Many of the drugs we use are natural compounds or their derivatives, obtained from plants, fungi, or bacteria. Unfortunately, these organisms produce them for their own needs, and don't always make enough for us to obtain them in sufficient bulk or purity. One of the things that motivates the field of synthetic biology is the hope that we can design an organism from scratch so that it will make useful compounds like these drugs. But a paper that will be released by PNAS later this week suggests that there may be an easier way to go about things: take an existing bacteria and delete anything it doesn't need to make the drugs.
One of the challenges of engineering bacteria to produce natural compounds is that the chemicals involved in the production process—the biosynthetic pathway, as it's called—may come from many different parts of a cell's metabolism. So, for example, the biosynthetic pathway may stitch together a piece of a sugar, part of a broken-down protein, and some lipid in order to make a useful drug compound. So, it's not simply enough to identify the enzymes that catalyze the steps in a biosynthetic pathway; you have to identify the raw materials, too, and ensure that your engineered bacteria makes all of those.
There are two potential approaches to this, each with its share of drawbacks. One is the ground-up approach, most closely associated with Craig Venter, who has partnered with Exxon-Mobil to develop biofuels based on his technology. Venter's group started with a bacteria that has the smallest known genome, called Mycoplasma genitalium (yes, as the name implies, it's a genital parasite), and then deleted every gene in it to determine which were essential for it to survive in culture. He's also developed the ability to construct an artificial genome, which should allow him to replace the one that the bacteria already has.
This synthetic approach has two significant advantages: complete control, and the fact that the bacteria is only using the bare minimum of energy to stay alive, meaning any excess will be funneled into making your molecule of choice. The downside, however, is that you have to have complete knowledge of every step of the biosynthetic pathway, including every place where a chemical is siphoned off from the basic metabolism. For many natural compounds, we're not there yet.
The new paper takes an opposite approach: start with a useful bacteria, leave its entire metabolism intact, and then strip out anything you're not interested in. In this case, the authors started with a species of Streptomyces, a soil bacteria that's already in use for industrial production. The soil environment is a very competitive one, and different Streptomyces species produce a variety of useful compounds; the authors list chemicals with "antibacterial, antifungal, antiviral, and antitumor activities but also antihypertensive and immunosuppressant properties." These chemicals, which are optional for the bacteria's survival in the lab, are derived from the core metabolism, and called secondary metabolites.
Conveniently, the genes for most secondary metabolites are primarily clustered in two locations in the genome. So the authors simply deleted them, taking 1.4 Megabases (about 20 percent of the genome) out with a single targeted mutation. A few other specific biosynthetic pathways were also lopped off, leaving a Streptomyces strain with little more than its core metabolism.
They then plugged a few biosynthetic pathways back individually. For example, they were able to take a cluster of genes that produced the antibiotic streptomycin and drop it into their strain. It produced the antibiotic at higher levels than it would normally, presumably because it didn't have additional pathways to divert chemicals to. Another experiment converted the lab strain to one that produced a compound with antitumor properties. They could even control the expression of these biosynthetic pathway by manipulating the genes that normally regulate its activity.
In addition to compounds normally produced by bacteria, they were able to tweak a plant gene and get the Streptomyces to produce a compound called amorpha-4,11-diene. That's a precursor to the plant product artemisinin, which is a potent antimalarial drug.
Because so much of the basic metabolism of the bacteria is intact, the researchers were able to get away with a minimum amount of genetic engineering, and didn't need to identify every step in a biosynthetic pathway to get things to work. But, because of the deletions, the bacteria were focused on generating a single product, which allowed them to produce it more efficiently (though probably not as efficiently as one with a synthetic genome could).
In the end, both of these approaches will have their uses, and it's possible the two might ultimately be integrated into a single process, with bacteria producing an intermediate, which synthetic organisms go on to shape into the final product. However things work out, it's good to have options.
PNAS, 2009. DOI: 10.1073/pnas.0914833107
reader comments
16