? Yes  
Impact This work contributed to a recent publication (ACS Chem. Biol. 2020,, A Priming Cassette Generates Hydroxylated Acyl Starter Units in Mupirocin and Thiomarinol Biosynthesis). Understanding starter unit selection has widespread application in bioengineering, synthetic biology and natural product research for increasing the chemical diversity of engineered high-value compounds of commercial and therapeutic value. 
Title Protein PDB coordinates for MupZ 
Description PDB code 6FXD: Crystal structure coordinates of MupZ from the mupirocin biosynthetic gene cluster from Pseudomonas Fluoroscens. 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
Impact The discovery that MupZ can direct an anti-Baldwin ring closure, and is promiscuous, has led to further investigations into its application as a biocatalyst. This relates to follow on funding studying Diels-Alderase enzymes and work with Industrial Partners AstraZeneca who have invested over £100K to begin to develop the application of these enzymes further in industrial processes. 
Title Protein PDB coordinates for mAcpD 
Description PDB code 6TG5: NMR solution structure of mAcpD from the mupirocin biosynthetic gene cluster from Pseudomonas Fluoroscens. 
Type Of Material Database/Collection of data 
Year Produced 2020 
Provided To Others? Yes  
Impact This ACP model is the first example of an extended structure encompassing additional structural elements that may have important applications in pathway engineering. Whereas this is too early to predict at present we will be able to update this record in due course.