Tuesday, June 25, 2013

Stop using that word: Accordance

Consider "accordance". It's a sleek, shiny word. But it's terribly misused in scientific manuscripts.

To pick on just one example, take the following text from a recent publication by Neil Kelleher's group at Northwestern (bold emphasis mine):
Our observations were in accordance with a previous study on nostocyclopeptide, where certain amino acids in the peptide sequence were found essential for the spontaneous macrocyclization of the peptidyl aldehyde intermediate into a cyclic imine.
And another example from a total synthesis of daptomycin by Xuechen Li:
The spectrum was in full accordance with those in the literature.
Not quite right.

"Accord" implies agreement. It's what writers usually mean when they use "accordance". For instance:
His dismal personal life was in accord with his excellent progress in total synthesis.
"Accordance", in contrast, implied obedience. It refers to compliance with a rule.
My one day of vacation per year is in accordance with group policy.
Examples of the correct use of "accordance" are actually hard to find. Consider this paper by Robert West at Wisconsin. An excerpt (bold emphasis mine):
In accordance with Bent’s rule, the increased R–O bond polarities of permethylated species lead to increased oxygen hybrid s-character and R–O–R bending angles in both ethers and siloxanes.
That one is arguably correct.

The difficulty is probably that "in accord with" doesn't flow quite smoothly. Perhaps a better option would be simply to say "in agreement with". Either way, reviewers probably won't catch it.

This has been a public service announcement.

Thursday, June 20, 2013

Carbenes: turns out, nature has them, too

Enzymes are like nature's little gloveboxes. It's really quite interesting what kind of chemistries are possible in aqueous, biological conditions just by manipulation of the local electronic and steric environment by the structure of enzymatic active sites.

Take carbenes. Everybody likes a carbene--it's a lone pair on carbon, but it's formally neutral, and it does interesting reactions like alkene insertion (cyclopropanation), C-H insertion, rearrangements, and the like.

One of the more interesting aspects of carbenes is their variable reactivity. They can exist in a singlet (depicted as lone-pair) or triplet (depicted as diradical) form. Depending on their electronic environment, they can react as nucleophiles (aided by a high-lying HOMO) or electrophiles (encouraged by lowering the LUMO). Their reactions can be stereospecific through concerted pathways (singlet carbenes) or non-stereospecific through stepwise mechanisms (triplet carbenes). All of this is tuned, not surprisingly, through the electronic/steric environment around the carbon in question.

Carbenes are typically highly reactive and short-lived, though examples of persistent carbenes are now well-known. N-heterocyclic carbenes (NHCs) make a prime example. The electron-rich di-adamantyl NHC shown above, for instance, was described in 1991 and can be crystallized (it melts, by the way, at 240 degrees Celsius).

Note the electronic nature of the carbene: the carbon is flanked by two nitrogens, each bearing lone pairs capable of donating electron density into the carbene's p orbital. This acts to stabilize the singlet state and imbues NHCs with admirable properties as metal ligands (electron rich sigma donors which bond quite strongly to metal centers) The most famous of these is probably Grubbs' second-generation catalyst, which bears an NHC in lieu of one of the phosphine ligands of the first-generation counterpart. Besides olefin metathesis, though, persistent carbenes (as NHCs) are quite useful ligands for tricky C-C cross-couplings. Specifically, the so-called Pd-PEPPSI complexes are useful for coupling of tetrahedral carbon centers to each other.

It turns out that nature utilizes carbenes as well. Take a look at vitamin B1--also known as thiamin. It's an essential coenzyme which, as it turns out, we can't make and must obtain in our diet. There's several forms consisting of various decorations, usually of the hydroxyl moiety. One of these is thiamin diphosphate (ThDP), which, if you didn't guess, is thiamin with diphosphate attached (it also goes by the name thiamin pyrophosphate, or TPP, which is definitely not confusing at all). ThDP is a coenzyme for pyruvate decarboxylate and pyruvate oxidase, among other enzymes.

If you look at thiamine, there's a place--right between that sulfur and its neighborly nitrogen--that seems like a nice candidate for a carbene. There's been a debate in the literature; that carbon must be deprotonated for catalytic activity, and it hasn't been clear whether the associated enzymatic reaction proceeds via a carbanion or the short-lived carbene. A recent report in Nature Chemical Biology provides evidence for the latter. The authors examined thiamin diphosphate with the enzyme pyruvate oxidase (from bacterial origins). Phosphate was employed as a mimic of the substrate (pyruvate) that would bind similarly but not form a covalent adduct--this was to see if substrate binding might correspond with the formation of a carbene.

Via circular dichroism (CD), and X-ray diffraction, the authors give evidence that although the coenzyme is C-protonated in its resting state, binding of phosphate results in accumulation of either the carbanion/carbene form. This is narrowed down to the carbene chiefly through analysis of the XRD structure. Not only was the electron density consistent, but the bond lengths and angles were similar to synthetic thiazolium carbenes previously reported. The authors mention similar results upon analysis of the ThDP/cyclohexane-1,2-dione hydrolase complex.

Essentially, under physiologically relevant equilibrium conditions, the thiamine/enzyme complex can accumulate a carbene. In water. That's quite cool.