Synthetic biology


(Author for correspondence: tel +44 1603 450407; email

4th New Phytologist Workshop, Bristol, UK, June 2012


Synthetic biology has been variously defined as:

  • Synthetic biology aims to use modular, well-characterised biological parts to predictably construct novel genetic devices and complex cell-based systems following engineering principles.

(Endy, 2005)

  • Synthetic biology is the design and engineering of biologically based parts, novel devices and systems as well as the redesign of existing, natural biological systems. It has the potential to deliver important new applications and improve existing industrial processes – resulting in economic growth and job creation.

(A synthetic biology roadmap for the UK, 2012,

  • Synthetic biology is the engineering of biology: the synthesis of complex, biologically based (or inspired) systems, which display functions that do not exist in nature. This engineering perspective may be applied at all levels of the hierarchy of biological structures – from individual molecules to whole cells, tissues and organisms. In essence, synthetic biology will enable the design of ‘biological systems’ in a rational and systematic way.

(Synthetic Biology: Applying Engineering to Biology: report of a NEST High Level Expert Group. Molecular Systems Biology (2007) 3: 158.)

There is no agreed definition of synthetic biology, but it is best understood as the rational design of biological systems and living organisms using engineering principles. The concept of ‘synthetic biology space’ (Channon et al., 2008) provides a useful tool that enables the sometimes seemingly disparate components, hierarchies and approaches encompassed by synthetic biology to be placed into a common framework.

Synthetic biology presents a big challenge and as such was the impetus for convening the 4th New Phytologist Workshop, highlights from which are discussed and presented here (Bristol, UK; Challenges and questions include can we deconstruct biology and modularize it – standardize it into parts? Can such parts be designed, engineered and combined in new ways to achieve novel functions? Major advances have been achieved – for example, the construction of a synthetic biosynthetic pathway for the synthesis of artemisinic acid, a precursor for the anti-malarial drug artemisin, in microbes (Ro et al., 2006; Westfall et al., 2012), or indeed the synthesis of a novel, functional microbial genome de novo (Gibson et al., 2010). Yet biological systems are complex, and while we should continue this empirical, ambitious and in some cases, as Dek Woolfson (University of Bristol, UK) said in his keynote talk, even fanciful approach to the design and engineering of biology, the biggest advances may come from understanding how biological systems function – that is, what goes wrong when we try to construct them, and why? Nobel Laureate and physicist Richard P. Feynman wrote, ‘What I cannot create, I do not understand.’ Synthetic biology represents a response to that problem.

‘A major challenge of synthetic biology is that real biology is messy and difficult to deconstruct and catalogue.’

Synthetic biology is much more than a platform – an inventory of parts, modules and circuits that can be slotted into chasses and used. It is a bold and new way of thinking that helps us to understand biological processes and systems. It encourages scientists from across the disciplines (social scientists, biologists, chemists, engineers, mathematical modellers and others) to work together to identify the grand challenges faced by society and collectively to find solutions (Fig. 1). To do this effectively it is essential that there is meaningful and productive engagement between scientists and the wider public.

Figure 1.

Network map showing the relationships between 527 clusters of authors with publications on synthetic biology in Web of Science. Node size is based on the number of publications for a given author. Reproduced with permission from Oldham et al. (2012).

Science has tended to fragment into ever more specialized areas of research. The branches of a tree may divide repeatedly to form the finest twigs – twigs that may snap off in the wind. This is the danger in the silo mentality that tends to prevail in science (Osbourn, 2006). Synthetic biology has great things to offer – parts, circuits, deliverables, drugs for industry and the developing world. Perhaps the greatest opportunity, however, is in bringing people together from different disciplines and communities; in doing things differently; in doing what it takes to make things happen; in remembering what made us do science at school and go into it as a profession – because it is exciting; because we have the privilege of being able to find out; because it connects us with the world around us; because we can make a difference.

The aim of this New Phytologist workshop was to bring together a wide range of scientists, from diverse backgrounds, to share and discuss current ideas and progress in their respective fields. This is a departure from the typical New Phytologist workshop, in the sense that it involved chemists, microbial biologists, plant scientists, mathematical modellers and social scientists, all with a common interest in synthetic biology. Most who attended the meeting had not seen all or even many of the other speakers previously, reflecting the eclectic mix.

The meeting began with a presentation by Professor Douglas Kell, Chief Executive of the UK's Biotechnology and Biological Sciences Research Council (BBSRC), which funds nonclinical biological sciences. Synthetic biology forms part of BBSRC's strategic priority area in Industrial Biotechnology and Bioenergy, and BBSRC together with the Engineering and Physical Sciences Research Council (EPSRC) have collectively put > £62 million of funding into this area to date. The BBSRC and the EPSRC aim to ensure that the UK is able to remain at the forefront internationally as the field develops. Advancing the field requires engineers and biologists to work with social scientists, mathematicians, computer scientists, chemists and life scientists to address the technical, social, ethical and economic challenges in a holistic manner. The UK Government's Department of Business Innovation and Skills (which supports both the BBSRC and EPSRC) has recently launched its new Synthetic Biology Roadmap ( This roadmap lays out the need for future funding in this area, and highlights its importance for Research Councils UK (RCUK), the UK Government and the UK economy. Implicit in this is the need for new ways of working required by scientists to achieve progress, and specifically in establishing multidisciplinary programmes. The New Phytologist workshop reflects this new approach, in which biologists, from different backgrounds, will increasingly develop joint programmes of work with engineers, physical scientists and mathematicians.

Engineering principles and applications

From an engineering perspective, synthetic biology insists on standardized parts (e.g. genes, proteins, circuits) that can be assembled using bioinformatics and simulation tools to build functionality. The Massachusetts Institute of Technology (MIT) ‘Registry of Standard Biological Parts’ ( contains an impressive inventory of components (Baker et al., 2006). However many of the components that have been deposited in repositories have not undergone the level of quantitative characterization that an engineer would require in order to design a system with predictive behaviour. A major challenge of synthetic biology is that real biology is messy and difficult to deconstruct and catalogue. Professor Dek Woolfson provided a keynote lecture, focusing on protein design and engineering, illustrating how synthetic biology draws upon more than simply parts, but biological design principles to create new parts and new systems. His laboratory is generating a toolkit of short (c. 30 amino acids) peptide-encoding DNA sequences for the self-assembly of novel proteins with new properties. This is based on a bioinformatics approach that distilled information on structure–function relationships from databases to identify an alphabet of helix–helix interaction motifs. Using rational computational design, the Woolfson group has successfully tested these relationships experimentally, using biophysical methods to make new materials from peptide self-assembly (Zaccai et al., 2011). Numerous applications include new molecular switches, sensors or nanoscale objects. A key aim at present is to understand the rules that link sequence with shape and function (Fletcher et al., 2012) .

Dick Kitney (Centre for Innovation and Synthetic Biology, Imperial College, London, UK) explained how the principles of modularity, characterization and standardization can be applied to overcome this complexity and transform basic science into applications using synthetic biology approaches. An ultimate aim of ‘extreme’ engineering-led synthetic biology is to deskill the design process and apply automation through the use of BioCAD tools and laboratory robots for assembly, characterization, testing and validation. The overall aim of these developments is to standardize procedures and processes so that they can be implemented at multiple locations. Rationally combining well-characterized parts and devices has enabled the development of complex systems in bacteria such as co-ordinated oscillation, pattern formation, edge detection and genetic logic gates (Elowitz & Leibler, 2000; Basu et al., 2005; Tabor et al., 2009; Wang et al., 2011). Such approaches are less well advanced for eukaryotes.

Rainer Breitling (University of Glasgow, UK) asked whether complex biology really can be simplified in engineering terms, commenting that there is a tendency to focus on design but not on how to run the system – where synthetic biology runs into systems biology. The analogy that he used was that of steam train workers who have learned to recognize, diagnose and fix the odd rattles and noises that the engines make when they are not running optimally. The moral is that there is a need to focus on design but be responsive to the need to ‘debug’ the system when things do not go as expected. Successful debugging will require modelling. Breitling used the highly complex area of metabolomics as his example. Major challenges when considering the manipulation of metabolism include the identification of novel metabolites, detection of side products and toxic intermediates, elucidation of global metabolic reorganization and identification of switches that regulate these processes, location of bottlenecks, and quantification of end products. The development of computational tools that can be used to identify all possible metabolic pathways is now allowing genome-scale metabolic modelling (Medema et al., 2011b). This will ultimately enable the synthetic design of biochemical pathways through integration of components into smartly regulated transcriptional units. Through engineering efforts, we are finding the constraints that systems and components operate under – this will undoubtedly provide new insights into the challenges faced by biology and the limits of how we can exploit this for novel designs.

Other empirical tests of the capability of synthetic biology involve the synthesis of multicomponent assemblages with life-like properties, and even artificial cells from abiotic components. These efforts draw inspiration from biology and the spontaneous formation of membranes to guide strategies to create novel supramolecular architectures from self-assembly of simple molecular building blocks (Aida et al., 2012). Interest in this area is not new to synthetic biology and goes back to seminal endeavours to understand the origin of life in the 1950s, when considerable attention was focused on trying to encourage components of simulated ‘prebiotic soup’ to form molecular assemblies and vesicles (Deamer, 2008). Sam Stupp (Northwestern University, Evanston, IL, USA) presented innovative strategies using supramolecular self-assembly to create peptide-based nanoscale filaments with chemical structures that mimicked aspects of the signalling machinery involved in tissue growth (Fig. 2a). These strategies have enormous potential for the formation of human tissues and organs in regenerative medicine and for the development of cell-like microscale objects that can be targeted for therapeutic purposes (e.g. artery repair, drug delivery). Cameron Alexander (University of Nottingham, UK) focused on a project to develop a chemical cell (‘Chell’). The goal of this project is to generate complexity in function from abiotic parts, each of which has a specific function that can be related to a real biological cell. The aim is to couple chemical information, a simple metabolism model, and a container to gate information flow into a ‘Turing test’ experiment that will enable the properties of the Chell (information flow and signalling) to be evaluated against real bacterial cells to establish ‘living’ artificial chemical systems (Cronin et al., 2006; Xue et al., 2011).

Figure 2.

Some examples of synthetic biology outputs: (a) peptide amphiphiles self-assemble into nanofibres (image courtesy of Mark Seniw and the Stupp Laboratory, Northwestern University, Evanston, IL, USA); (b) robot (image courtesy of Joseph Ayers, Northeastern University Marine Science Centre, Nahant, MA, USA); (c) structure of the antimalaria drug, artemisinin; (d) cryo-electron microscope reconstruction of a synthetic nanoshell produced by transiently co-expressing the coat protein precursor and 24 K proteinase of the cowpea mosaic virus (CPMV) in leaves of Nicotiana benthamiana (image courtesy of Kyle Dent and Neil Ranson, University of Leeds, UK).

One of the highest profile advances in the field of synthetic biology so far has been the generation of the first synthetic genome derived from the bacterium Mycoplasma by Craig Venter and co-workers. This artificial genome was assembled in yeast, implanted into Mycoplasma, re-booted and shown to be functional (Gibson et al., 2010). However we are still a long way away from designer genome engineering. As Tom Ellis (Imperial College, London, UK) pointed out, much of synthetic biology so far has been concerned with writing applications rather than complete operating systems. The construction of operating systems is limited by the paucity of well-characterized parts, and the parts that are available differ in multiple parameters and so are not ideal for large-scale rational design. For example, one of the limitations in the design of synthetic gene networks is the lack of suitable promoter sets. Tom presented a systematic analysis of promoters in yeast, including rational diversification of a promoter that passes multiple tests for ‘constitutiveness’ and the demonstration that this promoter can be subject to fine-tuned orthogonal regulation using Transcription Activator-Like Orthogonal Repressors (TALORS; Blount et al., 2012). An important aspect of genetic engineering in yeast is that gene expression and regulation can be dramatically altered depending on whether the gene/promoter in question is located on a plasmid or within the genome. Factors such as this highlight the context-dependent properties of biological parts, which will be critical in understanding their behaviour and using them for genome writing.

Joe Ayers (Marine Science Centre, Northeastern University, Nahant, MA, USA) moved from promoters and genes to much larger components and circuits. The segmental nature of invertebrates, and the ‘simple’ central nervous system wiring makes these creatures particularly amenable to simulation using robotics (Fig. 2b). Joe gave examples of invertebrate reflexes that allow reactive autonomy in response to gravity, obstacles and hydrodynamic and optical flow and showed how, by measuring rhythm across neuronal circuits, he has been able to develop lobster- and lamprey-based robots that operate under the control of central pattern generators (neural networks) rather than algorithms (Ayers et al., 2010). In a separate project he is developing an electronic nervous system to control the flight of ‘RoboBees’. Potential applications of robotic insects include crop pollination, search and rescue, exploration of hazardous environments, military surveillance, high-resolution weather and climate mapping and traffic monitoring.

Synthetic biology in microbes

Baker's yeast (Saccharomyces cerevisiae) has a genome size of 14 Mbp and contains around 6000 genes. Recent advances in DNA synthesis technology have enabled the synthesis of synthetic yeast chromosome arms (Dymond et al., 2011). Joel Bader (John Hopkins University School of Medicine, Baltimore, MD, USA) presented an ambitious international collaborative project that is in progress to make the first complete synthetic yeast genome, Sc2.0, is being designed by teams of undergraduates around the world according to an arbitrary set of design principles (; Cooper et al., 2012). The synthetic DNA has been designed to eliminate transposons and other potentially detrimental elements. Sc2.0 incorporates an inducible evolution system, synthetic chromosome rearrangement and modification by loxP-mediated evolution (SCRaMbLE) that will allow wide-scale genome restructuring under controlled conditions, so generating complex genotypes and a broad range of phenotypes. This ‘bottom up’ approach, coupled with large-scale analysis of recombination in naturally occurring strains through genome sequencing (a ‘top down’ approach), is likely to provide unprecedented insights into mechanisms of genome reorganization and may open the door to a new field – that of synthetic combinatorial genetics.

The need for technologies that enable the construction of novel and complex functions in biological systems is a major grand challenge in the field of synthetic biology since this will enable the rapid evolution of synthetic multi-gene traits. Susan Rosser (University of Glasgow, UK) showcased a synthetic system that she has developed, inspired by natural bacterial recombination systems known as integrons. Integrons are natural cloning and expression systems that assemble multiple open reading frames, in the form of gene cassettes, by using site-specific recombination and conversion to functional genes by expression from an internal promoter (Cambray et al., 2010). Dr Rosser demonstrated optimization of a recombinase-based platform using assembly of the Erwinia carotenoid pathway as a proof of concept. This approach has the potential to generate a robust technology enabling the engineering of multi-step processes for a variety of useful purposes. Notable examples include the production of renewable bio-fuels and biomaterials, the synthesis of small biomolecules for applications in specialty chemicals and crop protection, and bioremediation.

One of the best-known examples of a high-value chemical from plants for which there is urgent demand is artemisinin, a natural product produced by sweet wormwood (Artemisia annua) that is used to combat malaria (Fig. 2c). Escherichia coli and Saccharomyces cerevisiae have both been engineered to produce the artemisinin precursor amorpha-4,11-diene, and yeast has been further engineered to produce artemisinic acid, the feedstock for chemical conversion to artemisinin (Ro et al., 2006). Chris Paddon (Amyris Inc, Emeryville, CA, USA) highlighted the challenges encountered during metabolic engineering of yeast, including the need to up-regulate the mevalonate pathway, which is required for the generation of precursors. Further yield improvement to economically viable production levels was achieved using an automated approach to combinatorial optimization problems (Westfall et al., 2012).

The pressing need to discover new antibiotics in the modern era can be approached starting with computer-based design strategies that integrate genomic diversity with the proper gene regulatory circuitry. Engineering and optimization of novel metabolic pathways requires in the first instance identification of the genes for the pathway components. Eriko Takano (University of Groningen, the Netherlands) presented a variety of computational tools for the discovery and exploitation of antibiotics and other secondary metabolites from Streptomyces bacteria with the ultimate aim of developing ‘plug-and-play’ systems for synthesis of microbial natural products (Medema et al., 2011a,b, 2012).

Synthetic biology in plants

Agriculture must become more efficient in the very near future in order to be able to support the global population as it grows by 40% to beyond nine billion by 2050. Ray Elliot (Syngenta, Jealott's Hill, UK) introduced the concept of ‘sustainable intensification’ to emphasize that modern biotechnology and crop protection chemicals will have a critical and essential contribution in ensuring food security as well as in protecting the environment and helping to mitigate the effects of climate change. To date, seed improvement has for the most part relied on traditional breeding. Biotechnology-based approaches have tended to focus on single gene input traits. Synthetic biology opportunities include the integration of traditional crop improvement strategies with systems biology and predictive science, moving from genetic manipulation of small parts of the genome to a more engineering-based approach. The target traits are: disease and pest resistance; water and nutrient use efficiency; drought, flood, salt and heat tolerance; more efficient and nutritious plants that have increased yield but use less carbon dioxide; and better chemical synthesis (i.e. by using plants as green factories for production purposes). An ambition for the future is to be able to move what we can currently do in microbes into plants.

The application of synthetic biology approaches to plants has many challenges. However there are also advantages. Higher plants are multicellular and are the obvious starting point for the engineering of multicellular systems. Plants possess indeterminate and modular body plans, have a wide range of biosynthetic activities, can be genetically manipulated, and are widely used in crop systems for the production of biomass, food, polymers, drugs and fuels. Indeed, it is now feasible to consider creating new tissues or organs with specialized biosynthetic or storage functions. Jim Haseloff (University of Cambridge, UK) presented work on assembly of feedback regulated genetic circuits and developmental regulators that will allow the engineering of stable new patterns of gene activity in planta, and targeted reprogramming of the number and arrangement of cell types in plant organs. Manipulating the behaviour of individual cells can give rise to new morphologies, illustrating how cell-autonomous behaviour leads to the emergence of biological form (Dupuy et al., 2010). Approaches of this kind will provide a means to modify plant form and biosynthetic activities, with the ultimate prospect of producing neomorphic structures suited to bioproduction. Priorities identified by a BBSRC-funded network to investigate Synthetic Plant Products for Industrial applications (SPPIs) included the development of sets of biobricks and orthologous regulatory systems for plants, harnessing natural product diversity, remodelling subcellular storage as synthetic organelles, and making novel cell wall polymers (Rob Edwards, University of York, UK).

June Medford (Colorado State University, Fort Collins, CO, USA) provided a wonderful illustration of the extent to which a synthetic biology strategy has been applied to the development of ‘plant sentinels’ – plants able to sense and respond to specific chemicals in the environment. Obvious applications include plants that sense pollutants, pathogens or explosives, but sensing mechanisms can be potentially exploited to breed plants that, for example, flower synchronously when sprayed with chemicals. A giant step towards this goal has been realized by development and rapid optimization of an orthologous receptor and signalling system in bacteria that was subsequently transferred into plants. Ligand binding proteins, such as the bacterial periplasmic binding proteins, can be redesigned to bind specific ligands (e.g. TNT) in the cell wall at very low concentrations (as low as 23 parts per trillion), to then interact with a histidine kinase partner, and activate a PhoB response regulator which translocates to the nucleus and binds to the synthetic promoter of an ‘output gene’, generating an output phenotype (Antunes et al., 2009, 2011). One such output gene tested is a chlorophyllase gene, resulting in plants that bleach in the presence of TNT. This demonstrates very nicely how molecular components from microbes can be successfully engineered and utilized across kingdoms, in this case to develop plants with quite novel properties of industrial or agricultural value.

Alain Goossens (VIB/Ghent University, Belgium) presented a functional genomics-based strategy for large-scale gene discovery with the aim of developing a toolbox for metabolic engineering of plant metabolism. This platform exploits the fact that synthesis of many plant secondary metabolites is induced by jasmonates (De Geyter et al., 2012). Profiling of jasmonate-elicited tissues of model plants (e.g. Arabidopsis thaliana, Medicago truncatula) and medicinal plants (e.g. Madagascan periwinkle – Catharanthus roseus; Chinese thoroughwax – Bupleurum falcatum; ginseng – Panax ginseng; and yew – Taxus baccata) has enabled the establishment of an extensive collection of thousands of genes with predicted functions in different aspects of metabolism. Collections of this kind will provide parts for metabolic engineering of pathways for existing or novel plant-derived molecules with applications in the pharmaceutical, nutraceutical and agrochemical industries.

Cowpea mosaic virus (CPMV) has proved to be a very effective workhorse for synthetic biology applications (Keith Saunders and George Lomonossoff, John Innes Centre, Norwich, UK; Sainsbury et al., 2009, 2010). Modified non-infectious CPMV capsids can be readily produced in large quantities in Nicotiana benthamiana (a relative of tobacco) for use in bionanotechnological applications (e.g. drug delivery) (Fig. 2d). The untranslated 5′ and 3′ ends of the RNA-2 molecule of this bipartite positive-strand RNA plant virus have also been exploited to develop a novel transient system for high level expression of proteins in plants, again using N. benthamiana as the chassis. This has enabled expression and assembly of multiple proteins such as the heavy and light chains of antibodies, blue tongue virus-like particles and empty CPMV virus-like particles (eVLPs). The technology allows plants to be used as mini factories, manufacturing beneficial proteins and metabolites that can be used for medicinal and biotechnology purposes. The technique is faster than current methods and so it offers an extremely rapid and effective way of creating vaccines. Metabolic enzymes can also be expressed using this system, opening up opportunities to use N. benthamiana as a chassis for metabolic engineering (Mugford et al., 2009).

Bioethics and engagement

As with genetic modification (GM) in the past, there are groups with growing concerns about the use of synthetic biology. It is essential, therefore, for scientists who are working in the area of synthetic biology to be proactive in framing the debate at an early stage, as was arguably failed to be achieved by scientists working on GM in earlier decades. The involvement of social scientists in this framing process will be critical for productive engagement on synthetic biology-based research and will also be highly valuable in identifying and shaping effective pathways to impact commercialization for synthetic biology outputs. Social scientist Joyce Tait (University of Edinburgh, UK) commented that a central challenge is how to deal with idealogically motivated opposition in a way that does not allow extremists to take the lead in framing synthetic biology in the minds of uncommitted citizens. Making science more accessible to school children and the wider public will also play an important part in ‘demystifying’ synthetic biology and in raising awareness of its potential value to society. Jenni Rant (the SAW Trust, Norwich, UK) introduced the Science, Art and Writing (SAW) Initiative, a science engagement vehicle that draws on intriguing scientific images to underpin scientific investigation, coupled with exploration of scientific themes through art and creative writing ( This union of practical science, art and writing around a central scientific focus provides an innovative cross-disciplinary approach to science engagement that is accessible to all, young and old. The SAW Trust has worked with many scientists, supporting them in the design of outreach projects on topics as diverse as soil bacteria, photosynthesis, pathogens, natural products, diabetes and fractals. Jenni gave an example of how SAW had been used very effectively to explore synthetic biology with school children aged 7–9 years.

Where do we go from here?

What was very apparent to the attendees of this workshop was the excitement generated by thinking about and carrying out new science, and Synthetic Biology offers many new opportunities. More than mere technology, it requires new ways of thinking about science – how to communicate effectively with others from quite different disciplines, who speak in a different language – that of biology, chemistry, mathematics, engineering, social science. This is not easy, it does not come naturally to those trained in highly specialized disciplines, but is essential if we are to develop very new types of biological understanding.

The workshop highlighted the enormous value of bringing scientists from such disparate backgrounds together, and focusing on one problem – in this case, what synthetic biology can offer. Fundamental biological research has already resulted in an impressive resource of understanding of biological processes and what could be thought of as biological parts, for example genes and regulatory elements which can be used to engineer biological systems. Synthetic biology will contribute a complementary perspective from which to consider, interrogate and ultimately understand life processes providing new insights or even paradigm shifts in a way that may not be possible by analysis alone. We need more such workshops, so that more can experience for themselves the possibilities for this new science. These workshops could be focussed on how synthetic biology can be applied to specific problems or research areas and could be tied in to funding calls on specific themes, along similar lines to the BBSRC Enhancing Photosynthesis and Nitrogen Fixation Ideas Laboratories. We need funding to be directed towards multidisciplinary science. In the UK, the Research Excellence Framework, which the universities go through, does not make multidisciplinary working easy, as such work does not always readily get published in high impact (e.g. biological) journals. It is also difficult to find reviewers of papers and grants that cover diverse subject matter. These problems need to be overcome for progress to be made, and require those new ways of working – not just at the bench, but also at the level of grant and manuscript review.

Synthetic biology can be defined in different ways, but what is clear is that the potential for new understanding through building new biological components and testing their functions, and for developing new application, is enormous.


Videos of presentations from the 4th New Phytologist Workshop can be viewed online at A.E.O is supported by a joint Engineering and Physical Sciences Research Council(EPSRC)/National Science Foundation award as part of the Syntegron consortium (EP/H019154/1). A.E.O. and P.E.O. both receive support from the Biotechnology and Biological Sciences Research Council (BBSRC) Institute Strategic Programme Grant ‘Understanding and Exploiting Plant and Microbial Secondary Metabolism’ (BB/J004596/1) and the John Innes Foundation. K.L. and S.R. have funding from BBSRC and EPSRC. We thank Amanda Collis (BBSRC) and Lionel Clarke (Shell) for valuable discussion during the preparation of this report.