DNAzyme 10-23 - Based Nanomachines for Nucleic Acid Recognition

Summary

The DNAzyme-based nanomachines can be used for highly selective and sensitive detection of nucleic acids. This article describes a detailed protocol for the design of DNAzyme-based nanomachines with a 10-23 core using free software and their application in the detection of an Epstein-Barr virus fragment as an example.

Abstract

DNAzyme-based nanomachines (DNM) for the detection of DNA and RNA sequences (analytes) are multifunctional structures made of oligonucleotides. Their functions include tight analyte binding, highly selective analyte recognition, fluorescent signal amplification by multiple catalytic cleavages of a fluorogenic reporter substrate, and fluorogenic substrate attraction for an increase in sensor response. Functional units are attached to a common DNA scaffold for their cooperative action. The RNA-cleaving 10-23 DNMs feature improved sensitivity in comparison with non-catalytic hybridization probes. The stability of the DNM and the increased chances of substrate recognition are provided by a double-stranded DNA fragment, a tile. DNM can differentiate two analytes with a single nucleotide difference in a folded RNA and a double-stranded DNA and detect analytes at concentrations ~1000 times lower than other protein-free hybridization probes. This article presents the concept behind the diagnostic potential of DNA-nanomachine activity and overviews DNM design, assembly, and application in nucleic acid detection assays.

Introduction

One of the earliest methods for nucleic acid detection is complementary binding of selective oligonucleotides to the analyzed RNA or DNA sequences. This method employs nucleic acid fragments (probes) that can form Watson-Crick base pairs with a targeted DNA or RNA analyte¹. The complex is then differentiated from the analyte and the unbound probe by a variety of techniques. Certain methods, such as Northern blotting, in situ hybridization, or qPCR, are based on the phenomenon of complementary binding². The most common hybridization probes have two significant drawbacks: low sensitivity (they can reach micromolar to nanomolar levels) as well as low selectivity to single nucleotide variations (SNV), including point mutations. The issue requires a sophisticated approach since the solutions to these drawbacks conflict with each other³. To provide the best selectivity, the detecting oligonucleotide should be kept short enough to form unstable hybrids with mismatched analytes. Such short oligonucleotides are not able to bind targeted nucleic acids tightly and unwind their secondary structures, thus often failing to provide a detectable signal. It is especially challenging to detect CG-rich fragments in biological RNA, or double-stranded DNA (dsDNA). On the other hand, long probes are insensitive to SNV³. To break the deadlock, the concept of a binary probe has been developed⁴.

Binary sensors split a single detection probe into two pieces and specialize its fragments, keeping one of them long to unwind the hairpins or invade the dsDNA. The second binary probe fragment remains short to be sensitive to a mismatched base, thus providing a means for detecting SNV. The signal will be produced if the two binary sensor parts are joined together⁴. The full complex can be visualized via FRET⁵, molecular beacon probe^6,7, or G-quadruplexes with peroxidase-like activity^8,9,10, or else RNA-cleaving^11,12,13 DNAzyme. Binary sensors with a 10-23 DNAzyme core have been amply described. They have been adapted to detect single-stranded nucleic acid fragments created synthetically¹¹, or single-stranded amplification products^14,15. Detection without amplification is also possible, for example, in the case of 16S rRNA extracted from a cell culture, since this molecule is abundant in cells^12,16. Binary sensors with a 10-23 DNAzyme core can also be adapted for non-nucleic acid detection, such as for lead ions¹⁷. Nevertheless, low sensitivity is still considered to be one of the major drawbacks of binary 10-23 DNAzyme sensors, and various approaches, including cascades¹⁸ or alternative signal-enhancing methods¹⁹, have been developed to address this issue. Unfortunately, the above techniques drastically increase the cost and instability of the sensor components, and so they have not come into practice.

The experiment described in this work uses 10-23 DNAzyme-based DNA-nanosensors referred to as DNA-nanomachines (DNM)^20,21. These structures include binary sensors and also (1) DNA facilitators flanking the binary probe, unwinding the structured regions and invading into double-stranded DNAs, and (2) a DNA tile that holds all the DNM parts together in close proximity and increases the chances of signal formation (Figure 1). Altogether, the 10-23 DNAzyme-based DNMs decrease the limit of detection (LOD) down to the picomolar range²¹; they can be used to detect 16rRNA in cell lysates²² or report the presence of viral RNA without amplification^23,24. At the same time, the DNMs remain sensitive to the SNV and can even be used to genotype alleles in heterozygous organisms²⁵. The DNM method can be used as a complementary technique to any amplification type for product visualization or identification of the product in a complex mixture with high selectivity. The DNMs, unlike conventional TaqMan and beacon probes, do not require heating of the sample and so can be used in end-point detection protocols, making the overall detection cost-effective.

This article describes the in silico development of a DNM and demonstrates procedures for assembly and functional characterization of DNM. The work was performed using a synthetic fragment of Epstein-Barr virus (EBV) corresponding to the positions 13972-14154 of the viral DNA (OR652423.1) (see Table of materials).

Protocol

1. DNM design

1. Select a unique region within the genome of interest.
   NOTE: This work uses a region of the Epstein-Barr virus (EBV) genome in positions 13972-14154 (OR652423.1).
2. Create a primer set for the amplification of the selected fragment, for example, via Primer3 tool embedded in Ugene²⁶.
3. Open the UNAFold web tool²⁷ (see Table of Materials) and insert the selected region. Change the folding temperature to 55 °C, and ionic conditions to 250 mM of monovalent ions, and 200 mM of Mg²⁺.
   1. Open the resulting secondary structure of the ssDNA fragment and select a stable hairpin. Locate the DNA-nanomachines' binding site in the unstructured regions or, alternatively, on the loops of the hairpin structures to achieve a stronger binding of the target.
4. If the targeted analyte site is folded in a stem-loop structure, place Arm 1 and 2 as follows: Ensure that Arm 2 binds one side of the stem, the loop, and 3-5 nt of the second stem side; Arm 1 binds a fragment of the second stem side.
5. Open the DinaMelt tool²⁸ (see Table of Materials) and analyze the sequences of the arms and their reverse-complement sequences. Use the DinaMelt tool to make sure that the T_m of Arms 2 and 3 is above 65 °C: at least 10 °C above the assay temperature (55 °C).
   1. Keep the T_m of Arm 1 at least 1-3 °C above the reaction temperature (55 °C) if high recognition selectivity is to be achieved²⁹.
6. Combine Arms 1 and 2 sequences with the DNAzyme core's halves and the F-sub binding fragments¹³.
7. Analyze the secondary structure of the designed DNA strands using UNAFold²⁷.
   NOTE: 55 °C (assay temperature), 200 mM Mg²⁺, 250 mM monovalent ions (HEPES, Na⁺, K⁺).
   1. Use the design only if the folding ΔG is lower than - 4 kcal/mol. Try to avoid long CG-rich sequences. If the secondary structure is too stable, introduce a mutation to the Arm to increase the folding ΔG³⁰. Alternatively, placing Arm 3 2-3 nt away from Arm 2 is another strategy to change the Arm 3 sequence to avoid stable intermolecular structures.
8. Create a random sequence for the DNA tile fragment as long as Arm 3. Combine the Arm 3 sequence with the DNA tile sequence via (dT)_4-6 or ethylene glycol (e.g., triethylene glycol (TEG) or hexaethylene glycol (HEG)). The DNM version presented in the experiment uses (dT)₆.
9. Link Arm 2 with the sequence complementary to the DNA tile fragment selected in step 1.8 via a linker.
10. Analyze the secondary structure of the designed DNA strands using the UNAFold web app. Ensure that the folding energy of each DNA strand is above 4 kJ/mol at 55 °C, and avoid long CG-enriched sequences. Preferably, change the DNA tile sequence.
11. Draw the structure using any available vector graphics software. This article uses Biorender and Pixelmator Pro for visualization (see Table of Materials). Verify accuracy in 5'->3' directions for each strand.
    NOTE: Obtain the sequences from a reliable commercial vendor.

2. Functional test of the DNM

1. Preparation of buffer and reagents
   1. Prepare the reaction buffer composed of 50 mM HEPES (pH 7.5-7.8), 50 mM NaCl, 150 mM KCl, and 200 mM MgCl₂. Fill with the deionized water up to 50 mL.
   2. Prepare 10 µM aliquots of Dzb-Tile and Tile-Arm3 oligonucleotides. Prepare 1 µM solution of the Dza, a 100 µM F-sub, and 1 nM, 10 nM, 100 nM analytes in RNase/free water or sterile deionized water. The details of the oligonucleotide sequences are provided in the Table of Materials.
2. DNM assembly
   1. Thaw the oligonucleotide solutions completely before using. Gently mix by tapping (do not intensively vortex). Spin down (rapidly, for a few seconds) the solution using a microcentrifuge.
   2. Mix 1 µM each of Dzb-Tile and Tile-Arm3 in 200 µL of the reaction buffer in 0.5 mL microcentrifuge tubes.
   3. Mix the tube gently and spin down the solution.
   4. Wrap the lid of the closed tube with paraffin film. Alternatively, use screw cap tubes.
   5. Incubate the tube into a 500 mL beaker with boiling water for 2 min, then turn the heater off and let the temperature passively cool down overnight.
   6. Prepare the 12% native PAGE: Mix 1 mL of 10х TBE buffer, 3 mL of 40% AA:BA, deionized water up to 10 mL, 50 µL of APS, and 5 µL of TEMED (see Table of Materials). Move the mixture to the gel cassette and let it polymerize.
   7. Mix 1 µL of each sample, namely, the assembled DNM, Dzb-Tile, and Tile-arm3, with 1 µL of 4x loading dye and load them into the gel.
   8. Put the cassette into the chamber, fill it with 1x TBE buffer, and let the gel run for 90 min at 80 V.
   9. Dye the gel with Ethidium bromide and visualize it using a gel-documenting system (see Table of Materials).
3. LOD assay
   1. Prepare 160 µL of 200 nM F-sub in the reaction buffer. This is done to assess the basic background of the substrate stability.
   2. Prepare 7 aliquots of 160 µL of preassembled DNM, F-sub, and Dza at concentrations of 20 nM of DNM and Dza and 200 nM of F-sub in the reaction buffer. Out of the seven tubes, keep tube #1 as the blank background to assess the spontaneous assembly of the DNAzyme core.
   3. Add the analyte into tubes #2-7 at the final concentrations of 1 pM, 10 pM, 100 pM, 200 pM, 500 pM and 1000 pM.
   4. Gently mix, spin down the tubes, and classify the solutions into three 50 µL portions into a black 96-well plate. Seal the plate with an optically transparent film.
   5. Incubate the plate in a water bath at 55 °C for 1 h. Spin down the plate.
   6. Measure fluorescence at 525 nm (λ_ex = 495 nm). Calculate the mean and the standard deviation for each point.
   7. Calculate the blank-to-basic ratio by dividing the signal of the tube that contains F-sub and DNM by the signal of the tube that contains F-sub only. Ensure the ratio is within the 1.1-1.5 range.
   8. Calculate the signal-to-blank ratio by dividing the signal of the tube that contains F-sub, DNM, and the analyte to the signal of the tube that contains F-sub and DNM only. Consider the signal as true positive if it is higher than the background average plus three standard deviations.
   9. Repeat the experiment twice and ensure the standard deviation between the analytical repeats is less than 0.5.

Results

The aim of the first experiment was to show the assembly of the DNM before the synthetic fragment of the analyte. All the constituent DNM strands were added to the reaction buffer and assembled in the beaker. The assembled DNM complex was assessed for its correct size and homogeneity by native PAGE. Native PAGE shows the assembled DNM with the analyte in lane 1 and two DNM strands in lanes 2 and 4. If the DNA nanomachine were not assembled, 2 or 3 separate bands would be visible instead of a single low-mobility...

Discussion

The design of DNA machines is straightforward but requires some experience in designing hybridization probes or functional DNA nanostructures. It is appropriate to keep the analyte fragment as short as possible to diminish the number of possible secondary structures and simplify DNM invasion to the secondary structure. The CG content should preferably be below 60% to avoid stable intramolecular structures. Successful assembly of the DNM is achieved at slow cooling rates. In some cases, DNMs can be spontaneously asse...

Materials

Name	Company	Catalog Number	Comments
1.5 mL tube	Biofil	CFT011015
100 bp+ DNA Ladder	Evrogen	NL002
100-1000 µL pipette	Kirgen	KG-Pro1000
10-100 µL pipette	Kirgen	KG-Pro100
1-10 µL pipette	Kirgen	KG-Pro10
20-200 µL pipette	Kirgen	KG-Pro200
2-20 µL pipette	Kirgen	KG-Pro20
4x Gel Loading Dye	Evrogen	PB020
Acrylamide 4K	AppliChem	A1090,0500
Ammonium persulfate	Carl Roth	2809447
Biorender	Biorender		https://www.biorender.com
Bisacrylamide	Molekula	22797959
Boric Acid	TechSnab	H-0202
ChemiDoc imaging system	BioRad	12003153
Costar 96-Well Black Polystyrene Plate	Corning	COS3915
DinaMelt	RNA Institute		http://www.unafold.org/Dinamelt/applications/two-state-melting-hybridization.php
EDTA	Amresco	Am-O105
Ethidium bromide	BioLabMix	EtBr-10
HEPES	Amresco	Am-O485
Magnesium chloride	AppliChem	131396
Mini Protean Tetra Cell	BioRad	1658001EDU
pipette 10 µL tips	Kirgen	KG1111-L
pipette 1000 µL tips	Kirgen	KG1636
pipette 200 µL tips	Kirgen	KG1212-L
Pixelmator Pro	Pixelmator Team		https://www.pixelmator.com/pro/
Potassium chloride	Carl Roth	1782751
PowerPac Basic Power Supply	BioRad	1645050
RNAse/DNAse free water	Invitrogen	10977049
Sealing film for PCR plates	Sovtech	P-502
Sodium chloride	Vekton	HCh (0,1)
Spark multimode microplate reader	Tecan	Spark- 10M
T100 amplificator	BioRad	10014822
TEMED	Molekula	68604730
Tris(hydroxymethyl)aminomethane	Amresco	Am-O497
UNAFold	RNA Institute		http://www.unafold.org/mfold/applications/dna-folding-form.php
Water bath	LOIP	LB-140
Oligos used
Analyte:	Evrogen		direct order, standard desalting purification
GAGCACTTGGTCAGGCACACGG ACAGGGTCAGCGGAGGACGCG TGGCACAGCAGCCCGGGGTAG GTCCCCTGGACCTGCCGCTGG CGGACTACGCCTTCGTTGC
Tile-Arm3:	Evrogen		direct order, standard desalting purification
CCG GGCTGCTGTGCCATTTT TTGCTGACTACTGTCACCTCT CTGCTAGTCT
Dzb-Tile: AGACTAGCAGAGAGGTGACAG TAGTCAGCTTTTTTCGCGTCCTC CGCTGACCACAACGAGAGGAA ACCTT	Evrogen		direct order, standard desalting purification
Dza: TGCCCAGGGAGGCTAGCTCT GTCCGTGTGCCTGACCA	Evrogen		direct order, standard desalting purification
F-sub oligonucleotide: AAGGTT(FAM)TCCTCrGrU CCCTGGGCA-BHQ1	DNA-Syntez		direct order, HPLC purification