Quote:

---

Although the information that can be crammed into a genome is remarkably complex, the molecule underlying it all obeys remarkably simple rules. In every creature we've studied, there are four bases that always pair the same way: A with T, and G with C. These combinations are dictated by basic chemistry. The two pairings form different numbers of hydrogen bonds but keep the spacing between the two strands of DNA constant.

But synthetic chemists haven't been willing to sit still and accept what nature has given them. It's possible to come up with a variety of similarly shaped molecules that can pair up in a way that keeps the spacing of DNA constant but forms base pairs that are physically distinct from those of A, T, C, and G. In many cases, the enzymes that copy DNA don't really care about the precise chemical details; as long as there's a successful base pairing, they'll incorporate it into the DNA.

As a result, biochemists have managed to build DNA and RNA with a variety of synthetic bases and get enzymes to work with them—at least in test tubes. Getting these synthetic bases into a cell and fully incorporated into the processes that keep the cell dividing has been a significantly larger challenge. But now, researchers have reported that they've gotten bacteria to maintain a small piece of DNA that includes these synthetic bases. The work is a bit limited, but it's an important proof-of-principle.

The new paper, released by Nature today, uses the synthetic bases shown at top. Rather than hydrogen bonding to keep the two strands of DNA together, these strands nestle up against each other through hydrophobic interactions. (Their close spacing allows both to have larger, two-ringed structures; normal base pairs involve one two-ring base paired with a single-ringed base.)

As is often the case, this work grew out of an earlier, failed attempt to get these bases into bacteria. The original idea was that these bases are hydrophobic enough to work their way across a cell's membrane. So you could just put some in the media in which you're growing your cells, and the bases would get inside and be used by the normal DNA metabolism.

Great idea, but it didn't actually work. The authors stuck with the same bases as they tried something different, though: testing out a bunch of different transport proteins that normally act to pull bases across membranes. The researchers eventually found some proteins from algae that worked. Normally, these transport proteins sit on the membrane of the chloroplast and pull bases in from the rest of the cell, but it turns out that they could also work when they were sitting on the outer cell membrane and pulling bases in from the environment.

Once the researchers got the bases inside the bacterial cell, it was largely a matter of synthesizing some DNA that included the synthetic bases and putting that in the bacteria as well.

The process of copying DNA normally involves several different enzymes, and the authors had only tested the synthetic bases with the one that does repair and cleanup of the main replication process. So they targeted their synthetic bases to a region they knew would be involved in cleanup. After letting the bacteria divide multiple times, they went back and looked at the region containing the synthetic bases and showed that it was successfully copied multiple times with well over 99 percent fidelity. Their bacteria were using artificial base pairs.

This is clearly an intermediate step in a much larger project. The next step will be testing whether these synthetic bases also work with the main DNA copying enzyme, which would allow them to be used much more widely. After that, you'd want to start getting them incorporated into RNA as well, which would eventually allow us to expand the genetic code.

The nice thing about this is that there are multiple pathways that could lead to success. If these particular bases don't work especially well with other DNA- and RNA-copying enzymes, researchers could always try some of the other synthetic bases that chemists have dreamed up; the transporters used in this research seem to be pretty flexible about what they'd import. Alternately, the DNA- and RNA-copying enzymes are pretty well studied, and we know of a number of mutations that decrease their finickiness. Introducing these mutations may help the bacteria use the synthetic bases more widely.

Finally, it should be possible to just stick these bases in a bacterial strain that's prone to mutations, then try to force the bacteria to use them. Whatever grows out has a reasonable chance of using the synthetic bases.

The new work isn't the end of the story, but it's a key indicator of progress, and the approach used makes me optimistic that it won't be a dead end.

---