Three-dimensional DNA nanostructures to improve the hyperbranched hybridization chain reaction† †This work is dedicated to 100th anniversary of Nankai University. ‡ ‡Electronic supplementary information (ESI) available. See DOI: 10.1039/c9sc02281c

Nonenzymatic nucleic acid amplification techniques (e.g. the hybridization chain reaction, HCR) have shown promising potential for amplified detection of biomarkers.


Materials and instruments
All DNA and RNA oligonucleotides (Table S1) were purchased from Sangon Biotech Co., Ltd. (Shanghai, China) and purified by high-performance liquid chromatography. 1,3-Diphenylisobenzofuran (DPBF), 2',7'-dichlorodihydrofluorescein diacetate (DCFH-DA) and estradiol were obtained from Sigma-Aldrich. Lipofectamine 3000 transfection reagent was purchased from Invitrogen (USA). Methylene blue (MB) was purchased from Acros Organics. Deoxyribonucleases I (DNase I) was obtained from New England Biolabs Beijing Ltd. All reagents used in this work were of analytical grade and directly used without further purification. Deionized and sterilized water (resistance >18 MΩ·cm) was used throughout the experiments.
UV spectral measurements were performed on an Agilent Technologies Cary 60 UV/Vis spectrometer. All fluorescence measurements were carried out on a Shimadzu RF-5301 PC fluorescence spectrometer (Shimadzu Ltd., Japan). AFM characterization was observed by Bruker Multimode V8 Scanning Probe Microscopy (USA). The images of cells were visualized on an inverted confocal microscope (A1R Confocal System, Nikon).

Construction of TDN-assembled hairpins
TDNs with different valences were prepared by mixing corresponding oligonucleotide strands (Table S1) in TM buffer (20 mM Tris-HCl, 50 mM MgCl 2 , pH = 8.0). Unless otherwise stated, the concentration of a or a* is 1 µM and H1, H2 changes with different valences. For instance, the mixtures of qTDNH1 (a*, b*, c*, d* are 1 µM and H1 is 4 µM) were heated at 95 °C for 5 min and then cooled to 4 °C within 1 min. The obtained modules were then directly used for characterization without further fractionation or purification. To achieve the loading of MB in modules, MB was mixed with the prepared modules (e.g. qTDNH1 or qTDNH2) at 37 °C for 1 h and then purified by ultrafiltration (30 kDa molecular weight cutoff) to remove the free MB remaining in solution. The supernatants were collected to analyze the concentration of the released MB. The concentration of MB in modules was calculated as follows: c MB loading = (n MB total − n MB free )/V modules .

MiR-21-triggered DNA assembly
According to Table S2, differently valent TDNH1 and TDNH2 or h1 and h2 were mixed with miR-21. The mixture was then isothermally incubated at 37 °C for proper time. 6

Electrophoresis characterization
Agarose gel electrophoresis was applied to characterize the successful preparation of building modules and DNA assemblies.
The reaction solutions were separated by 2% agarose gel electrophoresis at 4 °C (80 V, 30 min). The gels were stained with GelRed (Takara) in advance and visualized with a Gel Image System.

Atomic force microscopy imaging
In air scan. The modules and DNA assemblies (1 μM, 10 μL) were deposited onto a freshly cleaved mica surface and left to adsorb to the surface for 5 min, washed with 30 μL of water for more than 10 times and dried with compressed air. A MultiMode V8 AFM (Bruker) system was used to image the samples under ScanAsyst-Air mode.

FRET measurement
To achieve the in vitro FRET sensing of miR-21, different concentrations of miR-21 were added in 100 μL of Tris-HCl solution (20 mM Tris with 10 mM MgCl 2 , pH = 7.4) containing 50 nM qTDNH1 Cy3 and 50 nM qTDNH2 Cy5 or 50 nM h1 Cy3 and 50 nM h2 Cy5 (H1 Cy3 , H2 Cy5 , h1 Cy3 or h2 Cy5 is 50 nM), followed by incubation for proper time at 37 °C. Then, the fluorescence spectrum of the mixture was collected from 550 to 800 nm with 540 nm excitation. All experiments were repeated at least three times. To investigate the kinetics of DNA assembly process, real-time monitoring of fluorescence intensity at 560 nm was performed after miR-21 addition. The in vitro FRET sensing of Let-7a was conducted by using the above-described process.
The FRET efficiency was evaluated using the equation of E=F A /F SUM 1-5 , where E is FRET efficiency, F A is the integral area of acceptor Cy5 (from 620 to 800 nm) in the fluorescence spectrum of DNA assembly, F SUM is the total integral area of the whole spectrum of DNA assembly.

Biostability characterization of modules and DNA assemblies
To compare the abilities of nuclease resistance, 1 μM h1, 1 μM qTDNH1, HCR products formed by h1 and h2, hyperbranched HCR products formed by qTDNH1 and qTDNH2 were respectively incubated with 0.5 U/mL DNase I at room temperature for different time. Then, 20 μL of reaction solution was mixed with 4 μL of loading buffer (6×), followed by 2% agarose gel electrophoresis running at 80 V for about 30 min in 1×TAE buffer and analyzed by a fluorescence image scanner.