Synthetic polymers self-assembling into catalytic structures

Modern organic chemists have access to a huge range of different chemical reactions and, these days, even the most complicated natural product could probably be synthesised if somebody wanted to. However, nature still does chemistry a hell of a lot better than even the best organic chemist; enzymes allow even the simplest organism to catalytically (and often asymmetrically) carry out organic reactions at room temperature in an aqueous environment. Being able an enzyme’s characteristics artificially would be, obviously, hugely advantageous. And there are a number of project on-going particularly using supramolecular chemistry, dendrimers or polymers. Another method, one of the first in a completely aqueous environment, has been described in a short communication in Angerwandte by the Palmans group. It utilises a simple polymer functionalised with L-proline allowing it to diastereoselectively catalyse an aldol reaction with cyclohexanone and p-nitrobenzaldehyde.

The group have successfully generated a polymer, P1a, that contains randomly distributed proline, PEG and BTA (benzene-1,3,5-tricarboxamide) residues. BTA polymers are known to stack into helical structures (due to intra-molecular hydrogen bonding), but are highly insoluble in water, the PEG containing branches therefore aid water solubility. Additionally the BTA component of the polymer be turned inwards, away from the water, while the PEG protrudes out into the aqueous medium; the hydrophobic interior, isolated from the bulk solvent, then acts as structure in which the proline catalysed aldol reaction can take place.

So how active is the polymer? Well it’s pretty good, conversion is about 30% after 24 hours with 94% d.r. and 70% e.e. After 120 hours conversion is up to 99% albeit with a slight reduction of d.r and e.e. Obviously, the nature of the polymer is crucial to understanding the reaction taking place. I haven’t worked in polymer chemistry, but the group do appear to have been very thorough in the analysis of their various polymer constructs with measurements of size (around 28 KDa for the P1a polymer), purity, polydispersity as well as NMR and UV analysis.

To examine their effects of polymers with different properties the group varied the amounts of BTA and proline as well as control polymers which tested whether the self-assembled structure is crucial for activity (by replacing BTA with a non-stacking alternative). This produced interesting results with some longer versions of P1a losing diastereoselectivity, but increasing reaction rate and conversion; others with additional hydrophobic groups were able to maintain some diastereoselectivity coupled with high conversion and reaction rates. Interestingly increasing the amount of proline reduced the observed d.r. Crucially, polymers which lacked BTA were inactive suggesting the polymer’s tertiary structure is key to the success of the reaction.

So what to make of this then? It is very impressive in a number of respects, but there are inevitably some shortcomings. The random nature of the polymer means you can’t get a single defined active site for a start and the d.r. and e.e. need to improve, but as a proof of the principle that simple polymers assembling to contain and control a chemical reaction it is really exciting. I would love to see some of these kinds of structures being used to do something other than an aldol reaction though. Obviously there are a number of good reasons for picking the aldol, it’s very reliable and you can do it asymmetrically for instance. However, in organic solvents several examples have been described using reactions not done typically by enzymes (see the links at the top). Surely this is the way this kind of technology has got to go, if there is one enzyme which is used very successfully in organic chemistry already it’s an aldolase; practically it’s always going to be a struggle to beat one of those. 

Finally, some arbitrary opinion and wild speculation (from me); the thing I always think about when I see stuff like this is: is this how life got going? If tiny amounts of a pretty simple polymer can catalyse organic reactions was that going on 3.5 billion years ago, on Earth, once you had some water to do your reactions in? There are people working in that area and although I don’t think that’s what Palmans’s research is about, maybe research into these kinds of polymers might provide some insights into how proteins originally evolved.