Fierce debate has erupted in the Journal of the American Chemical Society (JACS) over a phenomenon known as the anomeric effect. The controversy reminds us once again that while observations are usually verifiable, interpreting results is something all scientists need to play a part in.

The anomeric effect, familiar to many chemists, is important because it influences the shape of sugar molecules, which are, of course, biologically and medicinally relevant. In essence it means that sugars prefer their C1 hydroxyl groups to be in an axial orientation, whereas other cyclohexane based systems usually have their large substituents in an equatorial orientation. The reason for the anomeric effect is normally assumed to be hyperconjugation of a lone pair of electrons on the ring oxygen to the antibonding orbital of the carbon next door. The truth is though that academic dispute over the cause of the effect still exists.

Debate and discussion is integral to scientific progress

It’s difficult to study the anomeric effect, firstly because in solution it’s difficult to isolate solvent molecules’ contributions to it and secondly because there is a relative scarcity of methods which can probe the underlying parameters which give rise to the effect.

It was interesting then, when Oxford University’s Ben Davis managed to create a gas phase complex of separate anomers and resolve their electronic spectra. He did this by shining a laser on a dry carbohydrate (D-glucose) mixed with an oligopeptide receptor. This created a gas phase mixture of the two anomers complexed separately by the receptor. The complicating solvent issue was removed and the α- and β-anomers could be studied independently. The infra-red spectra of the two complexes differed in some key points – notably the frequency of the peptidic N-H bond signals. This indicates the hydrogen bonding interactions between the receptor and the sugar is different for each anomer.

The electron density on the ring oxygen should be different for each anomer due to the different degrees of hyperconjugation. In theory this would contribute to different hydrogen bonding strengths between the ring oxygen and the peptidic NHs.

Interesting stuff. Except that a shadow was about to fall over Davis’ work in the shape of US-based computational chemist Yirong Mo. Almost immediately after Davis’ work came out Mo and his colleagues published what you might call a stern rebuttal of their conclusions. What’s a little disappointing is that Mo is just one of several scientists to comment on Davis’ work but since his rebuttal is put in the strongest possible terms, it seems likely that it is this paper which will garner the most attention. Mo’s work consists of computational modelling of a very similar system to the one Davis used where the anomeric effect is disabled. Mo showed that in this case, his modelling still predicts the same spectral differences that Davis observed and, therefore, says that the changes are not caused by the anomeric effect. The abstract of Mo’s paper states that Davis’ ‘”sensor” cannot probe the anomeric effect as claimed’, essentially proposing that Davis’ experiment is flawed.

I spoke to Davis and Mo as well as computational chemistry experts Jeremy Harvey from the University of Bristol and Jonathan Goodman from Cambridge University. A very subtle argument emerged. ‘The bottom line here is that intermolecular interactions – as between the sugar and the peptide model – are complicated,’ Harvey says, ‘assigning their strength based on relatively simple concepts such as the anomeric effect is hard. Spectroscopy, and energy decomposition analysis […] provide lots of insight, but leave lots of room for ambiguity.’

Davis emphasised that his work is very much a reporting of experimental results and that he doesn’t insist upon any specific conclusions about the causes of the anomeric effect. But he says ‘it’s great to see our experimental results stimulating such a lively debate in the theoretical community’.

To the uninitiated (which, given the complexity of the arguments involved, is many people) it looks like Mo’s paper shows Davis’ to be completely wrong. That is, after all, what he says in his abstract. This could be an instance when a little post-publication peer review could make a big difference to people’s perceptions. Picking apart the arguments is complex and time consuming, so when people take the time to look at complex debates like this it would be great to see their ideas and conclusions posted alongside the articles. That would make it clear that, actually, there is a debate to be had here, and as scientists, we want to stimulate it.

Josh Howgego

VN:F [1.9.10_1130]
Rating: 10.0/10 (2 votes cast)
VN:F [1.9.10_1130]
Rating: 0 (from 0 votes)
Anomeric argument, 10.0 out of 10 based on 2 ratings
Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)