Intriguing Results for Nanopore Sequencing from Defunct Roche-IBM Partnership
By Allison Proffitt
December 9, 2014 | In a paper published online last month, researchers from Arizona State University and IBM’s TJ Watson Research Center presented the development of a solid state tunneling device sensitive to nucleotides that could prove to be another assault in the nanopore wars.
“This is probably the first time that we are able to use a manufacturable device to sense the identity of the nucleic acid bases,” says Gustavo Stolovitzky, Program Director, Translational Systems Biology and Nanobiotechnology, IBM Research. “We can identify the A, C, G, and T using a device that is electrical, rather than the usual way in sequencing which is typically using fluorescence.”
“The resolution of this technology is much higher than the sort of ion current technology that people like Oxford Nanopore have developed,” says Stuart Lindsay, professor of chemistry and biochemistry at Arizona State University.
The paper, published online last month in ACS Nano (DOI: 10.1021/nn505356g), detailed the construction of a fixed-gap tunnel junction. On a silicon substrate, two 10 nm layers of lead are separated by a 2 nm insulating layer. A hole is drilled through all three layers to expose the electrodes. As a proof of principal, a solution of nucleotides was washed over the junction and nucleotides were identified.
The result was a solid-state device sensitive to molecules dissolved in solution; in this case a solution of DNA nucleotides.
The work represents a step toward solid state sequencing. “This would enable the possibility of having a sequencer in which the readout is an electrical signal,” says Stolovitzky. “[It is] the proof of principle that this could be done using techniques that are the same techniques that are used to make microprocessors… There’s no lipid bilayer; there’s no polymerase, there is no biological protein… The thinking is it will not degrade over time.”
“We were able to make these tunneling measurements in a device that can be fabricated like a computer chip, and therefore in quantity where the costs associated with making the device would be essentially nothing. Very, very low indeed,” explains Lindsay.
But there’s much work to be done before the device could actually sequence anything.
“You still have quite a few problems to solve,” points out Yann Astier, nanobiotechnology research scientist at IBM Research. “One would be to align the DNA and guide it to the recognition electrode and make sure each base sits in the gap between the electrodes one at a time. There’s a lot of engineering still required to get the DNA to behave and let itself be read by this device.”
Future of Nanopores
Those are next steps, however, that IBM won’t be taking.
Stolovitzky calls the paper “the natural conclusion” of the work started as part of a now-defunct partnership between Roche, IBM, and Arizona State University.
Roche and IBM announced a partnership in 2010 to develop a nanopore next-gen sequencing system. In 2011, Roche licensed technology out of Lindsay’s lab at Arizona State in addition to other ongoing projects, including Roche’s acquisition of 454 Life Sciences. In April 2013, Roche shuttered its 454 division and the rest of its next-gen sequencing programs and ended the partnership with IBM.
“When Roche changed strategies, we moved on, and we are right now doing other kinds of nanotechnology projects,” Stolovitzky says, “not necessarily DNA sequencing, but encompassing other areas.”
Meanwhile, Roche acquired Genia last June, getting back into the nanopore sequencing game, and has renewed its partnership with Lindsay’s group at Arizona State.
“Roche has rekindled a sponsored research agreement with us and we are working with Roche on developing this as a sequencing device,” Lindsay says, adding that DNA sequencing is “the primary interest of Roche.”
Lindsay agrees with Astier that the main challenge is to position the DNA in a way that will be readable. “The next challenge here is for whatever is needed to get the DNA to move past the junction… But to do that, we need put a very tiny hole—that’s one approach—through the tunnel junction and that’s what we’re working on now.
“Assuming the translocation part of the reading head is a problem that’s solvable—and it may not be, but let’s assume it’s solvable—the thing about this technology is it gives a very distinctive signal for each individual base, or… indeed for amino acids. It’s a much more sensitive method of reading than looking at the fluctuations of current through the pore, which is what the companies like Oxford Nanopore or indeed Genia do. In that sense, it has much greater potential for being a more versatile chemical and analytical tool and much more robust.”
While DNA sequencing is where Roche’s interests lie, Lindsay believes that the device may be useful for other applications earlier. Lindsay’s lab also focuses on proteomics, and he sees much promise for amino acid detection.
“It’s a solid state device that can be stamped out like a computer chip and does have an extraordinary sensitivity… You’ll notice there’s a cryptic comment in the paper that because this thing can detect single molecules, it obviously has the potential to be a single molecule equivalent of something like an ELISA assay. I think that’s something that could happen actually sooner than sequencing,” he says.
“I think DNA sequencing in one sense is already a quite mature field, with genomes costing less than $1,000 available anyway, and the thing that makes that all happen is the polymerase chain reaction,” Lindsay says. “I think for diagnostics physicians are really going to want to look at the individual proteomes, and of course there’s not PCR there. I personally feel that the impact of single molecule reading devices will be enormous once we’re in a state where… the use of individual proteomic profiles is feasible.”