NEW YORK (GenomeWeb) – Researchers at the University of Oxford have demonstrated that optical sensing of DNA or RNA in nanopores can be highly parallelized, suggesting a path to high-throughput nanopore sequencing.
In a paper published online last month in Nature Nanotechnology, by the teams of Mark Wallace and Oxford Nanopore co-founder Hagan Bayley in the Department of Chemistry at Oxford, the scientists showed that they can optically measure the ion flux through many protein nanopores and discriminate between nucleic acids in those pores with single-base resolution.
In addition, they built an array of 2,500 bilayers, using micropatterned hydrogel chips, which could carry different types of nanopores or be used to measure many different samples in parallel.
The new approach might enable highly parallelized nanopore sequencing. Human genomes can already be sequenced within days using short-read technologies, Wallace said, but that is not fast enough for many applications, such as clinical use of the technology.
Current nanopore sequencing methods, including Oxford Nanopore's and others demonstrated by academic groups, measure the ion current through single isolated nanopores electrically and will be difficult to scale up to very large numbers of pores, the authors noted in their paper.
However, the ion current can also be measured optically, using an indicator dye on one side of the pore that becomes fluorescent when calcium ions flow through the pore and bind to it, a concept originally developed by Wallace's group in 2009. A year ago, two research teams in the US — one led by Amit Meller at Boston University, the other by Meni Wanunu at Northeastern University — already explored this approach for DNA sensing, using solid-state nanopores.
In their latest study, the Oxford researchers showed that they can optically measure the ion current from hundreds of protein nanopores in parallel, using total internal reflection fluorescence (TIRF) microscopy and so-called droplet interface bilayers (DIBs). "The use of this imaging in droplet interface bilayers gives us very high signal to noise, which is sufficient to resolve the small changes in blocking current that are associated with a single type of base," Wallace said.
"Each pore is a spot of light on an image," he explained, and "even without playing any additional tricks, we can resolve the location of those pores to within a few microns of each other." In their paper, the researchers made optical recordings at a density of 10,000 nanopores per square millimeter.
Initially, the researchers used the alpha-hemolysin nanopore to explore the limits of the technique. However that pore only allowed them to distinguish between oligonucleotides with or without a number of abasic sites.
To achieve single-base resolution, they switched to the MspA nanopore, which has been shown in the past to offer better resolution between the four types of bases than alpha-hemolysin. "When we switched to MspA, we were able to resolve the differences between G, T, A, and C in the system, purely using optical detection of the current," Wallace said.
In another part of the paper, they showed that they could distinguish between different microRNAs by hybridizing them to a DNA probe, forcing them to unzip and translocate through an alpha-hemolysin pore, and measuring the unzipping kinetics. Others had already shown that this was possible using electrical measurements. "We essentially took this idea … and translated it into a purely optical detection of many microRNAs in parallel," Wallace said.
Finally, they demonstrated that they could not only increase the number of nanopores measured in parallel but also the number of bilayers, by building an array of 2,500 bilayers from a micropatterned hydrogel chip.
Such a chip, Wallace said, could be used for screening applications, for example to test different DNA samples or to analyze the same sample with different types of nanopores. Such a strategy — measuring the same DNA with different nanopores — could also help to increase the accuracy of nanopore sequence data, he added.
The next step is to use the optical approach for actual nanopore sequencing. "Here, we're just showing that we can tell the difference between an A, C, G, or T on a piece of DNA in a static measurement," he said.
According to Wallace, it would only be a small step to DNA sequencing, using enzyme-based tethering methods to slow down the DNA, similar to what Oxford Nanopore and others have done. "Although there is scope for improvement, the paper shows that we don't actually have to do anything more than take what is already known to work electrically and apply it to this optical method," he said. "There are no major technical hurdles, other than an expansion of the complexity of the experiment."
His lab is currently working on obtaining DNA sequence information, and Wallace said he expects the first results "very soon."
Based on the results of their published study, the researchers estimated in their paper that they should be able to sequence DNA from an array of nanopores at a rate of a million bases per square millimeter per second someday, which could produce a human genome in as little as 15 minutes.
Those estimates are actually conservative, Wallace said, and "if we are willing to invoke the magic of super-resolution microscopy," the nanopore density could potentially increase by another factor of 1,000.
Wallace said there is intellectual property associated with the method, but he could not say whether it had been licensed to anyone for commercialization yet.
According to the paper, methods developed by Wallace have been licensed by Oxford Nanopore Technologies, but it was unclear whether this includes the techniques described in the article. The company did not respond to a request for clarification before deadline.