The concept is simple: thread a strand of DNA through a microscopic hole and detect each base-pair as it passes through. The potential advantages over current sequencing techniques are numerous: no PCR amplification, no chemical labeling, no optical instrumentation, long continuous reads, and a sub-$1000 genome. Since it was first described in a 1995 patent filed by sequencing guru George Church and colleagues (1), researchers have been keen on turning nanopore sequencing into a reality.
“Ordinarily if you just allow electrophoresis to drive the DNA through the pore, the consensus is with current technology that that’s too fast to resolve bases,” says nanopore sequencing developer Mark Akeson. “You want to slow it down enough to read base identity, but you don’t want to slow it down enough that it becomes impractical as a sequencing device.”
But now, a major funding agency has provided a major confidence boost to nanopore sequencing. When the National Human Genome Research Institute (NHGRI) awarded $14 million to support sequencing technology development earlier this year, most of the recipients were researchers working to overcome the major obstacle in nanopore sequencing: controlling DNA translocation through the nanopore. But the NHGRI has not placed all its eggs in one basket; each researcher has a unique idea to control DNA traffic through the nanopore.
Stuart Lindsay, director of the Center for Single Molecule Biophysics at Arizona State University, is one of nanopore sequencing’s strongest proponents. Although his previous research focus was on scanning probe microscopy (he founded a company called Molecular Imaging Corp. in 1993, which was later acquired by Agilent Technologies, Inc.), Lindsay’s research took an inspired turn a few years ago when micro-manufacturing techniques that could create nanopores began to emerge. He realized that these nanopores had the potential to allow researchers to study the physics of biomolecules in an unprecedented way.
“The single most important aspect of nanopore sequencing is that you ought to be able to get very long continuous reads of DNA sequence. It’s already well-established that you can pass a good 10 kilobases or so of single-stranded DNA though a nanopore. If this is your fundamental read length, the whole problem of constructing libraries for assembly and so on pretty much goes away,” says Lindsay.
Specifically, his design uses electrodes positioned at the opening of a metal or carbon nanotube to measure DNA translocation. In a 2010 paper published in Science (2), Lindsay and colleagues reported the creation of single-walled carbon nanotubes that could measure the transport of single-stranded DNA by means of its ion signature.
“Our technique, if possible, will be completely physical. It will be a direct electronic read-out of the DNA sequence, and the device itself will look more like a computer chip than something you’d see in a wet chemistry lab,” says Lindsay.
But as with most nanopore technologies, the control of DNA translocation through the nanopore remains a problem for Lindsay. So, the NHGRI has awarded Lindsay $4 million to continue developing his solution, which is called recognition tunneling.
In Lindsay’s recognition tunneling concept, two recognition molecules are covalently attached to the electrodes at the opening of a nanopore. As each base-pair exits the nanopore, they form a noncovalent bond with these recognition molecules, producing a unique electronic signature for each type of base that can be measured by the electrodes. With a small force, these noncovalent bonds can be broken, allowing the next base-pair to be analyzed.
With his technique, the DNA can be slowed down enough to be accurately read, while still maintaining lightning fast DNA sequencing. Using a preliminary version of recognition tunneling, Lindsay has read small fragments of DNA at a rate of 10–100 bases per second.
Recently, Lindsay’s nanopore technology has also caught the interest of commercial partners. In October, Roche licensed Lindsay’s nanopore technology with plans of incorporating his technology with IBM’s silicon nanopore array technology for DNA sequencing. The hope is that the collaboration will result in a commercial nanopore sequencing platform within a couple of years. “The fact that businesses are interested implies that the timescale for this is reasonable,” says Lindsay.
Polymerase Traffic Control
At the University of California, Santa Cruz, Akeson has a different idea to put the brakes on the speeding DNA: natural enzymes. For the past 12 years, Akeson’s team at the biophysics lab of the university’s Center for Biomolecular Science and Engineering has used a protein called a-hemolysin to develop nanopore sequencing. a-hemolysin is a pore-forming toxin used by the bacterium Staphylococcus aureus to obtain nutrients from host cells. But his group is also exploring the use of other protein nanopores as well.
“Part of our grant is to begin working with Jens Gundlach at the University of Washington who has developed the MspA pore from the Mycobacterium smegmatis,” says Akeson.
Similar to Lindsay, Akeson has also received an award from the NHGRI to advance his nanopore technique. But Akeson’s $3.5 million grant is supporting his work on optimizing processive enzymes to control DNA translocation.
“One of the great things about biology is that we have enzymes and enzyme complexes that replicate DNA very efficiently and move it at 3–5 angstrom precision and can do it tens of thousands of times,” says Akeson.
In short, Akeson’s approach exploits a viral DNA polymerase to control the movement of DNA through a nanopore. In a 2010 paper published in the Journal of the American Chemical Society (3), Akeson and colleagues reported that the bacteriophage f29 DNA polymerase was stable in the electric field of an a--hemolysin nanopore. The polymerase efficiently controlled the speed of the DNA strand, allowing for the detection of continuous base-pairs. The read lengths were only limited by the length of the DNA template.
With the proof-of-concept paper published, Akeson hopes to improve the resolution of his nanopore sequencer. To increase the signal differences from each different base, his team is now exploring increased salt concentrations. This will, in turn, require a DNA polymerase that is tolerant of high-salt environments. Working with other researchers at his university, Akeson will be investigating different polymerases from salt-tolerant organisms.
And just like Lindsay, commercial vendors have already begun collaborating with Akeson. Specifically, Oxford Nanopore, Inc. has supported the development of Akeson’s technology and has licensed it for the development of company’s GridION instrument. Although the company has not yet set a launch date, they have reported that the instrument is also showing promising results for electronic protein analysis.
Speeding Up Detection
“We have this transistor tip of a field effect transistor (FET) that we’ll build close to the nanopore. The FET device is able to sense things at much, much faster rates than any other sensor,” says Takulapalli.
FETs are semiconducting sensors that can detect electrical fields on their surface. As the exterior electrical field changes, the current within the FET conducting channel also changes. Because of their high-input impedance and low-output impedance, FETs are commonly used in electronics as a signal amplifier.
The concept of FET-based DNA sequencing is certainly not unprecedented. In 2004, South Korean researchers created a FET sensor that read DNA sequences by detecting the different electrical charge of each base (4). In 2006, two Japanese researchers also reported that label-free DNA sequencing could be performed using a FET to detect the molecular charges (5). However, neither of these devices were ever further developed for high-throughput sequencing.
With a $900,000 grant from the NHGRI, Takulapalli has begun laying out the blueprints for his FET detector, which will be the size of about a 1/1000 of a human hair and will read bases at the rate of 1 billion cycles per second. At those speeds, Takulapalli’s detector could possibly sequence an entire human genome in as little as 15 minutes. If it works as well as his physics on paper tells him it should, his technology will surely be revolutionary.
“Using that high speed if you are able to sense at 10–100,000 bases per second, it could be brilliant,” says Takulapalli. “That’s what my project aims to do.”
But Takulapalli’s device remains in the design phase. In a 2010 paper published in ACS Nano, Takulapalli reported a molecular sensing concept that would have high sensitivity and selectivity using a FET device (6). But such a device remains theoretical for now.
In the end, Takulapalli expects nanopore sequencers will begin infiltrating the market within the next 5–6 years, with his technology following sometime afterwards to further boost sequencing speed. After that, Takulapalli says that DNA sequencing might only require a trip to your local pharmacist.
“Right now, they have one-hour photo. Maybe in 10 or 15 years, they’ll have a one-hour genome sequence there,” says Takulapalli.
- Church, G.M., D.W. Deamer, D. Branton, R. Baldarelli, J. Kasianowicz. 1998. Characterization of individual polymer molecules based on monomer-interface interaction. US patent # 5,795,782 (filed March 1995).
- Liu HT*, He J*, Tang JY, Liu H, Pang P, Cao D, Krstic P, Joseph S, Lindsay SM, Nuckolls C (2010) Translocation of Single-Stranded DNA through Single-Walled Carbon Nanotubes. Science 327: 68.
- Lieberman KR, Cherf GM, Doody MJ, Olasagasti F, Kolodji Y, Akeson M. 2010. Processive replication of single DNA molecules in a nanopore catalyzed by phi29 DNA polymerase. J Am Chem Soc. 132:17961-72.
- Kim DS, Jeong YT, Park HJ, Shin JK, Choi P, Lee JH, Lim G. 2004. An FET-type charge sensor for highly sensitive detection of DNA sequence. Biosens Bioelectron 20:69-74.
- Sakata, T. and Y. Miyahara. 2006. DNA sequencing based on intrinsic molecular charges. Angew Chem Int Ed Engl. 45:2225-8.
- Takulapalli, B.R. 2010. Molecular sensing using monolayer floating gate, fully depleted SOI MOSFET acting as an exponential transducer. ACS Nano. 4:999-1011.
Stuart Lindsay is one of nanopore sequencing’s strongest proponents. Source: The Biodesign Institute at Arizona State University.