Regulation of DNA strand displacement using a G-quadruplex-mediated toehold

Abstract

We developed a toehold-mediated DNA strand displacement that is driven only by a G-quadruplex. This strategy is able to be regulated by adjusting the PEG volume fraction, the G-quartet number and the G-quadruplex split mode. This method will provide new possible applications for DNA nanotechnology.

---

The unique properties of the DNA structure allow for the development of DNA nanotechnology that has been adapted for use in a wide variety of fields, such as catalytic circuits,^1–3 DNA computation,^4,5 material assembly,^6–8 drug delivery and biosensing.^9,10 Several applications are related to toehold-mediated strand displacement, which was first reported by Yurke et al. as a DNA nanomachine.¹¹ By this mechanism, a single-stranded DNA bound to its complementary target strand, and a short single-stranded called toehold overhung from the duplex. The invading strand was complementary to the toehold, as well as its adjacent part in duplex. The displacement was initiated at the toehold part. Hybridisation of the invading strand and the toehold led to the dissociation of target strand and substrate and the combination of invading strand and substrate. From naissance to now, optimisation of this process has developed rapidly due to the stability of the DNA structures involved, its wide applicability and its broad capacity to be designed for the requirements of specific uses.^12–14 However, most of the toehold systems were restricted to double helix DNA structures.

G-quadruplexes are formed by Guanine-rich oligonucleotides and are stabilised by Hoogsteen hydrogen bonds.^15,16 Tang and colleagues¹⁷ first developed a toehold activation strategy based on a DNA tetraplex. G-quadruplex toehold strand displacement was controlled using a complementary single-stranded segment and Sr²⁺. However, the reaction remained slow without the help of an extended segment or Sr²⁺, which may limit the versatility of the method. PEG 200, regarded as a molecular crowder^18,19 or hydrophobic reagent,^20,21 impacted the thermodynamics of DNA duplexes, triplexes, G-quadruplexes and other structures. Studies have indicated that PEG influences DNA structures disrupting hydrogen bonding between base pairs. The stability of DNA G-quadruplexes increased in PEG environments, while DNA duplexes were destabilised.^22,23 In this paper, we develop the G-quadruplex toehold-based strand displacement in the presence of K⁺ and Na⁺ in a PEG environment. These conditions allow for the reaction to be controlled by both different concentrations of PEG and changes in the structure of the G-quadruplex. In this way, the method will have wider suitability and can be tailored to more biological applications.

The method is illustrated in Scheme 1: the DNA substrate complex (SC) was assembled by annealing the substrate strand (S), migration strand (M) and toehold strand (T). Strand M served as the target DNA and was complementary to part of strand S.

Scheme 1 Mechanism of G-quadruplex-mediated strand displacement reaction.

Strand T contained three repeats of GGG as the toehold domain; the rest of strand T was bound to strand S. The invading DNA (I) had one GGG repeat, which could dock to the toehold domain of strand T and form a 3 : 1 split mode G-quadruplex. The dissociation of the SC and association of strand I and S was slow in aqueous phase but quite fast in the presence of PEG 200. We first determined the optimal concentration of PEG 200 to promote G-quadruplex toehold-mediated strand displacement. Because PEG stabilises DNA G-quadruplexes (Fig. S1A†) and destabilise DNA duplexes (Fig. S1B†), finding the proper volume fraction of PEG was crucial to increase the yield and decrease the background signal. Li⁺ was utilised to identify strand displacement that was mainly driven by the formation of G-quadruplexes rather than by the natural dissociation of double stranded DNA. Fig. S2† indicated that when the concentration of PEG was lower than 20%, the data collected under conditions containing Li⁺ revealed less than 18% variability. When the fraction of PEG was increased to 40%, the reaction is thought to be mainly driven by the instability of DNA duplexes (Fig. S1b†). Therefore, 20% PEG was selected as the appropriate concentration. Besides PEG 200, different polymerization degrees of PEG, respectively PEG 400 and PEG 600, were investigated (Fig. S3†). As the polymerization degrees of PEG increased, the reaction was promoted according to the crowding theory.²⁴ Li⁺ was utilised to identify the proper volume fraction. PEG 400 and PEG 600 resulted in the same proper volume fraction 10%. The reaction yield of the G-quadruplex-mediated strand displacement could reach 80%, while in PEG 200, the yield was only 60%. The resulted indicated that PEG with different polymerization degrees could influence the strand displacement.

To investigate whether the strand displacement could progress as predicted, a native polyacrylamide gel electrophoresis (PAGE) experiment was preformed (Fig. 1). By mixing purified SC (lane 5) and strand I in the absence of PEG, a small amount of the band shifted (lane 6) when compared to the SC and a slight amount of a new band (strand M) appeared, revealing that the G-quadruplex induced strand displacement in aqueous phase was indeed happening, albeit slowly. When PEG was added to the system, the reaction proceeded effectively as is evident by the observed more significant band shift of SC and the appearance of the new band associated with strand M (lane 7). In fact, these data are quite similar to the complete condition (lane 8). Circular dichroism spectroscopy analysis was performed to verify the formation of G-quadruplex in the strand displacement (Fig. 2). According to the CD spectra results, SC has a positive peak near 270 nm and a negative peak near 240 nm, illustrating a typical DNA double helix. Strand I, containing one GGG repeat, has a positive peak at approximately 280 nm; this result is likely due to the high concentration (20 μM) and the possibility of forming G–G mismatches or intermolecular G-quadruplexes. However, addition of strand I to SC resulted in band enhancement and a positive peak shift towards 265 nm. After subtracting the spectra of SC and strand I from the reaction spectrum, we reveal a new spectra with a small but obvious positive peak at approximately 265 nm and a negative peak approximately 245 nm, which is characteristic of a hybrid parallel G-quadruplex structure.²⁵

Fig. 1 20% native PAGE analysis of the G-quadruplex toehold-mediated strand displacement at 30 °C for 2 h. Lane 1: 750 nM strand S, lane 2: 750 nM strand T, lane 3: 750 nM strand M, lane 4: 7.5 μM strand I, lane 5: 750 nM substrate SC, lane 6: 750 nM SC and 7.5 μM I in aqueous phase, lane 7: 750 nM SC and 7.5 μM I in 20% PEG, lane 8: 750 nM SC and 7.5 μM I after annealed. The gel was photographed after stained by ethidium bromide.

Fig. 2 CD spectroscopy analysis of G-quadruplex structure formation in the displacement reaction. Initial concentrations: [substrate SC] = 2 μM, [strand I] = 20 μM, 25 mM HEPES-NH₃·H₂O (pH 8.0), 20 mM K⁺, 200 mM Na⁺, and 20% PEG. The experiment was carried out after 2 h incubation at 30 °C.

To explore how the concentration of PEG would impact G-quadruplex toehold-mediated strand displacement, fluorescence spectroscopy experiments were designed. Strand M labeled with a FAM fluorophore was quenched by a Dabcyl fluorophore covalently linked to Strand S in a purified SC. As strand M is replaced by strand I, the fluorescence intensity is expected to increase. As shown in Fig. 3, only 15% signal enhancement was observed in the aqueous phase. However, when the volume fraction of PEG was increased to 20%, fluorescence signal enhancement by strand displacement increased to 75%. The reaction rate increased as the PEG volume increased from 0 to 20%. Therefore, the fluorescence kinetics proved that the displacement was modulated by the concentration of PEG.

Fig. 3 Fluorescence kinetic experiments of G-quadruplex toehold-mediated strand displacement with varying concentrations of PEG. The reaction was started by quickly mixing strand I (7.5 μM) with substrate SC (750 nM). Concentrations of PEG were: 0, 5%, 10%, 15%, and 20%. The results were normalized by the equation: (F–F₀)/(F_M–F₀). F_M is the fluorescence intensity of 750 nM labeled strand M, F is the fluorescence intensity of each sample and F₀ refers to substrate SC in the absence of strand I.

Furthermore, to verify whether the number of G-quartet and the split mode G-quadruplexes could influence the displacement, three types of G-quadruplexes, (TGGTTGGTTGGTTGGT), (TGGGTTGGGTTGGGTTGGGT) and (TGGGGTTGGGGTTGGGGTTGGGGT), denoted 2G, 3G and 4G, respectively, were evaluated using the same strategy. Two types of split modes of G-quadruplex were utilised, including 3 : 1 (as demonstrated above) and 2 : 2. We did not consider the 1 : 3 split mode because the high concentration and the G-rich structure of strand I might have resulted in a complicated intramolecular or intermolecular secondary structure, thus complicating the reaction. In total, there were six conditions, including the 3 : 1 split modes for the two, three, and four G-quartets (denoted as 2G_3:1, 3G_3:1, 4G_3:1, respectively) and the 2 : 2 split modes for the two, three, and four G-quartets (denoted as 2G_2:2, 3G_2:2, 4G_2:2, respectively). The results were shown in Fig. 4. For both split modes, enhancement of strand M fluorescence was observed to be less for 2G when compared to 3G. Only 18% signal enhancement was observed in 2G_2:2 compared to 88% in 3G_2:2. The increasing strand displacement reflected the stability of the G-quadruplex, which was in accordance with the values of the melting temperature (Tm) (Fig. S5†). However, when the G-quartet number reached four, the result was harder to interpret. The enhancement of strand M was less for 4 G when compared to 3G, which is not in agreement with the results shown in Fig. S5.† 4G_2:2 decreased signal enhancement slightly when compared to 3G_2:2, and with 4G_3:1, the enhancement was only 24%. One possible explanation for these results was that the toehold domain of SC in 4G_3:1 had the most guanine nucleotides out of all of the conditions, which may have caused it to fold into a complicated secondary structure, especially in the PEG environment. Fig. S6† demonstrated the predictable structure, which proved to be a parallel G-quadruplex structure with a high positive peak in the CD spectra at approximately 265 nm and a negative peak approximately 245 nm. The Tm value of this structure had reached 70.30 °C. Therefore, TGGGGT of strand I was not enough to induce the dissociation of the structure at 30 °C, and the strand displacement was hindered. The result suggested that the G-quadruplex toehold-mediated strand displacement could be regulated by not only the number of G-quartets but also the split mode of the G-quadruplex.

Fig. 4 Impact of number of G-quartet and the split mode G-quadruplex on G-quadruplex toehold-mediated strand displacement. Normalized fluorescence was calculated according to Fig. S4.†

Different forms of G-quadruplex, respectively 2G, 3G and 4G, and different G-quadruplex toehold-mediated strand displacements, including 2G_3:1, 4G_3:1, 2G_2:2, 3G_2:2, 4G_2:2, were evaluated by CD spectroscopy analysis. The results were shown in Fig. S7.† It was indicated that when number of G-quartet was two, it was prone to form a weak antiparallel G-quadruplex structure. And when number of G-quartet was three and four, it was prone to form relatively strong parallel G-quadruplex structure (Fig. S7A†). In the G-quadruplex toehold-mediated strand displacement, after subtracting the spectra of SC and strand I from the reaction spectrum, the new spectra followed the same regulation. 2G_3:1 and 2G_2:2 revealed weak antiparallel G-quadruplex structure, 3G_3:1, 3G_2:2, 4G_3:1 and 4G_2:2 revealed obvious parallel G-quadruplex structure (Fig. 2 and S7B†). Considering the result showed in Fig. 4, 3G_2:2, 3G_3:1, 4G_2:2 contributed to the most reaction yields, which may indicate that different G-quadruplex structures could affect the displacement reaction and parallel G-quadruplex structure provide a favorable factor.

Conclusions

In summary, we constructed a type of toehold-mediated strand displacement induced only by G-quadruplex DNA. The reaction could progress in the presence of K⁺ and Na⁺. This strategy can be regulated by adjusting the concentration and polymerization degrees of PEG and by changing the number of G-quartets or the split mode of G-quadruplex. Because the product of the reaction contains a G-quadruplex, which could be used as a DNAzyme,²⁶ it has the potential to be used as a signal response in catalytic circuits rather than as a waste product. Different from the structure of the DNA double helix, which is usually considered as two dimensional, a DNA G-quadruplex extends the DNA nanostructure to three dimensions. PEG 200, often used to simulate the cellular environment, may expand the applications of toehold-mediate strand displacement into drug delivery and biosensing.

Acknowledgements

The authors thank the National Basic Research Program of China (973 Program) (2012CB720600, 2012CB720603), the National Science of Foundation of China (no. 91213302, 30973605, 21072115), the National Grand Program on Key Infectious Disease (2012ZX10003002-014).

Notes and references

1. D. Y. Zhang, A. J. Turberfield, B. Yurke and E. Winfree, Science, 2007, 318, 1121.
2. R. Penchovsky and R. R. Breaker, Nat. Biotechnol., 2005, 23, 1424.
3. G. Seelig, D. Soloveichik, D. Y. Zhang and E. Winfree, Science, 2006, 314, 1585.
4. E. Winfree, J. Biomol. Struct. Dyn., 2000, 112, 263.
5. C. Mao, T. LaBean, J. H. Reif and N. C. Seeman, Nature, 2000, 407, 493.
6. F. M. Pohl and T. M. Joyin, J. Mol. Biol., 1972, 67, 375.
7. B. Yurke, A. J. Turberfield, A. P. Mills, F. C. Jr Simmel and J. L. Neumann, Nature, 2000, 406, 605.
8. J. S. Shin and N. A. Pierce, J. Am. Chem. Soc., 2004, 126, 10834.
9. S. M. Douglas, I. Bachelet and G. M. Church, Science, 2012, 335, 831.
10. J. Li, H. Pei, B. Zhu, L. Liang, M. Wei, Y. He, N. Chen, D. Li, Q. Huang and C. H. Fan, ACS Nano, 2011, 5, 8783.
11. B. Yurke, A. J. Turberfield, A. P. Jr Mills, F. C. Simmel and J. L. Neumann, Nature, 2000, 406, 605.
12. A. J. Genot, D. Y. Zhang, J. Bath and A. J. Turberfield, J. Am. Chem. Soc., 2011, 133, 2177.
13. Y. Xing, Z. Yang and D. Liu, Angew. Chem., Int. Ed., 2011, 50, 11934.
14. M. You, L. Peng, N. Shao, L. Zhang, L. Qiu, C. Cui and W. Tan, J. Am. Chem. Soc., 2014, 136, 1256.
15. D. Sen and W. Gilbert, Nature, 1988, 334, 364.
16. A. K. Todd, M. Johnston and S. Neidle, Nucleic Acids Res., 2005, 33, 2901.
17. W. Tang, H. Wang, D. Wang, Y. Zhao, N. Li and F. Liu, J. Am. Chem. Soc., 2013, 135, 13628.
18. Z. Y. Kan, Y. Yao, P. Wang, X. H. Li, Y. H. Hao and Z. Tan, Angew. Chem., Int. Ed., 2006, 45, 1629.
19. B. Heddi and A. T. Phan, J. Am. Chem. Soc., 2011, 133, 9824.
20. D. Miyoshi, K. Nakamura, H. Tateishi-Karimata, T. Ohmichi and N. Sugimoto, J. Am. Chem. Soc., 2009, 131, 3522.
21. M. C. Miller, R. Buscaglia, J. B. Chaires, A. N. Lane and J. O. Trent, J. Am. Chem. Soc., 2010, 132, 17105.
22. D. Miyoshi, S. Matsumura, S. Nakano and N. Sugimoto, J. Am. Chem. Soc., 2004, 126, 165.
23. K. W. Zheng, Z. Chen, Y. H. Hao and Z. Tan, Nucleic Acids Res., 2010, 38, 327.
24. N. Samanta, D. D. Mahanta, S. Hazra, G. S. Kumar and R. K. Mitra, Biochimie, 2014, 104, 81.
25. R. Jin, B. L. Gaffney, C. Wang, R. A. Jones and K. Breslauer, Proc. Natl. Acad. Sci. U. S. A., 1992, 89, 8832.
26. P. Travascio, A. J. Bennet, D. Y. Wang and D. Sen, Chem. Biol., 1999, 6, 779.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra09516b

‡ S. Feng and F. Wu contributed equally to this work.

---