Skip to main content
Premium Trial:

Request an Annual Quote

Viscosity Can Slow Down DNA Translocation Through Solid-State Nanopores


NEW YORK (GenomeWeb) – Increasing the viscosity of the medium through which DNA travels in a solid-state nanopore sequencing device can slow it down by two to three orders of magnitude, which is enough to discriminate individual nucleotides, according to a research team from the École Polytechnique Fédérale de Lausanne.

The study, which was published this week in Nature Nanotechnology, builds off previous work by the same group, in which they demonstrated that molybdenum disulfide (MoS2) was a good material for solid-state nanopore sequencing.

Developers of nanopore sequencing technology typically work with either biological nanopores, which are used in Oxford Nanopore's MinIon system, or solid-state nanopores fabricated from materials used by the electronics industry. Although biological nanopores are further along in commercial development, solid-state nanopores could still provide some advantages. Production of solid-state nanopores is more scalable, and the pores do not need a specific pH or specific biochemical reagents in order to operate. However, one of the main challenges has been to slow down the speed at which DNA travels through the pores while maintaining a good signal-to-noise ratio in order to distinguish the unique current traces from each individual base.

In this recent Nature Nanotechnology study, the researchers used viscosity to slow down the rate of translocation to within a range of 1 to 50 nucleotides per millisecond, depending on the size of the pore, compared to typical solid-state nanopore speeds of 3,000 to 50,000 nucleotides per millisecond.

Aleksandra Radenovic, assistant professor in the laboratory of nanoscale biology at EPFL and senior author of the study, told GenomeWeb that while the idea of using viscosity to slow down DNA translocation was not new, previous groups have primarily worked with glycerol or other water-based liquids, which "decreases the signal-to-noise ratio," she said.

Radenovic's team instead used room-temperature ionic liquids, which have also been used to extract DNA from blood, and that have "300 times the viscosity of water," she said.

First, the team built single-layer MoS2 nanopores. The advantage of building a nanopore with this material, is that it is thin — only one nanometer thick — which reduces the number of bases that can be present inside the nanopore, and it is more sensitive than silicon nitride pores with a signal-to-noise ratio greater than 10. In addition, DNA does not stick to the material, which has been a problem with other types of solid-state pores.

Next, the researchers established a viscosity gradient separated by the MoS2 membrane. On the side of the membrane containing the DNA, they used a room temperature ionic liquid called 1-Butyl-3-methylimidazolium hexafluorophosphate, or BmimPF6. On the other side of the membrane was a potassium chloride water solution.

Having two different solvents on either side of the membrane established a gradient, including a region in and around the pore that included a mix of the solutions.

Previous attempts to use glycerol to increase viscosity were not successful because glycerol has a low conductance, so researchers were only able to realize slight improvements to DNA translocation speeds — down to about 3,000 nucleotides per ms. By contrast, "it was possible to use pure BmimPF6 without compromising the conductance of the MoS2 nanopore," the authors wrote.

As a first test of the system, the researchers created larger a pore of 20-nm in diameter to characterize translocation without the effects of a smaller pore size, which also helps to slow DNA down.

They first tested 48.5-kb double-stranded DNA molecules using MoS2 pores in their viscosity gradient system and compared them to MoS2 pores in the standard KCl solution. Average translocation time was 130 ms in the viscosity gradient system and 1.4 ms in the KCl solution, about a two orders of magnitude improvement. In addition, there was no reduction in the signal amplitude.

The researchers also created 2.8-nm pores and tested 30-mer homopolymers, demonstrating that the different nucleotides each generated unique signal traces.  Next, they translocated single nucleotides through the pore. They performed eight separate experiments, collecting more than 10,000 events in the same pore. Before each experiment, the researchers flushed out the analytes from the previous experiment.

The researchers again saw distinguishable current traces from each of the single nucleotides.

"We've shown that we can identify single nucleotides," Radenovic said, and the next challenge will be to demonstrate direct sequencing from a single strand of DNA and figure out how long of a strand the system can sequence. "That's something we're working on," she added.

Other research groups have tested mechanical methods for slowing down the speed at which DNA races through solid-state nanopores, like optical or magnetic tweezers and DNA hairpins, while some have turned to lasers to slow down DNA. 

However, Radenovic said that the viscosity method has a number of advantages over these others.

For example, methods involving optical tweezers are "not scalable to chip production," she said. Her group has worked with optical tweezers to study DNA properties and proteins that are bound to DNA, but said that they are not ideal for sequencing applications.

Lasers can complicate the system, while the viscosity gradient is simpler and scalable, she said. In addition, she said the viscosity set-up should be compatible with all types of nanopore devices, even protein nanopores.   

Radenovic said her group plans to continue to develop the MoS2 system using the viscosity gradient to slow translocation and eventually hopes to commercialize it in collaboration with a commercial partner. "We would like to show our platform is comparable or better than biological pores," she said, and then "potentially entice a commercial developer to go with the product."