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

Related articles

• New proteins might be supernatural
• Seafloor vents origin of life
• Upcoming drug revolution takes a quantum leap
• Enzymes may allow universal blood transfusions
• Artificial gut may replace animal testing

As far as proteins go, enzymes are seriously hard workers. They play a key role in almost all chemical reactions occurring in nature, from cell division and replication to digestion of that juicy steak you ate for dinner last night. Acting as 'catalysts', their job is to make otherwise lethargic reactions occur at much faster, biologically useful speeds.

The power of enzymes is also harnessed commercially. Scientists have been able to modify existing enzymes for use in everything from baking bread and brewing beer, to the development of novel cleaning products and essential pharmaceuticals, including antibiotics and blood pressure medications.

But commercial applications are limited by the fact there are only so many enzymatic tools that nature can provide us with. "At the end of the day, although she does so very well, nature uses only a very small fraction of 'chemical space'," says biochemist Dan Tawfik, from the Weizmann Institute of Science in Rehovot, Israel.

Tawfik is part of an international research team aiming to break through this barrier. And they're well on their way, with the successful design and creation of two new, entirely synthetic enzymes recently reported in the journals Nature and Science.

"For the first time, we have been able to computationally design [and create] enzymes from scratch," says team member Daniela Röthlisberger, a biochemist at the University of Washington in Seattle, USA.

These enzymes are capable of dealing with substances for which no naturally occurring enzymes have evolved, she says. And in principle, the technique could be used to produce enzymes capable of "accelerating any desired reaction at all". The possibilities, the researchers agree, are vast and intriguing.

Tricky business

Making enzymes is a tricky business. Even the names of the techniques sound mind-boggling, with the process involving a mix of 'quantum mechanical computation', 'advanced protein engineering' and 'directed evolution'.

A key difficulty – one that has long plagued researchers – lies in trying to mimic the astonishing structural and functional complexity of natural enzymes, says Tawfik.

Like all proteins, enzymes are made up of subunits called amino acids. The twenty amino acids are like building blocks that can be used in any number and combination to form the chain that makes up a protein. This chain must in turn take on a highly specific three-dimensional structure, also guided by the sequence of amino acids.

In enzymes, the protein must fold correctly into its particular shape in order to create what is known as the 'active site'. This is the critical region of the molecule – its sequence and shape is what allows the enzyme to bind target molecules and perform its catalytic function.

So there's quite a lot to consider when building an enzyme from scratch. As such, one of the goals of the project has been "to beat the daunting odds that a new protein could be put together that would fold [correctly and] as predicted, and have a novel function," says theoretical chemist Kendall Houk, part of the research group at the University of California in Los Angeles.