Chemical biology

Synthetic metabolism goes green

Article metrics

An extension of synthetic biology to a medicinal plant involves the transfer of chlorination equipment from bacteria. This exercise adds implements to the enzymatic toolbox for generating natural products. See Letter p.461

Plants offer a wondrous diversity of natural products for the chemist to explore and manipulate1. Their genetic, developmental and ecological complexity makes them tough targets, but O'Connor and colleagues (page 461 of this issue2) now provide an impressive example of how a plant's biosynthetic pathways can be tuned to fruitful ends.

The organismal complexities posed by plants mean that they have largely been supplanted by microorganisms as a source of natural products; microbial genetics, including the coordinately regulated and sequential arrangement of genes encoding biosynthetic pathways, is much more tractable. By contrast, plant natural products are often built by unknown numbers of enzymes, are encoded by genes lacking the orderly arrangement in microorganisms, and their fate is in part determined by task-oriented cells working in concert for biosynthesis, transport and storage3. Regardless of the source, if a natural product is to see the light of day, intervention by chemists is often necessary to create or modify what are arguably some of the most chemically impenetrable scaffolds known4.

In their paper, O'Connor and colleagues2 show that cells of a medicinal plant, Catharanthus roseus (Madagascar periwinkle; Fig. 1), can be coaxed into serving as chemical factories by using a combination of genes from microorganisms, enzyme engineering and plant-cell transformation. In this way, they have produced chlorine-containing analogues of natural products called monoterpene indole alkaloids. This alkaloid family includes pharmacologically important compounds that form the backbone of treatments for Hodgkin's disease and acute lymphocytic leukaemia. The biosynthetic installation of a non-natural chlorine atom on particular alkaloid atoms opens up previously inaccessible routes to selective chemical changes4. One day, these approaches might be used to alter drug potency and specificity while reducing side effects.

Figure 1: Catharanthus roseus, more commonly known as the Madagascar periwinkle.


This plant is well known to gardeners. But, as O'Connor and colleagues2 demonstrate, it is also a rewarding subject for the synthetic chemist.

Currently, most metabolic-engineering efforts produce plant compounds by reconstituting, in microbial hosts such as the bacterium Escherichia coli or baker's yeast, one or two enzymes known to produce plant secondary metabolites. Indeed, spectacular successes for high-level production of the antimalarial agent artemisinin, a plant secondary metabolite, have been achieved when two key enzymes of plant metabolic pathways are transformed and optimized in a microbial host5. These secondary — or more appropriately 'specialized' — metabolites ensure ecological survival of a plant, and diverge from the ubiquitous primary metabolites required for basal plant physiology. However, little attention has been focused on secondary metabolic engineering in the native plant — particularly in cases where knowledge of the genes encoding the enzymatic toolbox is still incomplete, as is the case for the clinically valuable monoterpene indole alkaloids.

These alkaloids are structurally complex, and are constructed from the common building blocks of the amino acid tryptophan and the ten-carbon terpene geraniol. Although these building blocks are in themselves chemically unassuming, they are transformed by a minimum of 14 enzyme-catalysed steps to form hundreds of monoterpene indole alkaloids6. A key step in this metabolic pathway is the removal of the carboxyl group of tryptophan by the C. roseus enzyme tryptophan decarboxylase, producing carbon dioxide and the intermediate tryptamine. Tryptamine then combines with the geraniol-derived terpene product secologanin. Strictosidine, the outcome of this coupling catalysed by the enzyme strictosidine synthase, is converted by a multitude of enzymatic transformations to form the diverse alkaloids of C. roseus.

Previously, O'Connor's group7 showed that synthetic tryptamine analogues containing chlorine atoms and fed to cell cultures of C. roseus were easily taken up by the plant cells, and that these chlorinated tryptamines were then incorporated into alkaloid products. The lack of substrate specificity displayed by strictosidine synthase, and by a series of yet uncharacterized downstream enzymes, may seem surprising. But it is increasingly clear that enzymes in general have varying levels of substrate permissiveness and mechanistic flexibility, more commonly referred to as catalytic promiscuity8. Indeed, the substrate permissiveness and mechanistic flexibility observed by O'Connor's group in their earlier study7 is the rule rather than the exception in plant secondary metabolism1. Because this earlier research bypassed the need for tryptophan decarboxylase, in the new work2 O'Connor and colleagues had first to establish that, at least in a test tube, C. roseus tryptophan decarboxylase also displays tolerance to chlorine-containing tryptophan substrates.

Having established this substrate permissiveness, the investigators turned to a class of enzyme known as halogenases to add the chlorine atom to one of two carbon atoms of the tryptophan ring. In an ironic twist on conventional metabolic engineering, they employed two genes obtained from soil bacteria that each encode site-specific halide-transfer activity to distinct carbon atoms of the tryptophan ring9. One question remained — would the wild-type strictosidine synthase from C. roseus accept these two chemically distinct chlorine-bearing tryptamine molecules to afford a larger combinatorial collection of downstream alkaloid products? Although one chlorine-bearing tryptamine was accepted, the other was not. O'Connor and co-workers turned again to earlier results10 involving structure-based engineering of strictosidine synthase to broaden its substrate selectivity, thus coercing the synthase to accept either of the non-natural tryptamine analogues.

With the necessary biosynthetic toolbox in hand, the authors2 employed a commonly used soil bacterium, Agrobacterium rhizogenes, that can insert foreign genes into plants and plant cell cultures. They generated a type of plant cell culture — known as the 'hairy-root culture' — of C. roseus that contained the microbial halogenase genes as well as the mutant form of the C. roseus strictosidine synthase. The resulting cultures produced not only the expected chlorinated tryptophan, but also a variety of downstream halogen-containing alkaloids, thereby demonstrating a somewhat surprising level of metabolic promiscuity.

This proof-of-principle study2 uses a metabolic engineering route to produce collections of modified natural products in cells from overlooked plant hosts that possess the complex enzymatic machinery necessary for these specialized biosyntheses. Many of the crucial enzymes remain unknown and are therefore not genetically accessible for expression in commonly employed microbial hosts. The product yields2 are modest, but they compare favourably with yields from other test-tube-based reconstitutions of metabolic pathways and from the rudimentary efforts to move plant alkaloid biosynthesis into microbial systems. In addition, the cogent application of catalytic tools amenable to structure-based engineering, and capable of installing a variety of chemical handles on otherwise uncooperative natural products, should expand plant natural-product discovery into the domain of the medicinal chemist.

O'Connor and colleagues' paper2 provides clear directions for engineering greater catalytic promiscuity into the C. roseus tryptophan decarboxylase, thereby alleviating a metabolic bottleneck in the hairy-root culture system. As such, the work stands as an elegant example of how choosing what seems to be a circuitous experimental route may actually provide a more direct path to success for the synthetic biologist.


  1. 1

    Austin, M. B., O'Maille, P. E. & Noel, J. P. Nature Chem. Biol. 4, 217–222 (2008).

  2. 2

    Runguphan, W., Qu, X. & O'Connor, S. E. Nature 468, 461–464 (2010).

  3. 3

    Ziegler, J. & Facchini, P. J. Annu. Rev. Plant Biol. 59, 735–769 (2008).

  4. 4

    Deb Roy, A., Grüschow, S., Cairns, N. & Goss, R. J. M. J. Am. Chem. Soc. 132, 12243–12245 (2010).

  5. 5

    Ro, D.-K. et al. Nature 440, 940–943 (2006).

  6. 6

    O'Connor, S. E. & Maresh, J. J. Nat. Prod. Rep. 23, 532–547 (2006).

  7. 7

    Bernhardt, P., McCoy, E. & O'Connor, S. E. Chem. Biol. 14, 888–897 (2007).

  8. 8

    Tawfik, D. S. Nature Chem. Biol. 6, 692–696 (2010).

  9. 9

    Ryan, K. S. & Drennan, C. L. Chem. Biol. 16, 351–364 (2009).

  10. 10

    Loris, E. A. et al. Chem. Biol. 14, 979–985 (2007).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Noel, J. Synthetic metabolism goes green. Nature 468, 380–381 (2010) doi:10.1038/468380a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.