Recently I came across a number of attempts to explain the “handedness” of life – the fact that proteins consist only of L-amino acids – by the crystallization behavior of amino acids. The general idea is that something that happens at the transition between solution and crystal that favors one of the enantiomers over the other. I have to admit that it sounds a lot less esoteric than the idea that polarized starlight somehow influenced the formation of interstellar amino acids sufficiently to create such an effect, but frankly, I don’t buy it.
Take, for example, this paper recently published by F.L. Falcon on arxiv.org: He claims that L-Ala crystals are slightly larger than D-Ala crystals. In his scenario there is a saturated solution of racemic alanine in a hydrothermal setting, with a gentle temperature gradient. Forming Ala crystals are pushed by currents into colder areas, where the solution is even more saturated, and grow. But L-Ala crystals are larger, so they get pushed further into the cold regions and grow even more.
Then, Ostwald ripening takes place, enhancing crystal size further and depleting the solution of L-Ala, and that in turn leads to all D-Ala turned into L-Ala by epimerization. Neat, eh? Of course you see the problem immediately: Where’s all that alanine to come from to create a saturated solution? The stuff is highly soluble, on the order of 300 g/L depending on the temperature. This seems highly unlikely to me.
The other problem is the claim that L-Ala crystals are larger, based on X-ray diffraction data. Falcon thinks that parity violation influences the strength of hydrogen bonding, thus explaining the crystal size. I, on the other hand, can think of a number of possibilities much closer to home: Different growth conditions, traces of water, other impurities, temperature differences… As one colleague observed, it may be difficult to get this paper past the reviewers.
The second paper by Lee and Lin about the crystallization of aspartic acid actually made it to publication. In my opinion it is rather weak nevertheless, and this is why:
The authors get very excited about the fact that they did dissolve a mixture of D- and L-Asp crystals and, depending on conditions, sometimes got back a mixture of the separately crystallized enantiomeres and sometimes additionally found racemic crystals, which are thermodynamically more stable.
And yet, only a 10-min, 25 °C incubation treatment of the 1.0 mL equal volume mixture of an unsaturated aqueous solution of conglomerate aspartic acid (i.e., prepared by dissolving 5.0 mg of d-aspartic acid and 5.0 mg of l-aspartic acid in 1.3mL of water) and an unsaturated aqueous solutionof racemic aspartic acid (i.e., prepared by dissolving 10.0 mg of dl-aspartic acid in 1.7 mL of water) produced a mixture of conglomerate crystals and a racemic compound.
From this they boldly conclude
There was some sort of molecular complexation or “memory” existing in those aqueous solutions which could not be erased by instant mixing.
Yeah, right. My money, on the other hand, is on seed crystals getting everywhere and enthusiastic grad students trying again when it didn’t quite work. I could be wrong, though, so read it yourself.
Finally, what does all this have to do with the origin of life? The authors claim that something on early earth led to the separation of the racemic solution into a D/L-solution. The idea then seems to be that due to the low solubility, L-Asp could have acted as a template inducing other amino acids to assume the same configuration. Somehow.
I was rather enthusiastic about the whole idea, at least at the beginning. A discermible chemical mechanism explaining the chirality of proteins would feel far more satisfying to me than cosmic rays or pure accident, but crystallization, it seems, isn’t it…
Lee, T., & Lin, Y. (2010). The Origin of Life and the Crystallization of Aspartic Acid in Water Crystal Growth & Design, 10 (4), 1652-1660 DOI: 10.1021/cg901219f