Signal Amplification in Electrochemical DNA Biosensors Using Target-Capturing DNA Origami Tiles

Electrochemical DNA (e-DNA) biosensors are feasible tools for disease monitoring, with their ability to translate hybridization events between a desired nucleic acid target and a functionalized transducer, into recordable electrical signals. Such an approach provides a powerful method of sample analysis, with a strong potential to generate a rapid time to result in response to low analyte concentrations. Here, we report a strategy for the amplification of electrochemical signals associated with DNA hybridization, by harnessing the programmability of the DNA origami method to construct a sandwich assay to boost charge transfer resistance (RCT) associated with target detection. This allowed for an improvement in the sensor limit of detection by two orders of magnitude compared to a conventional label-free e-DNA biosensor design and linearity for target concentrations between 10 pM and 1 nM without the requirement for probe labeling or enzymatic support. Additionally, this sensor design proved capable of achieving a high degree of strand selectivity in a challenging DNA-rich environment. This approach serves as a practical method for addressing strict sensitivity requirements necessary for a low-cost point-of-care device.


DNA Origami Tile Materials
All staple strands constituting the used DNA origami tiles were purchased from Integrated DNA Technologies and the employed M13mp18 scaffold strand was obtained from Tilibit Nanosystems.
50× stock TAE (Tris/acetic acid/ethylenediaminetetraacetic acid (EDTA)) buffer was purchased from Thermo Fisher Scientific (Finland) and molecular grade agarose from Meridian Bioscience (Ohio, US). All other chemicals required in the DNA origami assembly, purification and characterization were sourced from Merck/Sigma-Aldrich (Finland). Milli-Q deionized water was used in all procedures. DNA origami annealing was carried out in a Biometra T-Gradient thermocycler. Agarose gel electrophoresis was performed using a BioRad Mini-Sub Cell GT System with a BioRad PowerPac Basic power supply and imaged with a Bio-Rad ChemiDoc MP Imaging System.

Concentrations were measured with a BioTek Eon Microplate UV/Vis spectrophotometer and a Take3
micro-volume plate. Transmission electron microscopy (TEM) sample grids (FCF400-CU) were sourced from Electron Microscopy Sciences, treated with a NanoClean 1070, Fischione Instruments plasma cleaner and imaged using a FEI Tecnai 12 TEM.

Design and Capture Strands of the DNA Origami Tile
The full caDNAno [S1] design of the DNA origami tile [S2] is shown in Figure S1. The modification strands are highlighted in turquoise and pink, for the front and back sides of the tiles respectively. Figure S1. caDNAno design of the DNA origami tile.
The following extended strands are used to replace corresponding core staples in the DNA origami tile to create Tiles B and C. The core staple segment of the replacement strands are highlighted in turquoise and pink (See Figure S1), while blue denotes the used capture sequence. ''tttttt'' is a poly-T 6 spacer. Sequences are given from 5' to 3' (the first column indicates the start and end positions of the strand in the caDNAno design).

DNA Origami Assembly
The DNA origami tiles were assembled by first mixing a~10× molar excess of synthetic staple strands with a circular 7,249-nucleotide (nt) long M13mp18 scaffold strand (p7249) in 2.5× folding buffer (FOB: TAE buffer supplemented with MgCl 2 and NaCl). The resulting solution contained 20 nM of scaffold and~200 nM of each staple strand in 1× FOB (1× TAE (40 mM Tris, 19 mM acetic acid, 1 mM ethylenediaminetetraacetic acid (EDTA)) with 20 nM MgCl 2 and 5 nM NaCl, pH~8.5). A list of all replaced staples is given above. The mixtures were heated to 90 ℃ and slowly annealed according to the following thermal ramp: Cooling from 90 ℃ to 70 ℃ at -1.5 ℃/min, from 70 ℃ to 60 ℃ at -0.75 ℃/min, and from 60 ℃ to 27 ℃ at -0.05 ℃/min.
The mixture was then centrifuged at 14,000× g for 30 min at room temperature. After centrifugation, the supernatant was removed by pipetting and the remaining DNA origami pellet was dissolved in its original volume of 1× FOB. The solution was then incubated overnight at room temperature to resuspend the DNA origami tiles. Finally, the concentrations of the purified DNA origami solutions were determined with an UV/Vis spectrophotometer.

Agarose Gel Electrophoresis (AGE)
AGE was used to verify the integrity of the DNA origami tiles (main manuscript Figure 1b)

Transmission Electron Microscopy (TEM)
The fabricated DNA origami tiles were also imaged with TEM (main manuscript Figure 1b), based on a sample preparation protocol by Castro et al. [S4] A 3 µL droplet of~20 nM origami solution was deposited on an O 2 plasma cleaned (20 s flash) formvar carbon-coated copper TEM grid and incubated for 1 min. After incubation, the droplet was drained with a piece of filter paper and sequentially negatively stained with 2% (w/v) uranyl formate that contained 25 mM of NaOH. The grid was first immersed in a smaller 5 µL uranyl formate droplet, immediately drained with filter paper and then immersed in a larger 20 µL droplet before incubating for 45 s. After incubation, the grid was once more blotted with filter paper and left to completely dry in ambient conditions for at least 30 min before imaging with TEM. For imaging, a 120 kV acceleration voltage was used. See the        Table S3.  PGE that report mean signals for any one of the above that exist out with 1.5 interquartile range (IQR), were discounted and not carried forward for further experimental work. Electrochemical circuit fitting of Nyquist Data from EIS measurements is required to extract analytical parameters of solution resistance (R s ), charge transfer resistance (R CT ), and capacitance (C). The simplified Randles circuit was chosen for circuit fitting of electrochemical data (provided in Figure S5). Square Wave Voltammetry (SWV) interrogation of the sensor design was also explored, given its potential for enhancing signal gain reported in literature. However, SWV performance is a direct function of measurement frequency, and pulse amplitude. Both of which require close refinement for specific probe architectures, monolayer packing densities, and the electron transfer rates of the redox reporter. [S5] To date, we have yet to undertake such a study to explore the electrochemical parameters required to facilitate a sensing enhancement by SWV, and current recorded data shows no meaningful improvement against DPV interrogation. For simplicity, SWV analysis has not been reported here. It was necessary to assess the performance of the functionalization protocols. FE were subject to electrochemical interrogation in the redox buffer, following Measurement Script 1. Again, any data point existing out with 1.5 IQR was noted as evidence of abnormal functionalization and this electrode was discounted from further study.

Electrode Preparation, Electrochemical Measurement, Functionalization, and Target
Target Detection. The detection method of this approach is centered around the capture of a DNA origami tile/target complex from solution by an immobilized probe on the electrode surface. As such, it was first necessary to incubate a solution of both Tile A, B, or C and target to allow this complex to form. In this study, the tile was held at a fixed concentration (dependent upon particular experimental aim) against a varying target concentration. This complex was allowed to form by a 30 min incubation at 37°C. After which, FE were incubated directly in this solution for a further 30 min at 37°C.
Following all Target incubations, electrodes were rinsed in 1× PBS (phosphate-buffered saline) for 10 s, and gently dried under a steady stream of argon gas. They were then immersed in a redox buffer for the electrochemical characterization of sensing performance using Measurement Script 1 ( Table S3).
To confirm the applicability of this sensor construction in detecting key DNA targets of clinical interest, bacterial DNA sequences central to antimicrobial resistance are employed in assay development. Here, immobilized probes, and capture strands are primer sequences for the amplification of a region of an artificial plasmid attributing to the blaOXA-1 β-lactamase gene; encoding extended-spectrum β-lactamases (ESBLs) and resistance to Oxacillin, across a host of gram-negative species. This blaOXA-1 β-lactamase gene sequence (OXA Fragment) serves as the complementary target sequence in this study.
This mixture was not heated, to allow for minimal secondary structure formation. The below mixture (Table S4) was spiked during incubation stages, to provide a high concentration of background DNA.