Over two decades ago, scientists created the first artificial ion channel using channel-forming proteins – known today as the nanopore.1 The first foundational demonstration of this phenomenon consisted of a protein nanopore embedded in a lipid membrane, acting as a low cost, rapid method to sequence DNA.1 Today, nanopores are often associated with sequencing – likely due to the immense success of companies like Oxford Nanopore. However, nanopores can also be made of inorganic materials with tuneable diameters and surface properties, presenting new opportunities in the fields of biosensing and diagnostics.
Whether used for DNA sequencing or sensing, nanopores rely on the principle of resistive pulse sensing. Molecules are driven through the nanosized opening between two chambers of salt solution using forces induced by electric fields. Upon applying an electric field across the nanopore, the voltage drives electrically-charged molecules, like DNA, through the nanopore. As molecules move through the pore, the liquid containing the salt ions is displaced. The drop in liquid volume in the nanopore correlates to an increase in resistance and thus a drop in current. This current drop can provide information about the charge, molecular weight and conformation of the analyte.2
Nanopores can also act as an ideal platform for assaying CRISPR-Cas enzymes. The Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) and CRISPR-Associated (Cas) protein system is a sequence-specific RNA-guided protein duo that enables binding to a specific DNA sequence and subsequent cleavage. The CRISPR-Cas system has recently emerged as a revolutionary and widely employed gene editing tool3,4. In 2019, Weckman et al demonstrated the ability to observe bound dCas9 in nanopores, providing the inspiration for the use of dCas9 as a diagnostic tool for clinical DNA samples. For use in diagnostic assays, however, we first needed to benchmark the specificity of the guides we were designing, in order to make this method sensitive and targeted. Thus, we created a DNA nanostructure to test this.
As physicists and engineers delving into the biology realm, we learned more and more about the limitations of the systems along the way. Despite immense research in this field, there are still aspects of the mechanism that are not well understood, meaning that the parameter space for the design of specific and efficient guide sequences is poorly outlined. I learned this the hard way when I started to realize, upon analysis of my nanopore data, that the guides had a mind of their own - While some guides would bind 80% of the time to the target sequence, others wouldn’t bind at all!
One major limitation with using the CRISPR-Cas toolset for gene editing is the off-target effects. Although guides are sequence specific, some are more selective than others. The human genome is huge and because of this, the 20 base pair region that the guides target can be very similar to other regions of the genome. In some cases, only one or two of the base pairs differ. Throughout this project, we realized we were not creating a control just to check the specificity for ourselves, but were in fact creating a powerful approach for scientists across different fields using CRISPR-Cas systems to design specific guide sequences.
Currently, testing the guides is somewhat limited to high-throughput screening in cells or gel electrophoresis techniques. However, using DNA nanostructures, which can be designed to include CRISPR guide target sequences and a DNA 'barcode', which allows for multiplexed readout of the molecules, one is able to create ‘an assay for an assay’ as a way to observe specificity of the guides. The signals generated by these nanostructures in the nanopore are unique and provide information about the binding – as seen in Figure 1.
Fig. 1. DNA nanostructures for assessing dCas9 specific binding to target sequences. A) Schematic of DNA Structure with DNA barcode region and dCas9 overhang. Both the sequences used for the dumbbell region that creates the barcode (blue) and the schematic of the target DNA region which acts as an overhang (green) are highlighted. Additionally, the components involved in the dCas9 RNP are shown. B) Solid-state nanopore with two different DNA nanostructures (11111 and 11001) mixed together in solution, both these structures are shown with and without dCas9 binding, highlighting the capability to determine specificity and binding efficiency. C) Current traces from the nanopore of the different DNA nanostructures from the nanopore when two probes are added.
Using this system, we were also able to look at how specific the guide RNA can be to mismatches and the effect that the position and identity of the mismatch have. While the CRISPR-Cas system is an extremely powerful tool, it does have its limitations, with the most significant one perhaps being the ease with which one can make a poorly-designed guide. Because our nanostructures allow scientists to directly probe the behaviour of guides, experimentalists can then develop guides sensitive enough to distinguish between samples with single base pair changes. This is particularly important in diagnostics, where a single base pair change in a gene can be the difference between resistance and vulnerability to antibiotics. Another key area of diagnostics is DNA concentration quantification. In diagnostics, concentration measurements are relevant for determining the severity of the disease. Using this technology, we also demonstrated that we can obtain a relative measurement of the concentration of DNA present in a sample.
Through using the combinations of the three technologies – nanopore sensing, DNA nanotechnology and CRISPR/Cas molecular engineering, we are able to develop a tool to measure binding efficiency of guides and their specificity providing key insight into the fundamental behaviour of guide sequences. By showing that the protein complexes are difficult to predict when it comes to specificity, due to what is called guide-intrinsic mismatch tolerance, we demonstrate the necessity of testing CRISPR guides before you can use them in the many applications that are out there!
1 Kasianowicz, J. J., Brandin, E., Branton, D. & Deamer, D. W. Characterization of individual polynucleotide molecules using a membrane channel. Proceedings of the National Academy of Sciences 93, 13770-13773 (1996).
2 Wanunu, M. Nanopores: A journey towards DNA sequencing. Physics of life reviews 9, 125-158 (2012).
3 Doudna, J. A. & Charpentier, E. The new frontier of genome engineering with CRISPR-Cas9. Science 346 (2014).
4 Ran, F. A. et al. Genome engineering using the CRISPR-Cas9 system. Nature Protocols 8, 2281-2308, doi:10.1038/nprot.2013.143 (2013).