Programmable DNA Nanoswitches for Detection of Nucleic Acid

Letter pubs.acs.org/acssensors

Programmable DNA Nanoswitches for Detection of Nucleic Acid Sequences Arun Richard Chandrasekaran, Johnny Zavala, and Ken Halvorsen* The RNA Institute, University at Albany, State University of New York, Albany, New York 12222, United States S Supporting Information *

ABSTRACT: Detection of nucleic acid sequences is important for applications such as medicine and forensics, but many detection strategies involve multiple time-consuming steps or require expensive lab equipment. Here we report a programmable DNA nanoswitch that undergoes a predefined conformational change upon binding a target sequence, flipping the switch from a linear “off” state to a looped “on” state. The presence of the target sequence is determined without amplification using standard gel electrophoresis to separate the on and off states. We characterized the nanoswitch on a variety of DNA sequences and fragment lengths, showing detection of fragments as short as 20-nt, and sensitivity into the low picomolar range. Specificity and robustness were demonstrated by detection of a single target sequence from both a randomized pool of high concentration oligonucleotides and from a solution of fetal bovine serum (FBS), with no false positive detection in either case. Furthermore, we optimized the process to take less than 30 minutes from sample mixture to readout. By leveraging the already ubiquitous technique of gel electrophoresis, our low cost approach will be especially accessible to researchers in the biomedical sciences. KEYWORDS: DNA nanoswitch, electrophoresis, nucleic acid detection, DNA biosensors, DNA nanostructures

T

for amplifying the output signal. Our nanoswitch design can provide detection of a sequence with a simple one-pot mixture, and also gives a direct, label-free, and amplification-free readout using gel electrophoresis, which is one of the most common tools in biology. The basic principle of the DNA nanoswitch is that a conformational change is induced upon binding of a specific DNA sequence (Figure 1). The “off” state of the switch is a linear duplex formed by a single-stranded scaffold (7249nucleotide M13) and a set of short complementary “backbone oligos” that hybridize to the scaffold. Two of these strands (detectors 1 and 2) are designed to have overhangs (a on detector 1 and b on detector 2) that are complementary to parts (a* and b*) of the target DNA sequence (key oligonucleotide). Recognition of specific targets is reported by a conformational change between two different states of the nanoswitch. On addition of a specific DNA sequence, target recognition and binding reconfigures the switch to form a loop, thus changing it to the “on” state. The on and off DNA nanoswitches migrate differently on an agarose gel (Figure 1, inset), indicating their relative quantity in the two possible states. The nanoswitch is constructed using techniques adapted

he ability to synthesize arbitrary DNA sequences and the remarkable specificity of Watson−Crick base pairing has led to the recent use of DNA as a nanoscale building material.1,2 On-demand synthesis of DNA provides programmability in design, while base pairing provides the structural “glue” as well as the ability of structures to self-assemble. These features of DNA have been exploited for the construction of two-3−5 and three-dimensional lattices,6−9 as well as for complex shapes that have been facilitated by the advent of DNA origami.10−12 Recently, active structures such as nanodevices13 and nanomachines14−16 have been developed as researchers focus on applying DNA nanotechnology to solve real-world problems in science. One such application is biological sensing, where thoughtfully designed nanosensors have the potential to make detection of specific biological materials simpler, cheaper, and faster. Recently, our lab was involved in the development of a simple two-state DNA nanoswitch for quantifying molecular interactions.17,18 Here we adapt that approach to detect specific DNA sequences, an area with widespread importance in biotechnology, medicine, and forensics.19 In medicine, for example, nucleic acids play a role as biomarkers for many diseases and their detection is crucial to the identification and diagnosis of these diseases.20 Various electrochemical DNA sensors21 and DNA−gold nanoparticle conjugates22 have previously been constructed for this purpose. However, these processes involve multiple detection steps and complex designs © XXXX American Chemical Society

Received: October 23, 2015 Revised: November 23, 2015 Accepted: November 25, 2015

A

DOI: 10.1021/acssensors.5b00178 ACS Sens. XXXX, XXX, XXX−XXX

Letter

ACS Sensors

was a notable decrease and eventually no detection. Next, we tested “asymmetric” key oligonucleotides wherein the same 30nt sequence was truncated only on the a* side (Figure 2C). Here, we found that detection remained nearly constant until the closure length was reduced to 7-nt, at which point there was a precipitous decrease. We additionally tested a temperature ramp (46 to 4 °C) annealing protocol that gave similar results for the asymmetric keys and a more gradual decline for the symmetric keys (Figure S3). The differences in the minimum closure lengths between the symmetric (∼10-nt) and asymmetric (∼7-nt) key oligonucleotides suggest that there may be an inherent benefit to designing asymmetric binding regions, likely due to the deterministic order in binding events. To better understand the effect of temperature on the DNA detection, we additionally performed UV thermal melting analysis on relevant components of the nanoswitches (Figure S4). We found that the detector strands are well bound to the nanoswitch by 40 bp with high melting temperatures (∼70 °C), while the key oligos bind to the detectors at lower temperatures (∼40 °C to ∼60 °C) that generally decrease as the key oligonucleotides shorten (Tables S1 and S2). The data show that the shorter targets do not fully hybridize with the detectors at elevated temperatures, and thus are effectively given less reaction time in the high-temperature annealing protocol. The drop off in detection for the shorter closure lengths likely reflects some inherent limitations imparted by the binding energies. From previous work, we estimated the effective “concentration” of two binding partners on the same nanoswitch to be ∼30 nM,18 implying that an interaction with a KD of 30 nM would be half looped and half unlooped at equilibrium. Previous measurements for the hybridization of a 7 bp DNA gave a dissociation constant in that same range (∼5 nM),23 and the dramatic dropoff in detection observed here around 7 bp (for the asymmetric closures) substantiates that detection of these short sequences is likely limited by interaction energy. To demonstrate the specificity and robustness of the nanoswitches, we showed selective activation of two nanoswitches with different key sequences, detection of a target sequence from a large pool of random sequence oligonucleo-

Figure 1. Design and operation of the two-state DNA nanoswitch. A double-stranded DNA is made with a single-stranded scaffold (pink), complementary backbone oligos (blue), and detector strands (yellow) that can be inserted at different locations. Addition of the key oligonucleotide (red) binds the overhangs of the two detector regions “a” and “b” thereby forming a loop. This conformational change can be read out using gel electrophoresis (inset).

from DNA origami10 (Figure S1), and is programmable for specific target sequences by simply integrating two new detector strands. To characterize detection, we tested different concentrations and lengths of the target DNA sequence (Figure 2). These experiments were performed by incubating ∼100 pM of nanoswitches with the key oligonucleotide at room temperature. To test sensitivity, we used a 30-nt key oligonucleotide and varied the concentration between ∼1 pM and ∼10 nM, with detection visible into the low picomolar range (Figure 2A). To test detection of various fragment lengths, we first tested “symmetric” key oligonucleotides wherein a 30-nt sequence with 15-nt on a* and b* were symmetrically truncated (Figure 2B). The key oligonucleotides were detected with various efficiencies, remaining mostly constant until the closure length was reduced to about 10-nt, at which point there

Figure 2. Response of the DNA nanoswitch. (A) Concentration dependence for detection of a 30-nt target. (B) Binding of the “symmetric” key oligonucleotides to the nanoswitch. Closure length was varied from 15-nt to 7-nt. (C) “Asymmetric” key oligonucleotides with closure lengths varying from 15-nt to 5-nt (with b* being constant at 15-nt). B

DOI: 10.1021/acssensors.5b00178 ACS Sens. XXXX, XXX, XXX−XXX

Letter

ACS Sensors

Figure 3. Specificity and robustness of detection. (A) Agarose gel shows the sequence specific activation of the nanoswitch. Switch X (orange) turns on only in the presence of key oligonucleotide X (orange) and switch Y turns on only in the presence of key oligonucleotide Y with no background detection of the incorrect strand. (B) The nanoswitch does not detect competing random sequences, but does detect the correct sequence from the random pool, even at a large molar excess (40× shown here with similar results over the full range shown in Figure S5). (C) Robust detection of the key oligonucleotide spiked into fetal bovine serum (final concentration of 20% FBS).

tides, and detection of a target spiked into a biological fluid (Figure 3). First, we designed two nanoswitches (Switch X and Switch Y) to detect two different target sequences of similar length (Key X and Key Y, respectively). We showed sequence specific activation of the two nanoswitches only in the presence of the correct key (Figure 3A), with no false positive detection of the off-target sequences. We further showed specificity of the nanoswitches by incubating them with a pool of random oligonucleotides (15-nt complementing one detector and 10 random bases to potentially interact with the other detector). Again, we found no false positive detection of the random oligonucleotides, even at concentrations as high as 50 μM where the expected concentration of single and double mismatches are 1.4 and 19 nM, respectively (Figure 3B: lane 3, Figure S5 and experimental methods). Next, we showed detection of a target sequence out of a large pool of oligonucleotides with the same size but random sequences (Figure 3B: lane 4 and Figure S5). The detection signal of the target sequence was largely unchanged by increasing the “background noise” from the random oligonucleotides in solution. This “zero background” detection is in stark contrast to many surface-based capture assays, where inadvertent detection of off-target sequences can pose a significant challenge. Furthermore, we showed robust detection of the target oligonucleotide spiked into a fetal bovine serum (FBS) solution (Figure 3C), illustrating the potential of performing detection from complex biological fluids. This nanoswitch assay has several important benefits including a simple one-step mixture and a detection scheme which is amenable to biologists already familiar with the gel electrophoresis workflow. To further increase the accessibility of the technique, we characterized and reduced the time required to detect target sequences. We examined the association rate of a key oligonucleotide onto the nanoswitch, and the separation of looped and unlooped nanoswitches in the gel. Association kinetics were measured by conducting a time series at room temperature (Figure 4A). We detected DNA binding in as little as 15 min, and detection was complete by 1 h. To measure the separation time, we ran gels at a higher

Figure 4. Fast detection and read-out of nucleic acid sequences. (A) Normalized fraction of nanoswitches in the on-state after specific time intervals. Gel results are shown within the graph at different time points. (B) Gel readout at different gel running times. Separation of the on- and the off-state is clearly visible even after 20 min. The experiment was done at both 25 and 4 °C. (C) Scheme showing the fast, label-free detection strategy. Detection of the target sequence happens in 10 min and can be read out using gel electrophoresis within 20 min.

voltage (150 V) and imaged them in 10 min increments (Figure 4B), finding that 10−20 min at 150 V was sufficient to detect the looped nanoswitches. From these results, we further showed that we can detect a specific DNA sequence from solution in as little as 30 min from start to finish (Figure 4C). This short time for detection and the minimal requirements for the method make this a fitting approach for point-of-use detection, especially considering the availability of hand-held bufferless gel systems (e.g., Invitrogen E-gel system). In summary, we have shown that we can effectively detect specific DNA sequences of various lengths with picomolar level sensitivity using DNA nanoswitches. These nanoswitches can C

DOI: 10.1021/acssensors.5b00178 ACS Sens. XXXX, XXX, XXX−XXX

Letter

ACS Sensors be pre-made for various targets and can act as a “smart reagent” for a mix-and-read detection assay. Target sequences can be detected even from a pool of randomized oligonucleotides, or from fetal bovine serum with no false positive detection, illustrating their potential use in biological sensing applications. These DNA nanoswitches are inexpensive, costing less than one penny in materials for the amount used in a typical gel lane in this paper. Since the scheme to use them is nontechnical, requiring only common gel electrophoresis materials, the technique is expected to be highly accessible to many researchers.

■

DNA Crystals with a Triple-Helical Veneer. Angew. Chem., Int. Ed. 2014, 53, 3979−3982. (9) Zhao, J.; Chandrasekaran, A. R.; Li, Q.; Li, X.; Sha, R.; Seeman, N. C.; Mao, C. Post-Assembly Stabilization of Rationally Designed DNA Crystals. Angew. Chem., Int. Ed. 2015, 54, 9936−9939. (10) Rothemund, P. W. K. Folding DNA to Create Nanoscale Shapes and Patterns. Nature 2006, 440, 297−302. (11) Douglas, S. M.; Dietz, H.; Liedl, T.; Högberg, B.; Graf, F.; Shih, W. M. Self-Assembly of DNA into Nanoscale Three-Dimensional Shapes. Nature 2009, 459, 414−418. (12) Chandrasekaran, A. R. DNA Origami and Biotechnology Applications: A Perspective. J. Chem. Technol. Biotechnol. 2015, n/a. (13) Idili, A.; Vallée-Bélisle, A.; Ricci, F. Programmable pH-Triggered DNA Nanoswitches. J. Am. Chem. Soc. 2014, 136, 5836−5839. (14) Douglas, S. M.; Bachelet, I.; Church, G. M. A Logic-Gated Nanorobot for Targeted Transport of Molecular Payloads. Science 2012, 335, 831−834. (15) Modi, S.; Swetha, M. G.; Goswami, D.; Gupta, G. D.; Mayor, S.; Krishnan, Y. A DNA Nanomachine that Maps Spatial and Temporal pH Changes Inside Living Cells. Nat. Nanotechnol. 2009, 4, 325−330. (16) Andersen, E. S.; Dong, M.; Nielsen, M. M.; Jahn, K.; Subramani, R.; Mamdouh, W.; Golas, M. M.; Sander, B.; Stark, H.; Oliveira, C. L.; Pedersen, J. S.; Birkedal, V.; Besenbacher, F.; Gothelf, K. V.; Kjems, J. Self-Assembly of a Nanoscale DNA Box with a Controllable Lid. Nature 2009, 459, 73−76. (17) Halvorsen, K.; Schaak, D.; Wong, W. P. Nanoengineering a Single-Molecule Mechanical Switch using DNA Self-Assembly. Nanotechnology 2011, 22, 494005. (18) Koussa, M. A.; Halvorsen, K.; Ward, A.; Wong, W. P. DNA Nanoswitches: A Quantitative Platform for Gel-Based Biomolecular Interaction Analysis. Nat. Methods 2015, 12, 123−126. (19) Butler, J. M. Forensic DNA Typing: Biology, Technology, and Genetics of STR Markers, 2nd ed.; Elsevier Academic Press: New York, 2005. (20) Rosi, N. L.; Mirkin, C. A. Nanostructures in Biodiagnostics. Chem. Rev. 2005, 105, 1547−1562. (21) Pei, H.; Lu, N.; Wen, Y.; Song, S.; Liu, Y.; Yan, H.; Fan, C. A. DNA Nanostructure-Based Biomolecular Probe Carrier Platform for Electrochemical Biosensing. Adv. Mater. 2010, 22, 4754−4758. (22) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Selective Colorimetric Detection of Polynucleotides Based on the Distance-Dependent Optical Properties of Gold Nanoparticles. Science 1997, 277, 1078−1081. (23) Halvorsen, K.; Agris, P. F. Cross-Platform Comparison of Nucleic Acid Hybridization: Toward Quantitative Reference Standards. Anal. Biochem. 2014, 465, 127−133.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acssensors.5b00178. Methods, additional experimental data, and DNA sequences used (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions

ARC and KH designed experiments, analyzed data, and wrote the manuscript. ARC and JZ performed experiments. Notes

The authors declare the following competing financial interest(s): ARC and KH are inventors on a patent application covering aspects of this work.

■

ACKNOWLEDGMENTS This research was supported by The RNA Institute. The authors thank Dhruv Patel for assisting with some experiments, and Drs. Mehmet Yigit, Maria Basanta Sanchez, Sri Ranganathan, Jia Sheng, Prash Rangan, and Wesley P. Wong for useful discussions. We thank Dr. Gabriele Fuchs for providing the fetal bovine serum.

■

REFERENCES

(1) Seeman, N. C. DNA in a Material World. Nature 2003, 421, 427−431. (2) Jones, M. R.; Seeman, N. C.; Mirkin, C. A. Programmable Materials and the Nature of the DNA Bond. Science 2015, 347, 1260901. (3) Winfree, E.; Liu, F.; Wenzler, L. A.; Seeman, N. C. Design and Self-assembly of Two-Dimensional DNA Crystals. Nature 1998, 394, 539−544. (4) He, Y.; Tian, Y.; Chen, Y.; Deng, Z.; Ribbe, A. E.; Mao, C. Sequence Symmetry as a Tool for Designing DNA Nanostructures. Angew. Chem., Int. Ed. 2005, 44, 6694−6696. (5) Yan, H.; Park, S. H.; Finkelstein, G.; Reif, J. H.; LaBean, T. H. DNA-Templated Self-Assembly of Protein Arrays and Highly Conductive Nanowires. Science 2003, 301, 1882−1884. (6) Zheng, J.; Birktoft, J. J.; Chen, Y.; Wang, T.; Sha, R.; Constantinou, P. E.; Ginell, S. L.; Mao, C.; Seeman, N. C. From Molecular to Macroscopic via the Rational Design of a Self-Assembled 3D DNA Crystal. Nature 2009, 461, 74−77. (7) Sha, R.; Birktoft, J. J.; Nguyen, N.; Chandrasekaran, A. R.; Zheng, J.; Zhao, X.; Mao, C.; Seeman, N. C. Self-Assembled DNA Crystals: the Impact on Resolution of 5′-Phosphates and the DNA Source. Nano Lett. 2013, 13, 793−797. (8) Rusling, D. A.; Chandrasekaran, A. R.; Ohayon, Y. P.; Brown, T.; Fox, K. R.; Sha, R.; Mao, C.; Seeman, N. C. Functionalizing Designer D

DOI: 10.1021/acssensors.5b00178 ACS Sens. XXXX, XXX, XXX−XXX