Tiny holes have been drilled through individual layers of graphene — atomically thin sheets of carbon — using an electron beam. These nanopores might be useful for the ultrarapid sequencing of single DNA molecules.
The idea that DNA could be sequenced by running a strand through a tiny hole — a nanopore — and reading off the bases by electrical detection was suggested 14 years ago1. Since then, significant progress has been made towards this goal2, provoked by the US National Institutes of Health's $1,000-genome challenge3. Recent developments in base identification seemed to give the upper hand to protein nanopores. However, three papers4,5,6now report that nanopores fabricated from graphene — sheets of carbon only one or a few atoms thick — might have crucial advantages for this application.
Graphene7, a hugely extended aromatic molecule of fused six-membered carbon rings, is a material with extraordinary electrical and mechanical properties. On page 190 of this issue, Garaj et al.4 describe how they used an electron beam to bore holes ranging in diameter from 5 to 23 nanometres into graphene one or two layers thick. They then mounted the graphene in a chamber with an aqueous salt solution on each side of the film, and measured the current carried by the salt's ions when a voltage was applied across electrodes immersed in the solutions. The conductances of the nanopores scaled with their diameters, as expected for pores for which the thickness is much less than the diameter8. On the basis of the conductance values, the authors calculated that the effective insulating thickness of the graphene was only about 0.6 nm. This is much smaller than that of other materials that have been used for DNA analysis — 30 nm for typical silicon nitride pores9, for example, and 10 nm for α-haemolysin protein pores10.
Garaj et al. went on to measure the current carried by a graphene nanopore of diameter 5 nm while double-stranded DNA passed through it. They observed spikes in the current traces, which denoted current blockades corresponding to the transit of both folded and unfolded DNA. Similar blockades had previously been observed in analogous experiments with silicon nitride pores11. A basic solution of high ionic strength ensured that translocation of only a minority of the DNA molecules was hindered by adherence to the graphene surface. Using the mean amplitude of the current blockades, the authors were again able to calculate the effective thickness of the graphene film, confirming it to be about 0.6 nm.
The graphene used by Garaj et al.4 was prepared by a process known as chemical vapour deposition (CVD). Writing in Nano Letters, Schneider et al.5 report similar data for the translocation of double-stranded DNA through graphene nanopores, but using films that were made by exfoliation (the removal of graphene sheets from bulk graphite). They say that such films have fewer defects than those made by CVD.
In contrast to Garaj and colleagues' findings, Schneider and colleagues' data suggest that the conductance of the pores scales with the square of the pore diameter. This indicates that Schneider and colleagues' film was thicker than expected, perhaps because the authors coated it with 6-mercaptohexanoic acid to prevent DNA sticking to the graphene surface.
A third study of DNA translocation through graphene nanopores, also published in Nano Letters, is reported by Merchant and colleagues6. They worked with CVD-produced graphene that had a thickness of 3–15 atomic layers (rather than just one or two layers, as used by Garaj et al.4), containing pores 5–10 nm in diameter. Consistent with the other studies4,5, the authors observed current blockades associated with translocations of both folded and unfolded DNA, despite the high leak currents (perhaps caused by pinhole defects in the graphene) seen in this case.
So where do we stand with respect to the development of a graphene-nanopore device for DNA sequencing? In one manifestation of nanopore sequencing, single-stranded DNA would be sequenced by observing base-specific modulation of the ionic current as individual bases pass a recognition point in a pore1,2. A perceived advantage of graphene monolayers is that the entire thickness of a nanopore is comparable to the dimensions of a base, and might, therefore, form only one recognition point rather than participate in multiple contacts with DNA in the pore (Fig. 1). In the work considered here4,5,6, the double-stranded DNA moved at velocities of about 10 nanoseconds per base — too quickly to permit the resolution of current blockades arising from individual bases. Garaj et al.4 therefore resorted to computation to estimate the likely spatial resolution that could be achieved with a 2.4-nm pore at slower translocation speeds, and judged it to be as good as 0.35 nm — compatible with the identification of single bases.
But even if the translocation of single-stranded DNA through a pore can be slowed to a velocity of milliseconds per base, at which current blockades for individual bases should be measurable by ionic-current recording2, graphene-nanopore sequencing might still fall victim to one of several difficulties. For example, the graphene nanopores exhibit high levels of current noise4,5,6. This can be remedied, but only at the expense of thickening the device6. Furthermore, there is no experimental evidence that graphene nanopores will distinguish between different bases, and so the edge of the pore might need to be chemically modified to slow, pause and orient the translocating bases.
A second proposed means of base identification uses the quantum-mechanical phenomenon of electron tunnelling through DNA12, with graphene as a 'trans-electrode' (Fig. 1). Measurements of tunnelling currents or other electrical characteristics might allow extraordinarily rapid base identification at speeds of microseconds per base2,12,13. Crucially, both the ionic-current and tunnelling-current approaches for DNA sequencing are likely to need pores of 1.5 nm diameter or less (Fig. 1), much narrower than those so far described.
DNA sequencing using graphene nanopores will also undoubtedly require the application of new chemistry and physics. The nature of graphene surfaces, and especially that of the periphery of graphene nanopores, is poorly understood. Graphene surfaces are likely to be elastically corrugated14 and to contain various defects, such as those that cause the leak currents in the present work4,6. And what chemical groups are present around the periphery of graphene nanopores after exposure of the pores to air and water? Perhaps they resemble the mélange of groups found in graphene oxide15: carboxylates, hydroxyls, epoxides, alkenes, dienes and more. Or maybe the electron beam that creates the pores induces structural reorganization of the peripheral groups, generating new configurations such as five-membered rings16. Either way, are these peripheral structures susceptible to further rearrangement or hydrolysis, which would cause the nanopores to be irreproducible or unstable? And can they be covalently modified to facilitate base recognition?
Importantly, the surfaces of the graphene might need passivation to make them chemically inert, and the hole might need to be chemically modified independently of the surface — a taxing challenge. In terms of physics, ultrarapid DNA sequencing will require large arrays of electrically addressable nanopores. This challenge might be more readily tackled by using nanoscale gaps in graphene ribbons17, rather than nanopores in graphene sheets13.
Single-molecule DNA sequencing using arrays of nanopores offers the possibility to obtain genome sequences in less time than it takes to unravel a stethoscope. This potentially revolutionary technology will be unrivalled when numerous genome sequences are required from a single person — to personalize cancer treatment, for example. The first experimental steps4,5,6 taken with graphene nanopores suggest a compelling approach for clearing the remaining hurdles2 in the implementation of nanopore DNA sequencing.
Kasianowicz, J. J., Brandin, E., Branton, D. & Deamer, D. W. Proc. Natl Acad. Sci. USA 93, 13770–13773 (1996).
Branton, D. et al. Nature Biotechnol. 26, 1146–1153 (2008).
Schloss, J. A. Nature Biotechnol. 26, 1113–1115 (2008).
Garaj, S. et al. Nature 467, 190–193 (2010).
Schneider, G. F. et al. Nano Lett. 10, 3163–3167 (2010).
Merchant, C. A. et al. Nano Lett. 10, 2915–2921 (2010).
Geim, A. K. Science 324, 1530–1534 (2009).
Hille, B. Ion Channels of Excitable Membranes 3rd edn (Sinauer, 2001).
Kim, M. J., McNally, B., Murata, K. & Meller, A. Nanotechnology 18, 205302 (2007).
Song, L. et al. Science 274, 1859–1865 (1996).
Dekker, C. Nature Nanotechnol. 2, 209–215 (2007).
Thundat, T. Nature Nanotechnol. 5, 246–247 (2010).
Postma, H. W. C. Nano Lett. 10, 420–425 (2010).
Meyer, J. C. et al. Nature 446, 60–63 (2007).
Dreyer, D. R., Park, S., Bielawski, C. W. & Ruoff, R. S. Chem. Soc. Rev. 39, 228–240 (2010).
Chuvilin, A., Kaiser, U., Bichoutskaia, E., Besley, N. A. & Khlobystov, A. N. Nature Chem. 2, 450–453 (2010).
Wang, X. & Dai, H. Nature Chem. 2, 661–665 (2010).
Hagan Bayley is the founder, a director and a shareholder of Oxford Nanopore Technologies, a company engaged in the development of nanopore sequencing technology. Work in the Bayley laboratory at the University of Oxford is supported in part by Oxford Nanopore Technologies. Work on this News & Views article was not supported by Oxford Nanopore Technologies.
1. Hagan Bayley is the founder, a director and a shareholder of Oxford Nanopore Technologies, a company engaged in the development of nanopore sequencing technology. Work in the Bayley laboratory at the University of Oxford is supported in part by Oxford Nanopore Technologies. Work on this News & Views article was not supported by Oxford Nanopore Technologies. 2. Work in the Branton and Golovchenko laboratories at Harvard University is supported in part by Oxford Nanopore Technologies. The work by Branton, Golovchenko and colleagues discussed in this News & Views article was not supported by Oxford Nanopore Technologies.
About this article
2D Materials (2019)
An Overview of Molecular Modeling for Drug Discovery with Specific Illustrative Examples of Applications
Scientific Reports (2018)