DNA is pretty great for a lot of reasons. There is the whole “central component of life,” of course, but DNA is also a pretty great building material. Making super-tiny, nano-scale objects seems like a really hard problem, but biology does this all the time. Life’s standard nano-brick is protein, but proteins are difficult to re-engineer, whereas DNA is much simpler and more predictable. With only two pairs of bases and very predictable base-pairing rules, DNA structures are reasonably well-behaved, and DNA origami has been touted as a promising way to make useful nanoscale things from motors to boxes to smiley faces.
Impressive, but even this milli-crane still has about 5 orders of magnitude to go before hitting DNA origami scale. Image via Elfer, under the GFDL or CC-BY-SA-3.0 license
Recently, Jonathan Burns and Astrid Seifert, working in Stefan Howorka’s lab in London, figured out how to make designer DNA channels that let small molecules pass through membranes. If put in lipid vesicles, nano-pores like these could allow prolonged or selective drug release.
Normally, charged DNA is hydrophilic and stays away from the oily interior of a lipid membrane, but Howorka’s lab stuck their DNA pores into membranes by attaching cholesterol anchors to the 3’ ends of the DNA strands. As cool as these nano-pores were, they were still too tall to be truly stable in a bilayer, and they couldn’t open and close in response to a chemical signal. Now, the group has improved their channel to make it not only even more nano, but also DNA-sensing smart.
Right-Sizing the DNA Nano-Pore
The first step was to re-design the pore down to the 9 nm width of the lipid bilayer. The team did this using six single-stranded, 50 bp DNA oligos. Each oligo is like a chain link, with its 5’ and 3’ ends almost closed up on each other. Each link connects two neighboring links, and the whole bundle comes together like an open barrel.
The new nano-pore design not only self-assembled and inserted itself into membranes, but it also allowed current to pass, showing that it was indeed a membrane channel. The pore could be clogged with PEG molecules up to about 2 nm in diameter, showing that its interior channel was about that wide.
Making the DNA Nano-Pore Smart
With a more stable, membrane-width pore, the team next wanted to make the channel open and close in response to an external molecule. To do this, they extended two of the chain links with sequences complementary to a “lock” DNA strand, which would bind the extended links and cross over the top of the pore, thus blocking it. A third “key” oligo bound the lock, thus removing it and opening the channel. The closed pore still let some current through, but it very effectively blocked larger, biologically-sized molecules like sulpho-rhodamine. As expected, the key removed the lock and allowed the rhodamine through again.
Interestingly, the pore was selective for certain molecules. For example, a very similar dye with 3 negative charges rather than rhodamine’s 2 negatives and a positive, could not pass. Presumably, this is because DNA is negatively charged, so an all-negative molecule would be kept out by electrostatics.
Where Next?
This impressive advancement for designer nano-pores could be very useful for drug delivery. DNA is cheap and easy to work with, and the design is relatively simple, so it wouldn’t be surprising to see many other labs adapting the channel for their own problems. For example, using DNA aptamers for the “lock” could potentially let the channel respond to various small molecules. The authors even suggest using DNA analogs that are neutral or positively-charged to change the channel selectivity.
For more, you can pore over the original paper at Nature Nanotechnology, 2016.