By Monica Heger
In the increasingly competitive field of nanopore sequencing, researchers from the Kavli Institute of Nanoscience in the Netherlands have made a big step forward, becoming the first group to demonstrate the translocation of DNA through a graphene nanopore.
The researchers said the study, detailed in Nano Letters earlier this month, represents an important step toward nanopore-based DNA sequencing. Their next step is to use the graphene as not only the membrane, but also as the electrodes, and to slow down the DNA so they can read each individual base as it passes through the pore.
"The biggest achievement here is that we are able to detect DNA molecules in a graphene membrane," Greg Schneider, lead author of the paper and a postdoctoral research scientist in Cees Dekker's lab at the Kavli Institute told In Sequence. Previously, researchers have shown that DNA can translate through solid-state nanopores, but until now, had not demonstrated it through a graphene nanopore.
"It's a really important step toward ultimately sequencing DNA using solid-state nanopores," said Bo Zhang, an assistant professor of chemistry at the University of Washington, who has also been working on building graphene-based DNA sequencers.
Graphene, a hexagonal one-atom-thick layer of graphite, is an ideal material to be used in a nanopore-based DNA-sequencing device, according to Schneider, because it is both a good electrical conductor and also thin enough that only one DNA base can fit inside the pore at a time as the molecule translocates. One problem with other materials used for solid-state nanopores, such as silicon, is that they are between 20 nanometers and 30 nanometers thick, and up to around a hundred bases can be present in the pore at once, so the electrical signal is the average of many bases, Schneider told In Sequence.
By contrast, one layer of graphene is only 0.3 nanometers thick, smaller than the distance between two bases of DNA. But graphene poses its own challenges, such as fabricating a single free standing layer.
In the Nano Letters study, the researchers made use of graphene's thinness to demonstrate that a double-stranded molecule of DNA could translocate through a nanopore. One key to building the device was a technique the team published in a previous study to create a free-standing single layer of graphene.
They first grew a layer of graphene on a silicon oxide wafer. They then coated the wafer and graphene with a hydrophobic polymer, and dipped the assembly into water. Because both the polymer and graphene are hydrophobic, and the silicon wafer is hydrophilic, the water wedged the graphene and polymer off the silicon oxide wafer, and the graphene floated to the surface. The graphene could then be positioned onto a support structure, to have nanopores drilled into it.
Zhang said that the wedging technique should be particularly useful for the field of nanopore sequencing. "It's a new way of placing graphene, and represents a very interesting step forward," he said.
The team experimented with different sized pores ranging from 5 nanometers to 25 nanometers in diameter. They then added a 48 kilobase pair double-stranded DNA molecule to one side of the nanopore and applied a 200 millivolt voltage across the membrane. As the DNA passed through, the team observed current spikes. Three different types of signals were observed, which the researchers determined were correlated with three different translocation events: nonfolded DNA passing through the pore, fully folded DNA — where the DNA is pulled through the pore from the middle of the molecule — and partially folded DNA.
The average translocation time for nonfolded DNA was 2.7 milliseconds, just slightly larger than the translocation time for a solid-state nanopore made from silicon nitride. The team cannot yet distinguish between each individual base, and in order to have an accurate readout, would need to first demonstrate translocation of single-stranded DNA.
Schneider said the group is now working on making a number of improvements to the nanopore. Aside from using single-stranded DNA, he said they also want to slow down the DNA, which will make it easier to read the bases. "If we can slow it down by a factor of 10 I think we are in business," he said, although did not discuss details about how they planned to slow down the DNA.
Zhang said controlling both the rate of translocation and the direction would likely be the most challenging step. "For each nucleotide, you only get a couple of microseconds to do the analysis," he said. "That's not long enough to get enough information."
Additionally, DNA translocation is characterized by randomness: "it can go both forward and backward, and the random motion is hard to control," he said.
Zhang added that the size of the nanopore would likely have to be reduced to 2 nanometers or less to obtain enough resolution.
Another, less important issue, Schneider said, is making sure the DNA doesn't stick to the graphene as it passes through the pore. Graphene is hydrophobic, and DNA has a tendency to stick to it, which can clog the pore. "We're working on optimizing the efficiency of sliding the DNA through the pore," he said, though he added that it didn't seem to be a very big problem.
Schneider added that the team will also be working to make use of graphene's electrical properties, and use the material not just as the nanopore, but as the electrical probes as well. To do that, he said they will have to create "two pieces of graphene separated by a nanodistance. Then connect those to a generator, and use it like two electrodes."
Other groups, such as Zhang's lab and Henk Postma's group from California State University Northridge, are also working to use graphene as both the membrane and the probes (IS 3/16/2010).
In the recent Nano Letters study, the team measured DNA translocation by measuring the flow of ions through the pore. Different bases block a different amount of ions. Ideally, though, "we want to use the electrical properties of graphene," Schneider said. "As DNA goes through the pore, we want to measure the electrical resistance of the DNA, instead of the ionic resistance," he said.
Schneider would not speculate about when the group might have a fully functional sequencing device based on graphene nanopores. "Getting a DNA strand through graphene is already something that is not that simple," he said. He also declined to comment on the potential for licensing the technology.
While the Kavli group is the first to demonstrate DNA translocation through a graphene nanopore, other groups are close on the researchers' heels. Slaven Garaj, whose team at Harvard University is also working on developing a graphene nanopore-based sequencer, has a study in review at Nature. Garaj said he could not comment on the Kavli study because of Nature's embargo policy.
In their study, which is already available from the arXiv e-print service, the Harvard researchers show that a layer of graphene, when immersed in ionic solution, takes on new electrochemical properties that make it a trans-electrode. The membrane’s effective insulating thickness is less than one nanometer, making graphene "an ideal substrate for very high-resolution, high-throughput nanopore-based single molecule detectors," the authors write.
Meanwhile, industry has also shown an increased interest in solid-state nanopore-based sequencing technology. Oxford Nanopore, which is focused on developing biological nanopores as its first commercial product, also licenses technology from Harvard University on developing sold-state nanopores. Startup NobleGen recently licensed technology from Boston University researchers who have developed an optical readout system, and said they plan to commercialize a solid-state nanopore sequencing device (IS 5/25/2010). And IBM recently partnered with Roche to develop and commercialize a nanopore sequencer based on its DNA transistor technology (IS 7/6/2010).