A computer graphic showing part of the active site of retro-aldolase, a new enzyme designed to break unnatural carbon-carbon bonds.
Credit: Jason DeChancie/UCLA
Mind-boggling calculations
To make things as easy as possible, the researchers focused on designing enzymes for two well described chemical reactions: one responsible for breaking carbon bonds, the other involved in removing protons from carbon. This way, they were able to fairly confidently characterise the geometric structures required of their synthetic enzymes' active sites, says Röthlisberger.
The next step was to figure out which amino acids could be used to generate these conformations – a monumentally complex task.
Röthlisberger's group at the University of Washington, led by biochemist David Baker, developed a computer program to model vast numbers of sequences and predicted enzyme shapes. This stage was so computationally intensive that the researchers set up an online network enabling thousands of volunteers worldwide to donate their computers' downtime to the effort.
From here the rest of the project was relatively easy. "As soon as we could obtain the design model [in the computer], we could apply standard molecular and biochemical methods to go from the amino acid sequence to the DNA sequence to the expressed genes," says Röthlisberger.
And so the team were able to produce the first functional synthetic enzymes. But while naturally occurring enzymes speed reaction rates by many billion (or even trillion) fold, the synthetic enzymes gave more conservative boosts – around 100,000 fold.
"The acceleration [achieved] is really rather modest by comparison to Nature," admits Houk – but it's still incredibly exciting. The team has managed to show, for the first time, that three-dimensional protein structures can be modelled and translated very closely into functional catalysts, he says.
'Fake' enzymes, real applications
Synthetic enzymes could be designed to improve on the functions of existing natural enzymes, but Röthlisberger believes they will prove most useful where no natural counterpart exists - when applied to reactions not known in the natural world.
Houk sees synthetic enzymes as being particularly useful in the field of medicine, in the synthesis of pharmaceuticals, for example. "One could imagine a process group designing and making an enzyme to carry out efficient reactions to produce large quantities of a drug precursor," he says. "Unnatural enzymes might also be used to directly carry out reactions that destroy toxins or do other active processes that might be of direct medicinal value."
The enzymes might also play a role in defence against biological warfare. Enzymes could ultimately be designed, Houk says, to catalyse the reactions needed to destroy biological threats, like bacterial or viral components.
Biochemist Sue Brown, from Australia's CSIRO in the Australian Capital Territory, describes the work as "very exciting research ... taking the best of nature and modern chemistry and combining them". Brown, who was not involved in the project, thinks one of the most important aspects of the new enzymes is that they are "environmentally friendly".
Whichever industry they're used in – be it pharmaceutical, agricultural or chemical – Brown says the synthetic enzymes "have great potential to be used in … reducing the use of solvents and heavy metals, or degrading harmful chemicals such as pesticides and toxins".
