Desymmetrizing dinitrile substrates

Desymmetrization of dinitrile substrates using nitrile-active enzymes like NHases to get a chiral compound has been known about for ages, (Turner and Sugai & Ohta published early papers on this in 1993).

A nice later example of this is a biocatalytic synthesis of a (S)-methylDOPA precursor by Sugai and Ohta again:-

Realization of the synthesis of a,a-disubstituted carbamylacetates and cyanoacetates by either enzymatic or chemical functional group transformation, depending upon the substrate specificity of Rhodococcus amidase by Masahiro Yokoyama, Mieko Kashiwagi, Masakazu Iwasaki, Ken-ichi Fuhshuku, Hiromichi Ohta and Takeshi Sugai in Tetrahedron Asymmetry doi:10.1016/j.tetasy.2004.04.047

In the recent book “Practical Methods for Biocatalysis and Biotransformations” edited by Whittall and Sutton, there is a section on desymmetrization using nitrile active enzymes (though two are whole cell methods and the last is use of a Codexis nitrilase) written by Wijdeven, Kielbasinski and Rutjes (Section 5.5, p186-189).

NHase reviews in recent books

I have recently come across two book chapters which would be of interest to those interested in nitrile hydratase (and indeed nitrilases) enzymes.

Biocatalysis for the Pharmaceutical Industry: Discovery, Development, and Manufacturing edited by Junhua (Alex) Tao, Guo-Qiang Lin and Andreas Liese has a chapter on “Applications of Nitrile Hydratases and Nitrilases” written by Grace DeSantis and Robert DiCosimo.

Biology of Rhodococcus edited by Héctor Alvarez, whilst sounding a touch unpromising has a chapter co-written by Ludmila Martınkova entitled “Catabolism of Nitriles in Rhodococcus” which runs through a lot of the synthetic potential that the NHases from this type of prokaryote possess in detail I haven’t seen altogether in one place before.

Wild Type Fe-type Nitrile Hydratase from Comamonas testosteroni Ni1

The PDB from Comamonas testosteroni Ni1 is currently waiting (3/7.12) to be processed at the Protein Data Bank but will be called 4FM4.

Connecting B to A

Looking back on the bioinformatic analysis of eukaryotic nitrile hydratases in Marron, Akam and Walker's PLOS One paper to be found here, I thought it would be interesting to look at the link between the two segments which are separate subunits in prokaryotic NHases.
I mocked up this model of where the link had to go from two prokaryotic subunits which had most similarity to the Monosiga brevicollis NHase. The dangling ends of the right hand side of the structure indicate where the linker has to go.

As the easiest to access, I chose to align the protein sequences suggested by the transcriptions of ESTs from the four organisms held within the Broad Institute's Origins of Multicellularity project, namely M. brevicollis, Thecamonas trahens, Salpingoeca rosetta and Sphaeroforma arctica. (NB order in alignment: MB, SR, TT and last SA)

As Marron's paper comments:
"This histidine-rich region is prominent in T. trahens (12 residues), S. arctica (11 residues) and M. brevicollis (17 residues), but shorter in [...] S. rosetta (2 residues)."
That seems quite a lot of variation with only M. brevicollis having a string of pure histidines. In contrast the sequences are around the metal binding segment are quite similar.

New paper, new crystal structure

There is a new paper just released online in Biochemical and Biophysical Research Communications which is very interesting:

The Fe-Type Nitrile Hydratase from Comamonas testosteroni Ni1 Does Not Require an Activator Accessory Protein for Expression in Escherichia coli by Misty L. Kuhn, Salette Martinez, Natalie Gumataotao, Uwe Bornscheuer, Dali Liu  and Richard C. Holz.

http://dx.doi.org/10.1016/j.bbrc.2012.06.036

This paper reports something that has been something we have wondered about for a while… how important is that activator protein that is commonly cloned into E.coli clones alongside the DNA for the alpha and beta subunits. They show that one of the reasons for poor expression or activity for iron centred E. coli clones could well be down to codon bias. This isn’t obviously the full story because they find that whilst you don’t need the activator for the NHase from Comamonas (which they call CtNHase), you sure do for the one from Rhodococcus equi TG328-2.

The icing on the cake for this paper is that they have a crystal structure (though as of today 25/6.12 it isn’t on http://www.rcsb.org/) of CtNHase. This shows a slightly different arrangement of side chains in the active site but possibly more interestingly this active site isn’t at the end of a long dark tunnel but is relatively solvent-exposed, allowing easy direct access to the axial position on a bound iron. It only got assayed by the standard acrylonitrile assay but one wonders whether it would be rather less sensitive to steric crowding around the nitrile being hydrated than is usual with NHases.

PS This paper also has my favourite use of the word “recently” in a communication… to reference 12 from 2003.

Data to enzyme

I was talking at the "Leading Industrial Biotechnology: Focus on Biocatalysis" meeting last week in York, and I put up this slide which caused a bit of comment amongst those of a chemistry background in the audience who didn't know the power of state-of-the-art molecular biology for those looking for new enzymes not available from catalogue based suppliers. Moral of the story: find yourself a molecular biologist and you will have more (and more interesting) enzymes available than you can imagine.

The Importance of Oxidation in VCTLCSC

Just before my sabbatical on this blog, I was looking at the length of bonds which linked the metal to the sulphurs of the cysteine residues to test the state of evidence in the small molecule cystallographic record on how oxidation might change such bond lengths. There wasn't much data around for Co and Fe based systems to draw many inferences from. However, a new paper is just out looking at a model system based around Ru which has been synthesized in the various oxidation states of sulphur with the hope of giving some clues on what sulphur oxidation might offer to NHases.

Influence of Sequential Thiolate Oxidation on a Nitrile Hydratase Mimic Probed by Multiedge X-ray Absorption Spectroscopy by Jason Shearer, Paige E. Callan, César A. Masitas, and Craig A. Grapperhaus in Inorganic Chemistry (DOI: 10.1021/ic202453c).

They use a range of techniques including multiedge X-ray absorption spectroscopy and some serious hardcore DFT calculations to look at/predict how ionic and how Lewis acidic the complexed metal becomes as the sulphur ligands are sequentially oxygenated. Basically you get a harder Lewis acid as you move from naked sulphur through to a mixed sulfenato/sulfinato species. As the abstract graphic shows....

In the final few paragraphs, they speculate what this means for the mechanism of nitrile hydratase hydration (they say "hydrolysis" but that means the complete breaking up the whole bond to acid and ammonia in my book, not just breaking a couple of pi bonds with a molecule of water.). They show versions of the two most commonly proposed models- metal-bound hydroxide attacking a free floating nitrile, and metal-bound nitrile getting walloped by a hydroxide. They suggest that increasing hardness of the metal centre should favour the attachment of an oxygen based ligand over the nitrogen of a nitrile, and hence favour the metal bound hydroxide model. They end with the comment that harder Lewis acidity would actually aid NHase catalysis whatever was bound to the metal.