On the 9th of May, Professor Syma Khalid presented a seminar on her research groupās efforts to develop and utilise computational models of biomimetic nanopores in order to sequence single-stranded DNA (ssDNA).
The research was performed in collaboration with Oxford Nanopore Technologies; a company who manufacture a variety of DNA sequencing devices that utilise biomimetic nanopores embedded in membranes; these range from high-throughput laboratory machines to more portable devices able to plug into a mobile phone or USB connection. Small machines such as these have revolutionised the way we look at DNA sequencing: what was once a
multi-million dollar venture can now be performed on a machine costing as little as $820; a price that comfortably satises the long sought after goal of the so-called ā1000 dollar genomeā project.
The nanopores are engineered to mimic the ability of the biological porins that control the ow of compounds into and out of the cell whilst retaining a much simpler structural composition than their biological counterparts. This simplied composition presents two main benefits: simplifying the manufacturing process for industrial purposes and reducing the complexity of the interactions within the pore, allowing for more accurate measurement of
the quantities used to sequence the ssDNA.
In order to understand how these nanopores are used to sequence DNA, we must first understand the composition of the ssDNA itself: ssDNA is comprised of four monomer units called nucleotides, numerous copies of these nucleotides are then joined together in a specic sequence that denes the function and characteristics of the cell in which it resides. Crucially, these four nucleotides have diferent shapes and sizes and thus, when placed inside a nanopore, will occlude the pore to difering extents. This means that if there is a constant ion flow through the pore in the absence of the ssDNA, we should be able to slowly feed the ssDNA through the pore, measuring the variation in the ion flow as each base passes through; there should then be four distinct magnitudes of measured ion flow corresponding to the presence of each of the four nucleotides.
Enacting this process of DNA translocation can be considered as three steps: rst the pore must capture the DNA from solution, the DNA must then enter the pore and must subsequently undergo controlled translocation through the sequencing region. This method is not without its complexities though; in order to relate the ion flow to the presence of a specic base in the pore one must ensure that only one base is present within the sequencing region of the pore at a time. Furthermore, the DNA must be fed through the pore at a sufficiently slow pace such that the variations in ion flow can be seen as distinct levels, as opposed to a constantly changing flux.
It was found that the choice of anchoring residues used to embed the pore into the membrane allowed for the control of the DNA translocation process; arginine, for example, provides an interaction site that explicitly binds with DNA. This hard binding, however, can induce kinks in the DNA conformation; leading to more than one nucleotide potentially entering the sequencing region at once. A more desirable interaction is to instead place charged residues along the inside of the nanopores which transiently bind with the DNA thus slowing down the translocation process without the risk of undesired conformational changes. Another possibility is the engineering of hydrophobic gates within the pore; these are regions that remain hydrophobic until an electric potential is applied across them, a process called electrowetting. These hydrophobic gates also have the effect of slowing down the DNA without inducing conformational changes.
It was found that hourglass shaped pores (i.e with a central constricted sequencing region) provided the means to enforce the optimal rate of ssDNA translocation, with a hydrophobic gate across this central constriction being necessary to avoid coiling of the ssDNA within the pore. The group obtained reliable measurements of the distinct ion flux levels, proving that such a system can indeed be used to sequence DNA. Further avenues of research lie in
performing umbrella sampling techniques to determine the location of the free energy barriers to permeation of the nucleotides, allowing for the further optimisation of the transient binding sites within the pore.
Posted by Iain Smith and Yen-Ye Soh