One-step total synthesis
24 July 2022 - Orgo
Somewhere last year, chemists at the French BioCIS introduced the concept of one-step total synthesis. They showed that the reaction of butane-2,3-dione with L-proline catalyst will yield the complex polycyclic molecule nesteretal A in just a single step (DOI). Okay, the yield was 1% and the reaction time several weeks but the reason I decided to complain sort of about the article on twitter was not that but the entire notion of a one-step total synthesis.
Total synthesis can not be just one step on principle. It is supposed to be hard multistep no-cutting-corners labor. Imagine a hollow-eyed grad student looking at just 1 mg of product with the realization that there are 20 more step to go and that funding is running out next month. A successful total synthesis will get you awarded a PhD but only after 4 years of serious work. If the scientific community does not intervene, next thing you know a PhD Can be earned in a single week.
This year a new offender: chemists at the University of Adelaide have floated a total synthesis of peshawaraquinone (DOI). Again one-sep, just react 3-methylcrotonaldehyde and 2-hydroxy-1,4-napthoquinone with DMAP in toluene and a complex polycyclic molecule emerges. But a one-step total synthesis? Twitter!
All joking aside, what is peshawaraquinone and what makes the peshawaraquinone synthesis unusual? The compound was first isolated in 2015 by researchers from the University of Peshawar from the local plant Fernandoa adenophylla (Link). Based on X-ray crystallography the compound is chiral with 6 chiral centers although the analysis did not include optical activity. A systematic name is included in the original article but Opsin is unable to handle it. The plant is used in traditional medicine for a whole host of ailments. In one study the action of peshawaraquinone was compared to that of pain reliever tramadol in a molecular docking study (DOI).
As a reaction mechanism for the new peshawaraquinone synthesis the authors propose a biomimetic cascade with a Knoevenagel step, a retro-cyclisation, a Michael addition and a 3+2 cycloaddition. The authors also explain that the reaction is unusual because the reaction delivers just two diastereomers (an epimeric pair) from potentially a large number of diastereomers. They recall one classic synthesis from the nineteenseventies, that of carpanone as a result of dimerization of carpacin as a single diastereomer (DOI).
This blog takes an interest in organic reactions and is maintaining an organic reaction database. This reaction must certainly be included but now the question is bubbling up, how to graphically depict it?
If a systematic name offers no guidance then there is the chemical drawing that accompanies the article. The peshawaraquinone diastereomers are fitted with the not unfamiliar unwedged hashed bond (||||||||) as opposed to regular wedged bonds. Some bonds are pointing upwards and some of them downwards not because the molecule contains chiral centers but information needs to be relayed about cis-trans configurations in the annulated bicyclic constructs in it. Think achiral cis and trans decalin. But this notation has problems. First of all it is not at all recommended. From the IUPAC “graphical representation of stereochemical configuration” by Jonathan Brecher (2006) we learn:
“The most common (but still obscure) of these conventions says that bold and hashed wedges should be used for absolute configuration, unwedged bold and unwedged hashed lines represent relative configuration and racemic character”
and
“Given the lack of consensus among different systems and the poor general acceptance of any given one, the safest approach currently is to depict multiple structure diagrams explicitly, optionally accompanied by additional descriptive text ”
meaning explicitly drawing both enantiomers of a single diastereomer or adding a descriptive text such as “and enantiomer”. Of course 2006 is a long time ago but have there been new recommendations or proposals since?
The second problem with the unwedged bond is that the SMILES line notation and software libraries (rdKit) that I use for creating the reaction drawings do not support this type of bond. Graphics software like Chemdraw or BKChem of course support it but the whole point of curating a database with a million records is NOT having to draw that many images in an editor. So decision time! Of course peshawaraquinone will be immortalized in the CRD database and to handle the single diastereomer conundrum I will stick to IUPAC recommendation number two, that is add the reaction tag “and enantiomer”.