By Monica Heger
Single-stranded DNA behaves completely differently than double-stranded DNA as it translocates through a nanopore, researchers reported last month in Nano Letters — a finding that could have implications for nanopore sequencing.
Most researchers developing nanopore sequencing methods are looking to use single-stranded DNA to ensure that the correct base is being read as it passes through the nanopore, yet because it is trickier to work with, much of the work characterizing nanopores themselves has been done with double-stranded DNA. In this new study, researchers from the Kavli Institute of Nanoscience in the Netherlands found that ssDNA moves more slowly than dsDNA as it travels through the nanopore, but that the translocation time increases exponentially with increased applied voltage and also produces a greater than expected blockage current.
"They managed to unravel the translocation mechanism quite well," said Henk Postma, assistant professor of physics at California State University, Northridge, who is working on developing a nanopore sequencer based on graphene nanogaps (IS 3/16/2010). "It's really quite neat, and I was rather surprised to see it."
The team first constructed nanopores around 8 nanometers in diameter in a silicon nitride membrane, sandwiched between two compartments containing a salt solution and buffer.
In nanopore sequencing, when an electric field is applied across the membrane, ions flow through the pore. But if DNA is passing through the pore, the flow of ions is temporarily stopped, producing a blockade current, and in theory, each base generates a different one. In order to ensure that the correct base is being read, it's necessary to use ssDNA, because it would be difficult to distinguish the orientation of base pairs. However, working with ssDNA is trickier than dsDNA because it tends to fold itself up. So, the researchers in this experiment wanted to characterize how ssDNA behaved as it moved through a nanopore.
The team built constructs of dsDNA with ssDNA tails of three different lengths — 3.7, 1.9, and 0.7 kilobases. Using atomic force microscopy, they observed that the ssDNA tails hybridized to themselves, forming complex structures that would have to unravel before passing through the nanopore.
As the hybrid molecule passed through the nanopore, the researchers observed different levels of current spikes, which corresponded to the double-stranded and single-stranded portions of the molecule. They found that the double-stranded end entered the pore first about 80 percent of the time, and produced a lower current spike than the single-stranded end. As the length of the single-stranded tail increased, the amplitude of the current increased.
The larger signal and translocation time was due to the structure of ssDNA, which hybridizes with itself, forming a complex, three-dimensional structure. Unlike dsDNA, which produces a constant blockage current even as applied voltage is increased, the blockage current for ssDNA increases as voltage increases. "At higher voltages, the ssDNA blob (which does not fit through the pore without unraveling) is pushed closer to the pore due to the larger electrophoretic forces, thereby increasing the access resistance," the authors wrote. "Clearly, the physics in both cases is different."
Next, the group tested single-stranded viral DNA. They translocated the DNA through a 7-nanometer pore and observed peaks that were around 10-fold larger than expected based on the volume of DNA, again due to the fact that the ssDNA hybridizes with itself.
Stefan Kowalczyk, lead author of the paper, and a doctoral student in Cees Dekker's biophysics laboratory at the Kavli Institute, said the findings were promising for nanopore sequencing, because the larger blockage spikes and longer translocation time would help reduce signal-to-noise and make it easier to detect individual bases.
Stuart Lindsay, director of the center for single molecule biophysics at Arizona State University, said he was surprised that the ssDNA was able to unwind at a neutral pH. "One might expect that the structures would be so complicated that knots might be encountered that would never unwind. Evidently, this is not the case," he wrote in an e-mail.
However, he added that the complex structures could still be problematic in trying to read the bases. "Regions that form hydrogen-bonded tertiary structures stick at the entrance, moving through more slowly than less complicated regions. One would like a uniform translation rate to read sequence," he said.
Kowalczyk said their next steps will be to increase translocation times even more and to make the pore thinner, in order to make the nanopore more sensitive at detecting individual bases.
Postma added that while the work is an important step, it is still a generation or two away from sequencing. "This is where we need to go with single-molecule nanopore sequencing," he said. "It's important that we understand the mechanism, and this work puts that on more solid ground."