Aptamer-Based Sensing Platform Using Three-Way DNA Junction

Dec 6, 2013 - We proposed a new three-way DNA junction-driven strand displacement mode and fabricated an aptamer-based label-free fluorescent sensing ...
4 downloads 8 Views 2MB Size
Download PDF
Letter pubs.acs.org/ac

Aptamer-Based Sensing Platform Using Three-Way DNA JunctionDriven Strand Displacement and Its Application in DNA Logic Circuit Jinbo Zhu, Libing Zhang, Zhixue Zhou, Shaojun Dong, and Erkang Wang* State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun, Jilin 130022, P. R. China University of Chinese Academy of Sciences, Beijing, 100049, P. R. China S Supporting Information *

ABSTRACT: We proposed a new three-way DNA junction-driven strand displacement mode and fabricated an aptamer-based label-free fluorescent sensing platform on the basis of this mechanism. Assembling the aptamer sequence into the three-way DNA junction makes the platform sensitive to the target of the aptamer. A label-free signal readout method, split G-quadruplex enhanced fluorescence of protoporphyrin IX (PPIX), was used to report the final signal. Here, adenosine triphosphatase (ATP) was taken as a model and detected through this approach, and DNA strand could also be detected by it. The mechanism was investigated by native polyacrylamide gel electrophoresis. Furthermore, on the basis of this molecular platform, we built a logic circuit with ATP and DNA strands as input. Aptamer played an important role in mediating the small molecule ATP to tune the DNA logic gate. Through altering the aptamer sequence, this molecular platform will be sensitive to various stimuli and applied in a wide field.

I

toehold end is essential for this reaction. Distributing binding and toehold domain on different strands will make the design flexible and the reaction easily tuned. DNA aptamer is an important separation and analysis tool in target molecule sensing.26,27 If the aptamer sequence is incorporated into the three-way junction structure, the molecular platform will be sensitive to its corresponding target. This strategy can be used to build a DNA logic circuit that can receive input signals in many other forms. Extension of the input range will promote the application of the logic circuit in a wide field. Herein, we introduced a novel DNA three-way junction driven-strand displacement mode and built a label-free fluorescent sensing platform with the help of aptamer. The structure of the molecule and mechanism of the reaction are illustrated in Scheme 1. The binding and toehold domains are distributed on two different strands, respectively. The third strand hybridizes with them to form the Y-shaped junction

n recent years, DNA has been demonstrated as an excellent application prospect in wet computing, using biomolecules or living organisms to perform computing functions, for its outstanding data storage, parallel processing capabilities, and flexibility in design.1−8 Toehold-mediated strand displacement reaction (SDR) is a very useful tool in DNA computing to receive and process the DNA strand input signals.9−11 Since it can replace the DNA nucleases to release the output strands, it has been used to construct various logic gates, circuits, and even neural networks.12−20 However, the classic strand displacement mode is insufficient for building more powerful computing systems, which can proceed the information in a more efficient way and receive more input signals in different molecular styles, not merely DNA strands. To meet the growing needs of designing complex logic circuits and application in biological environments, new modes to tune the SDR are constantly emerging in recent years.21−23 DNA three-way junction is a Y-shaped DNA structure composed of three strands hybridized with each other.24,25 The multiarm structure gives us more choices to build sensing platforms. For toehold-mediated DNA strand displacement, the © 2013 American Chemical Society

Received: October 7, 2013 Accepted: December 6, 2013 Published: December 6, 2013 312

dx.doi.org/10.1021/ac403235y | Anal. Chem. 2014, 86, 312−316

Analytical Chemistry

Letter

Scheme 1. Schematic Diagram of the Sensing Platform Based on Three-Way Junction-Driven Toehold-Mediated SDRa

a

The binding domain and toehold domain are colored red and green, respectively. The G-rich sequences are colored blue.

strand S draws the two G-rich segments G1 and G2 together to induce the formation of split G-quadruplex, which can bind with PPIX and dramatically enhance its fluorescence. The fluorescence spectra are shown in Figure 1. Strand S was

through the toehold end and displace the output strand. Taking the adenosine triphosphatase (ATP)-binding aptamer (ABA) as the toehold domain-carrying strand, the molecular platform will be sensitive to the input ATP.28,29 At the same time, we can also use its complementary strand to hybridize ABA. Thus, ATP and DNA strand both can be used to tune the SDR. Through altering the aptamer strand, this sensing platform will be able to detect various other targets. Protoporphyrin IX (PPIX) usually aggregates into micelles with low fluorescence in aqueous solution, whereas its fluorescence can be dramatically enhanced upon binding to G-quadruplex.30,31 Split G-quadruplex enhanced fluorescence of PPIX has been utilized to capture the output strand and give off a final fluorescent signal in our previous works.12,32 We continued to use it as a signal reporter in this work, because this label-free fluorescence readout technique is very simple, economic, and easy to design. Furthermore, from another perspective, the whole system can work as a simple logic circuit with ATP and DNA as input. Since the gates in the circuit are all cascaded by DNA strands, this device has potential to expand into a complex computing system with specific function. Upon integrating different aptamer sequences into the circuit, this molecular platform will be sensitive to various stimuli and applied in more extensive areas. The three-way DNA junction-driven SDR is the core reaction of the molecular platform, whose route is shown in Scheme 1. Strand S, which can draw G1 and G2 together to form the split G-quadruplex, hybridizes with strand I by its binding domain at first. A small part of the binding domain is placed on strand ABA, which can accelerate the hybridization of ABA and SI complex and help strand I capture the S in solution at the same time. Length of the toehold end is a crucial factor for the SDR.33 Taking the length and sequence of ABA into account, a toehold domain with a length of 8 bases is designed here to keep the reaction carrying out smoothly. Since binding domain and toehold domain are distributed on strand I and ABA, respectively, strand R fails to replace S from I without the mediation of toehold end. Strand ABA can bind to the SI complex to draw these two domains close and provide the toehold end for R. Formation of the three-way junction structure is a causative factor of the SDR. Upon the mediation of toehold end on ABA, R hybridizes with ABA and I to form the DNA junction and S is given off. Subsequently, the output

Figure 1. Fluorescence emission spectra of the complexes of PPIX and different DNAs: (a) only PPIX; (b) G1, G2; (c) G1, G2, S; (d) G1, G2, S, I; (e) G1, G2, S, I, ABA; (f) G1, G2, S, I, R; (g) G1, G2, S, I, R, ABA; (h) G1, G2, S, I, R, ABA, ATP; (i) G1, G2, S, I, R, ABA, cABA. The spectra were collected in TEK buffer with a final concentration of 1.2 μM for PPIX, 0.16 μM for G1 and G2, 0.1 μM for S, 0.12 μM for the other strands, and 480 μM for ATP.

blocked by I, and the fluorescence intensity was at a low level at first. Adding either ABA or R was insufficient to release S (curve c, d), unless both of them were input. Strand S was replaced when both ABA and R were present and induced a high fluorescence signal (curve e). These results proved that the three-way DNA junction-driven SDR indeed happened as we designed. Assembling the aptamer sequence into the Y-junction enables the molecular platform to be sensitive to its corresponding target. ATP is chosen as a model, and its aptamer ABA is used in this work. As shown in Scheme 1, ABA binds to the SI complex to supply the toehold end for strand R. Since aptamer can selectively and tightly bind with its target, the binding between the SI complex and ABA can be easily 313

dx.doi.org/10.1021/ac403235y | Anal. Chem. 2014, 86, 312−316

Analytical Chemistry

Letter

inhibited by ATP. Thus, the fluorescence intensity will drop down to a low level when enough ATP is added (Figure 1, curve f). In Figure 2, the fluorescence intensity decreases with

Figure 2. Fluorescence emission spectra of the sensing platform (PPIX, G1, G2, S, I, R, ABA) with different concentrations of ATP (from curve a to g): 0, 80, 160, 240, 320, 400, and 480 μM. PPIX, G1, G2, S, I, and R were added in curve (h) as control. The spectra were collected in TEK buffer with a final concentration of 1.2 μM for PPIX, 0.16 μM for G1 and G2, 0.1 μM for S, and 0.12 μM for the other strands. Inset: dependence of the normalized fluorescence intensity (FI) at 630 nm on the ATP concentration. The FI of the sensing platform without ATP was set as 1. The data were collected from three independent experiments.

Figure 3. Native 15% polyacrylamide gel analysis of the formation of three-way junction structure and SDR. DNA strands added in every lane are indicated in the table. Concentrations for each DNA strand in PAGE are all 0.2 μM. ATP is added with a concentration of 3.2 mM.

first identified from lane 2 to 7. The mobility of the bands got lower after each addition of DNA strand, which was caused by the increase of weight and volume of the DNA complex. Through comparing the bands in these lanes, we can draw the conclusion that I, R, and ABA bind together to form the Y junction in lane 7. Strand S was tightly blocked by I, and the band of the SI double-stranded complex was obviously in lane 8. Addition of R affected this band a little in lane 11, which meant that only R was insufficient to replace S from I. However, when R and ABA were both added in lane 10, the band of SI complex got very weak and a new band appeared at the same position of Y junction in this lane. This phenomenon proved that SI complex was disbanded and the two input strands hybridized with the strand I to form the Y junction. Thus, the formation of the Y junction structure is a key factor of the SDR. Effects of cABA and ATP for this system were also investigated, respectively. In lane 13, the bands of SI and ABA/ cABA complex were both stable, but no band of Y junction was observed. This phenomenon indicated that the SDR was inhibited by the absence of free ABA to form Y junction. A similar case occurred when ATP was input in lane 15. The differences were that a weak band of Y junction appeared, and the band of SI complex became thin in lane 15 compared with lane 11 and 13. It should be due to the binding between ABA and ATP being looser than ABA and its fully complementary strand cABA. Although part of SI complex disbanded and a small amount of Y junction was formed in lane 15, most S still hybridized with I and the fluorescence intensity was kept at a low level (Figure 1, curve f). Therefore, the PAGE results and fluorescence data match well for each other. It is indeed a reliable sensing platform that functions as we designed. To further demonstrate the practicability of the three-way DNA junction driven-SDR mode, we built a simple logic circuit with ATP and DNA strands as input based on this platform. The schematic diagram is shown in Figure 4a. First, strand S and the G-rich segments worked as a YES gate. When free S

the addition of ATP, and there is a linear relationship between the fluorescence intensity and ATP concentration from 80 to 480 μM with a detection limit of 40 μM (three times the standard deviation of the blank solution). This result proves that the more ATP is input the less S is output. To exclude other interference possibilities, we tested the effects of ATP itself for this system. The control experiment indicated that ATP in the concentration range used in this work had little influence on the fluorescence enhanced by the split Gquadruplex (Figure S1, Supporting Information). Its selectivity was also investigated. As shown in Figure S2, Supporting Information, CTP, GTP, and UTP all failed to inhibit this SDR. Furthermore, due to the natural property of DNA, the input strand ABA can be blocked by its complementary strand cABA (Scheme 1). The presence of cABA also leads to a low output signal (Figure 1, curve g), and its effect on the SDR is greater than ATP. It should be attributed to the fully complementary duplex of cABA/ABA which is much stabler than ATP/ABA complex. The fluorescence intensity also linearly decreases with the addition of cABA in the range from 20 to 120 nM with a detection limit of 6.9 nM (three times the standard deviation of the blank solution) (Figure S3, Supporting Information). Analysis of the DNA strand cABA demonstrates that this sensing platform can be used to detect the specific sequence. Through altering the sequence of strand ABA, it will detect more other target complementary strands or some diseaserelated sequences. Native polyacrylamide gel electrophoresis (PAGE) was used to identify the mechanism of this three-way DNA junctiondriven SDR. Figure 3 shows the PAGE result, in which the DNAs contained in different lanes have been figured out in the table. Formation of the three-way DNA junction structure was 314

dx.doi.org/10.1021/ac403235y | Anal. Chem. 2014, 86, 312−316

Analytical Chemistry

Letter

fluorescent sensing platform using this strand displacement reaction. In this novel SDR, the toehold and displacement domain distribute on different strands, and the strand displacement is driven by the formation of the three-way DNA junction structure. Through assembling aptamer sequence into the three-way junction, the formation of this junction structure can be tuned by the corresponding target of the aptamer. Taking ATP and its aptamer ABA as a model, we fabricated a sensing platform based on the three-way junctiondriven SDR in this work. When ABA was regarded as general strand, this platform was sensitive to its complementary strand. This result proved that it can also be used to detect DNA strand. Linear correlations were obtained over the ATP concentration in the range from 80 to 480 μM with a detection limit of 40 μM and over the DNA strand concentration in the range from 20 to 120 nM with a detection limit of 6.9 nM. Compared with the conventional fluorescent biosensors based on aptamer,35,36 the detection limit of this platform may be unspectacular, but this design can avoid the limitation of the sequence and folding configuration of aptamer. The aptamer sequence can be changed flexibly according to the target in this platform. Thus, this aptamer-based sensing platform has good universality and expansibility. Furthermore, it was an ideal candidate to build the DNA computing systems which could receive various biomolecules as input. We have built a simple logic circuit based on this platform with ATP and DNA strands as input in this work. By combining suitable aptamer and DNA logic circuit together, this molecule system would possess more powerful functions and solve more complex problems.

Figure 4. (a) Schematic diagram of the logic circuit with strands R, cABA, and ATP as input. (b) Normalized fluorescence intensity at 630 nm of the different input modes of the circuit. The fluorescent analysis was performed in the TEK buffer with a final concentration of 1.2 μM for PPIX, 0.16 μM for G1 and G2, 0.1 μM for S, 0.12 μM for I, ABA, cABA, and R, and 480 μM for ATP. Fluorescence intensity at 630 nm of curve (c) in Figure 1 is set as 1. Threshold value is set at 0.45 to judge the positive and negative output signals. The three binary numbers represent the input status of R, ATP, and cABA, respectively. The data were collected from three independent experiments.

■

ASSOCIATED CONTENT

S Supporting Information *

Materials, DNA sequence, detailed procedures of the method, and parts of the experimental results. This material is available free of charge via the Internet at http://pubs.acs.org.

strand was present, the split G-quadruplex would be formed and enhance the fluorescence intensity of PPIX. Second, an AND gate was used to control the release of S. The characterization of this gate is that the high positive output (1) results only if both the inputs to the gate are high.34 Strand R and ABA were two inputs of this gate. Strand R provided a toehold end, and ABA displaced S upon the mediation of the toehold. These two inputs cooperated with each other to release S as the output signal. Absence of any one of them would lead to the low fluorescence signal (Figure 1, curve c, d). Third, using ATP and cABA to bind ABA, a NOR gate was formed and cascaded to the AND gate. For a NOR gate, the high output (1) is only given off if both the inputs to the gate are low.34 As was mentioned above, whether ATP or cABA was enough to inhibit the SDR by binding with ABA (Figure 1, curve f, g), the normalized fluorescence results for different inputs of the circuit were shown in Figure 4b. It is worth noting that a threshold value of 0.45 was set to judge the fluorescent output signal. The corresponding input concentrations of ATP and cABA are about 400 μM and 70 nM, respectively. These concentrations can be used to judge the input state. The fluorescence result matched well with the truth table given in Table S2, Supporting Information. It is a significant attempt to apply a small molecule and its aptamer as inputs to tune the output signal of logic circuit. If some relationship between the final output signal and drug release is established, a target biomolecule sensitive smart drug delivery system will be expected to be realized. To sum up, we introduced a three-way junction-driven strand displacement mode and built an aptamer-based label-free

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the National Natural Science Foundation of China (Grants 21075116 and 21190040) and 973 projects (2010CB933600 and 2011CB911000).

■

REFERENCES

(1) Schaus, T. E.; Yin, P. Nat. Nanotechnol. 2013, 8, 546−548. (2) Lipton, R. J. Science 1995, 268, 542−545. (3) Benenson, Y.; Paz-Elizur, T.; Adar, R.; Keinan, E.; Livneh, Z.; Shapiro, E. Nature 2001, 414, 430−434. (4) Katz, E., Ed.; Biomolecular Computing - From Logic Systems to Smart Sensors and Actuators; Willey-VCH: Weinheim, 2012. (5) Katz, E.; Privman, V. Chem. Soc. Rev. 2010, 39, 1835−1857. (6) de Murieta, I. S.; Miro-Bueno, J. M.; Rodriguez-Paton, A. Curr. Bioinf. 2011, 6, 173−184. (7) Willner, I.; Shlyahovsky, B.; Zayats, M.; Willner, B. Chem. Soc. Rev. 2008, 37, 1153−1165. (8) Han, D.; Zhu, G.; Wu, C.; Zhu, Z.; Chen, T.; Zhang, X.; Tan, W. ACS Nano 2013, 7, 2312−2319. (9) Qian, L. L.; Winfree, E. Science 2011, 332, 1196−1201. (10) Zhang, D. Y.; Seelig, G. Nat. Chem. 2011, 3, 103−113. 315

dx.doi.org/10.1021/ac403235y | Anal. Chem. 2014, 86, 312−316

Analytical Chemistry

Letter

(11) Yurke, B.; Turberfield, A. J.; Mills, A. P.; Simmel, F. C.; Neumann, J. L. Nature 2000, 406, 605−608. (12) Zhu, J.; Zhang, L.; Li, T.; Dong, S.; Wang, E. Adv. Mater. 2013, 25, 2440−2444. (13) Seelig, G.; Yurke, B.; Winfree, E. J. Am. Chem. Soc. 2006, 128, 12211−12220. (14) Wickham, S. F. J.; Bath, J.; Katsuda, Y.; Endo, M.; Hidaka, K.; Sugiyama, H.; Turberfield, A. J. Nat. Nanotechnol. 2012, 7, 169−173. (15) Qian, L.; Winfree, E.; Bruck, J. Nature 2011, 475, 368−372. (16) Li, B.; Ellington, A. D.; Chen, X. Nucleic Acids Res. 2011, 39, No. e110. (17) Li, W.; Yang, Y.; Yan, H.; Liu, Y. Nano Lett. 2013, 13, 2980− 2988. (18) Zhu, J.; Zhang, L.; Dong, S.; Wang, E. ACS Nano 2013, 7, 10211−10217. (19) Zhang, L.; Zhu, J.; Zhou, Z.; Guo, S.; Li, J.; Dong, S.; Wang, E. Chem. Sci. 2013, 4, 4004−4010. (20) Guo, W.; Yuan, J.; Wang, E. Chem. Commun. 2011, 47, 10930− 10932. (21) Zhang, D. Y. J. Am. Chem. Soc. 2011, 133, 1077−1086. (22) Genot, A. J.; Bath, J.; Turberfield, A. J. Angew. Chem. 2012, 52, 1189−1192. (23) Xing, Y.; Yang, Z.; Liu, D. Angew. Chem. 2011, 50, 11934− 11936. (24) Shlyakhtenko, L. S.; Potaman, V. N.; Sinden, R. R.; Gall, A. A.; Lyubchenko, Y. L. Nucleic Acids Res. 2000, 28, 3472−3477. (25) Zhao, Y.; Qi, L.; Chen, F.; Zhao, Y.; Fan, C. Biosens. Bioelectron. 2013, 41, 764−770. (26) Xing, H.; Wong, N. Y.; Xiang, Y.; Lu, Y. Curr. Opin. Chem. Biol. 2012, 16, 429−435. (27) Sefah, K.; Phillips, J. A.; Xiong, X.; Meng, L.; Van Simaeys, D.; Chen, H.; Martin, J.; Tan, W. Analyst 2009, 134, 1765−1775. (28) Wang, L.; Zhu, J.; Han, L.; Jin, L.; Zhu, C.; Wang, E.; Dong, S. ACS Nano 2012, 6, 6659−6666. (29) Du, Y.; Li, B.; Wei, H.; Wang, Y.; Wang, E. Anal. Chem. 2008, 80, 5110−5117. (30) Li, T.; Wang, E.; Dong, S. Anal. Chem. 2010, 82, 7576−7580. (31) Li, Y.; Geyer, C. R.; Sen, D. Biochemistry 1996, 35, 6911−6922. (32) Zhu, J.; Zhang, L.; Wang, E. Chem. Commun. 2012, 48, 11990− 11992. (33) Zhang, D. Y.; Winfree, E. J. Am. Chem. Soc. 2009, 131, 17303− 17314. (34) Zhu, J.; Li, T.; Zhang, L.; Dong, S.; Wang, E. Biomaterials 2011, 32, 7318−7324. (35) Leung, K. H.; Lu, L.; Wang, M.; Mak, T. Y.; Chan, D. S. H.; Tang, F. K.; Leung, C. H.; Kwan, H. Y.; Yu, Z.; Ma, D. L. PLoS One 2013, 8, No. e77021. (36) Jin, F.; Lian, Y.; Li, J.; Zheng, J.; Hu, Y.; Liu, J.; Huang, J.; Yang, R. Anal. Chim. Acta 2013, 799, 44−50.

316

dx.doi.org/10.1021/ac403235y | Anal. Chem. 2014, 86, 312−316