Protein export

Currently, vascular surgeons perform heart-bypass operations with arteries taken from the legs, whose diameters are usually an imperfect match for the coronary arteries to which they have to be joined.

The more familiar applications for super-strong fabrics woven from synthetic spider silk include bullet-proof vests more resistant to penetration than Kevlar, super-lightweight parachutes that could fold up to pocket size, or near-indestructible lightweight sails for high-performance racing yachts.

Voigt's team chose Salmonella bacteria because they have syringe-like transmembrane channels that inject toxins into the cells of their mammalian hosts. These structures can be co-opted to export proteins from the cell interior into the growth medium.

As a toolkit component, the syringes can be employed for any biosynthetic process requiring the protein product to be exported.

They can also modify genes in a way that allows the host microbe to produce the "right stuff" from animal genes - normally, bacteria are unable to perform processes like glycosylation, the addition of sugar molecules that fine-tune protein function.

Voigt's team is collaborating with another group at Berkeley that is designing microfluidic devices to mimic spider spinnerets - the tiny nozzles that secrete silk protein fibrils, to be woven into super-strong threads.

Bacteria are preferred to yeast as biofactories. Voigt says their simpler, more stable genetic architecture is more accommodating of transgenes. Yeasts tend to eject transgenes from their chromosomes.

As for applications, blue sky is the limit. "Biofuels is a big one, and a big part of what we're doing," Voigt says.

"We're trying to develop cells capable of secreting cellullase enzymes to digest waste agricultural products and convert them into fuels like alkanes and longer-chain alcohols, which have greater energy density than ethanol."

There is also potential to create designer algae or bacteria to capture carbon dioxide from coal and oil-fired fired power stations and synthesise it into biofuels - effectively using greenhouse gas emissions from fossil fuels as a renewable energy source.

As an example of the potential of the new technology, Voigt and his colleagues described a potentially revolutionary cancer therapy, based on re-engineered E. coli bacteria, in the Journal of Molecular Biology in 2006.

They engineered the microbes to express the invasin protein from Yersinia pseudotuberculosis, a cousin of the plague bacterium Y.pestis. Invasin allows Yersinia to invade its host's cells.

They propose to place expression of the invasin gene under the control of environmental sensors selected to distinguish between normal and cancerous cells by detecting biochemical cues exclusive to cancerous cell lines - when the microbe detects the cancer, it activates the invasin gene and penetrates the cancerous cell.

They also equipped the E. coli with a quorum-sensing biochemical circuit from the luminescent bacterium Vibrio fischeri lux, and a hypoxia-sensitive promoter from the fdhF (formate dehydrogenase F) gene from E. coli.

They also experimented with an arabinose-inducible araBAD promoter sequence, as an alternative to the hypoxia sensor.

The quorum-sensing circuit allows the bacteria to congregate in large numbers, while the fdhF promoter activates synthesis of a cytotoxic agent under hypoxic conditions - tumour tissues are typically hypoxic because of inadequate blood supply.

The engineered bacteria invaded the cancerous cells at densities greater than 108 bacteria per millilitre, after activation in an anaerobic growth chamber. Where the arabinose-inducible promoter was used, activation was via growth in a 0.02 per cent arabinose solution.

Voigt says that this approach could be used to engineer bacteria to sense the microenvironment of a tumour, and respond by invading the cancerous cells and releasing a cytoxic agent - presto, a highly selective tumour-killing therapy, with none of the toxic effects of chemotherapy or radiotherapy.