Re-engineering a tiny enzyme

Gerrit Poelarends was looking for an enzyme that would make a key step in the production of important pharmaceuticals greener and much more efficient. It led him to a tiny protein called 4-oxalocrotonate tautomerase. Over a period of five years, he managed to map the possibilities it offers and re-engineer it to perform the required step. The results were published in Nature Communications on 8 March.

There is a class of widely used pharmaceuticals based on the compound gamma-aminobutyric acid (GABA), which acts as an inhibitory neurotransmitter in the brain. Derivatives are used to treat conditions such as epilepsy or neuropathic pain. The synthesis of GABA derivatives involves a step in which the carbon atoms of two building blocks need to be connected. This requires some complex and fairly dirty organic chemistry. ‘So I asked: Is there an enzyme that would do the trick?’ explains Gerrit Poelarends, Associate Professor of Pharmaceutical Biotechnology.

The short answer was ‘no’. But the catalyst used in the reaction, which occurs in fairly obnoxious organic solvents, is a derivative of an amino acid called proline. ‘So it made sense to assume that an enzyme that has proline in its active site, the part of the enzyme that performs the reaction, might do the trick.’

Fortunately, Poelarends had been working with just such an enzyme, which bears the name ‘4-oxalocrotonate tautomerase’. It is a very small enzyme, made up of just 62 amino acids, whereas the average enzyme is easily made of hundreds of amino acids. ‘We tried it and the enzyme did indeed catalyze the crucial step in GABA formation.’ There were just two problems: the enzyme wasn’t very active and it produced the wrong end product.

That could have killed the whole idea, but for Poelarends and PhD student Jan-Ytzen van der Meer it was just the beginning of a very productive research project. ‘We wanted to see how we could engineer the enzyme to do what we wanted.’ Because it is such a small enzyme, they could use a systematic approach: ‘Nature uses twenty different amino acids to build proteins. So we decided to change each of the 61 amino acids of the tautomerase into 15 to 19 other amino acids.’ The first amino acid, which is the proline in the active site, was left intact.

A commercial firm produced just over a thousand mutants of the tautomerase gene, which were expressed in cells. The mutant enzymes were then individually tested for activity. ‘This produced a kind of a map that showed us several ‘hot spots’, where changing an amino acid would increase the required activity.’ Combining the mutation from several hot spots resulted in additional increases.

But there was also the problem of the enzyme producing the wrong end product. ‘The reaction can produce two variants which are each other’s mirror image’, Poelarends explains. These two ‘enantiomers’ differ from each other in the same way that our left hand differs from our right. And the enzyme produced the wrong ‘hand’. This is a serious problem, as only one ‘hand’ will be effective for most pharmaceuticals. So they tested their mutants for enantioselectivity or ‘handedness’, and again found hot spots.

The results of this work were published in the journal Nature Communications. ‘What we present is a first step – the end result needs to be improved further before the enzyme can be of interest to the pharmaceutical industry’, says Poelarends. As his research team also elucidated the 3D structure of the improved mutants, he has a pretty good idea about how to proceed to make the necessary improvements using ‘rational protein design’.

But the results so far are impressive enough. Activity is up about 20-fold and the enantiomeric excess of the desired product is in the upper 90 percent region. ‘Five years ago, there were no known enzymes that would catalyze this step in the production of GABA derivatives. We found an enzyme with a very low activity and the wrong end product and turned it into this.’

Over the last year, Poelarends secured three research grants to continue his work – from the European Union, the Dutch research organization NWO and a European Research Council ‘proof of concept’ grant, totalling well over a million euros. The ‘final product’ he is aiming for is a cell that can be fed basic nutrients and will synthesize useful compounds such as GABA. Such a biological system would do away with polluting or energy-intensive chemical synthesis.

‘And there may be more applications for this enzyme’, explains Poelarends. This tautomerase is a highly ‘promiscuous’ enzyme, which means it can catalyze many different reactions, albeit at low efficiency. Using his library of mutants, Poelarends can easily make new activity maps that will help him to redesign the enzyme for other purposes, such as the synthesis of amino acids for new peptide antibiotics. ‘It’s a great system. Our investment over the last few years is certainly paying off.’

Reference: Jan-Ytzen van der Meer, Harshwardhan Poddar, Bert-Jan Baas, Yufeng Miao, Mehran Rahimi, Andreas Kunzendorf, Ronald van Merkerk, Pieter G. Tepper, Edzard M. Geertsema, Andy-Mark W.H. Thunnissen, Wim J. Quax and Gerrit J. Poelarends Using mutability landscapes of a promiscuous tautomerase to guide the engineering of enantioselective Michaelases, Nature Communications, 8 March, DOI 10.1038/NCOMMS10911 .