A team of researchers led by the University of Pennsylvania has built solid-state nanopores that can distinguish between short DNA homopolymers, a step towards sequencing DNA with this type of pore.
The researchers, led by Marija Drindic's team at UPenn in collaboration with groups at Columbia and Brown University, published their results online in ACS Nano earlier this month.
While they stressed that actual DNA sequencing will require additional improvements in sensitivity of the nanopore and the signal-to-noise ratio, the authors wrote that "homopolymer differentiation represents an important milestone in the development of solid-state nanopores."
"The protein nanopores still beat the solid-state nanopores by a country mile," said John Kasianowicz, a project leader at the National Institute of Standards and Technology, and a nanopore expert. "However, that doesn't mean the solid-state nanopores will never ever get there, and this paper represents the first significant step forward in that direction."
So far, biological nanopores – naturally occurring protein pores – have been leading the field of nanopore sequencing. Last year, researchers showed for the first time that they can read DNA with single-base resolution using a bacterial nanopore, MspA (IS 3/27/2012). In addition, Oxford Nanopore Technologies is developing a commercial sequencer that is based on protein nanopores (IS 2/21/2012), though the company also has a research group that focuses on solid-state pores and detection methods (IS 1/17/2012).
Many other research groups have also been concentrating on solid-state nanopores, using different materials, among them silicon nitride and graphene, and various detection methods, ranging from optical detection to measuring tunneling currents and ion current blockage.
According to the study authors, solid-state pores might be easier to manufacture and offer higher signal levels than biological pores, so they might no longer require the DNA passing through the pore to be slowed down, as is currently needed for biological nanopores.
In their study, the researchers showed that they can distinguish between single-stranded homopolymers – each 30 nucleotides in length – of adenine, cytosine, and thymine, which each produces a slightly different ion current signal.
"It's not perfect, there is still some overlap, but they are different," said Kasianowicz, who was not involved in the study, adding that the researchers still have issues with noise and cannot control the rate yet at which the DNA passes through the pore.
The scientists did not include guanine oligonucleotides because these form intramolecular quadruplex structures, but according to Gabriel Shemer, one of the lead authors of the study and a postdoc in Drndic's lab, the researchers hope to be able to distinguish guanine in the context of other bases in the future.
According to Drndic, who is a professor of physics and astronomy at UPenn, one of the main achievements that allowed them to obtain high signal levels was to manufacture small silicon nitride nanopores, with a diameter between 0.8 and 2 nanometers, in thin membranes between 5 and 8 nanometers thick – roughly the same size as biological nanopores.
While their fabrication method, which uses a transmission electron microscope, is not novel, they refined their protocols – which involve thinning a silicon nitride membrane first and then drilling a small pore into it – to make the pores smaller than before, thus increasing the signal.
The researchers then coupled these small pores with high-bandwidth electronics − an amplifier specifically developed for nanopore studies with collaborators at Columbia University and Brown University. "This is the first amplifier that's been developed for quick nanopore measurements, so that we can measure small molecules quickly without the need to slow them down," Shemer explained.
Last year, the teams published a different type of amplifier that is integrated with a nanopore on a chip (IS 3/27/2012). The one used in their latest study "is kind of off-the-shelf, you just plug it in and work with it, which is similar to commercial ones, but much better," Shemer said.
Being able to distinguish between different types of homopolymers is "a step toward sequencing" with solid-state nanopores, Shemer said, a step that researchers already achieved for biological nanopores 13 years ago.
"While this is an important milestone, we're just differentiating between different molecules, not within a single molecule," Drndic said. In order to achieve higher resolution, and to move to measurements within single DNA molecules, they will need to improve the pores and to increase the signal-to-noise-ratio further. "Lowering the noise, increasing the signal, that's the whole game," she said.
"This is interesting work," said Cees Dekker, a researcher at the Kavli Institute of Nanoscience at Delft University of Technology in the Netherlands, who is also working on nanopore sequencing. However, while the results are nice, "it likely will be challenging to really sequence DNA with this approach," he said. "A related approach, using solid-state nanopores based on graphene, may be more promising."
Both Dekker and Drndic, as well as others, have been working independently on nanopores made from graphene, which is only a single atom-layer thick, showing two years ago that they can thread DNA through graphene pores (IS 7/20/2010 and IS 8/3/2010).
But Kasianowicz said that the DNA passed way too fast through these graphene pores, so the researchers were unable to resolve even much longer, double-stranded DNA molecules.
"The cool thing about this pore is, for reasons that are unclear to me, it's able to slow the DNA down to the extent that you can actually detect the events," he said. "That's significant."
Drndic said her group is still pursuing graphene as a material for nanopores, but she noted that it has a number of challenges, for example its hydrophobicity. "Ultimately, you may not need a single-atom thickness," she said, so her lab is working on different approaches.
She declined to speculate how long it will take to show proof of principle for DNA sequencing using solid-state nanopores and ion current measurements, but she is optimistic that it will not take 12 years, the time it took researchers to move from distinguishing homopolymers to showing single-base resolution for biological nanopores. "We have the whole biological pore science that we can learn from, so we're not starting from zero here," she said, "so I think the progress, in principle, could go faster."