Synthetic biology seems to have taken another big step. Many labs over the years have tried out expanding the genetic code in various ways, but all these run in various in vitro systems. Now the first organism has been engineered with a working unnatural base pair, according to this paper in Nature from the Romesburg group at Scripps.
The base pair in question is d5SICS and dNaM, shown at left, and a history of how they were developed is here. This class of interaction was found by screening thousands of possible combinations, and it’s notable that there’s no hydrogen bonding going on between the two residues. (It’s worth keeping in mind that the current AT/CG base pairing system was presumably also arrived at by screening a wide variety of candidates until something worked!)
There are a number of tricky steps needed to get this to work:
However, expansion of an organism’s genetic alphabet presents new and unprecedented challenges: the unnatural nucleoside triphosphates must be available inside the cell; endogenous polymerases must be able to use the unnatural triphosphates to faithfully replicate DNA containing the UBP within the complex cellular milieu; and finally, the UBP must be stable in the presence of pathways that maintain the integrity of DNA.
A transporter spliced in from algae can bring in the unnatural triphosphates, as it turns out, but a next step would be getting enzymatic machinery inside the cell to make them. But the existing enzymes can handle them once they’re available, and replicate plasmids containing these pairs, which also don’t get tagged as DNA errors and snipped out by any of the endogenous repair mechanisms. So another bridge has indeed been crossed.
Romesburg has started a company, Synthorx, to try to take advantage of the chemical biology possibilities in this work. (I realize that I’m probably supposed to think “Syntho-Rx” when I see that, but my brain persists in saying “Syn-thorks”.) I can imagine, down the road, some very interesting assay development possibilities that follow from this technique, with what might be very high signal/noise ratios, so this is worth keeping an eye on.
Update: a criticism of the press coverage of this paper, which has indeed not been very well informed.