An electrochemical study based on thymine–Hg–thymine DNA base pair mediated charge transfer processes

Abstract

A DNA monolayer composed of a double strand of 25 base pairs was adsorbed onto a Au surface. Matched base pairs were present at the gold surface and distal part, while (TT)_n mismatched pairs were present at the middle of DNA monolayer. Methylene blue was attached to the end DNA monolayer via a C₆ alkyene linker and acted as the electrochemical probe. Upon the addition of mercury ions (Hg(II)), thymine–Hg–thymine ((T–Hg(II)–T)_n) was formed through metal coordination between DNA duplexes. The charge transfer (CT) properties of the DNA monolayer were studied using Laviron's theory, and it was found that when n ≤ 6, the kinetics of CT followed the order: matched DNA base pairs < DNA duplexes with (TT)_n < DNA duplexes with (T–Hg(II)–T)_n, and the CT kinetics increased with increasing n. The conformation of the DNA monolayer adsorbed onto the Au (111) was investigated using atomic force microscopy (AFM), and it was found that a duplex structure was retained when n ≤ 6. The formation of (T–Hg(II)–T)_n brought about a smoother DNA monolayer surface than that composed of matched DNA base pairs. However, an interesting phenomenon was found when n ≥ 12, that is the (T–Hg(II)–T)_n complex was formed between different DNA strands, and this inter-DNA (T–Hg(II)–T)_n structure caused DNA monolayer deformation, so CT could not occur along the DNA base pairs.

---

Introduction

The molecular π-stacking of base pairs within double-strand DNA (dsDNA) has been shown to mediate charge transfer (CT) reactions, and this has received substantial attention because of its biological significance in DNA damage and DNA repair mechanisms as well as the development of DNA electronic devices.¹ Because this CT property of DNA depends on coupling within the base pair stacking, DNA CT is sensitive to static and dynamic perturbations in base pair structures.² Several techniques or types of assemblies have been employed for studying DNA-mediated CT:³ (a) an electrochemical method, where molecular self-assembly of dsDNA on an electrode was used to mediate electron transfer between the electrode and the redox active intercalator bound to the DNA at a site remote from the electrode surface;^4–6 (b) DNA-mediated CT (hole) between a photoexcited fluorophore and quencher;^7,8 and (c) DNA-mediated CT (oxidative radicals) between a photoexcited oxidizer and oxidizable groups (DNA bases, e.g. guanine).^9,10

Comparatively, electrochemical detection of base pair stacking perturbations may enable diagnostic applications of DNA CT due to the simplicity of the electrochemical methods.² Usually, electrochemical reduction of the distantly bound intercalator was conducted to study the CT through the base pair stacking under a negative potential. Perturbation in base pair stacking would cause a significant loss in CT efficiency, and the sensitivity to base pair stacking provides the foundation for the application of DNA CT in probing DNA structures.¹¹ Thus, DNA mediated CT processes have been applied in sensing DNA structures such as base mismatch in fully hybridized DNA duplexes,¹² conformation of the DNA (A-, B-, and Z-form DNA),¹¹ base lesions,¹³ and binding of proteins¹⁴ or cisplatin to the DNA.¹⁵

This DNA structure-dependent CT electrochemical sensitivity may find application in biological processes¹⁶ such as repair of photo-induced DNA lesions by the CT process, localization of oxidative lesions and the corresponding mechanisms for antioxidative protection.¹⁶ In the last case, it is well known that oxidative lesions of DNA caused by reactive oxygen species (ROS) are the most widespread form of damage in living organisms. Hydroxyl radicals (˙OH), singlet oxygen, and superoxide anions are the main ROS ingenuously inducing DNA oxidation damage.^17–19 Based on this mechanism, we developed a TiO₂@γ-Fe₂O₃/(AT)_n/dsDNA/Au composite monolayer assembly to read the diffusion distance of ˙OH.²⁰

A lot of studies have been reported on DNA base pairing via metal coordination acting as the driving force and as an alternative to hydrogen bonding.²¹ The metal coordination is a stronger bonding, and DNA duplexes possessing such metallo-base pairs usually exhibited higher thermal stability than natural hydrogen-bonded DNAs. Such type of metal-responsive functional DNA molecules may find use in DNAzymes and DNA machines.²¹ It is well-known that in the presence of Hg(II) ions, a thymine–thymine (TT) mismatch pair forms a neutral metallo-base pair (T–Hg(II)–T), as originally proposed previously.^22,23 These T–Hg(II)–T complexes have higher thermal stability in comparison with matching DNA base pairs such as A–T.²³ The T–Hg(II)–T chemistry was of significance in probing defects in DNA and the assay of Hg(II) in the environment. However, studies on the effect of metal-mediated DNA base pairing on CT processes of DNA are rarely reported. The conductance measurements of DNA duplexes containing one H–Cu(II)–H base pair have been reported to be comparable to those of the natural DNA duplexes, indicating that the H–Cu(II)–H base pair favors CT similarly to natural DNA matching base pairs.²⁴ Theoretical and fluorescence studies found that for a DNA duplex containing T–Hg(II)–T, the overlap of the bases was favorable for CT at low temperatures. CT was driven by Hg changing the spatial overlap of bases.²⁵ At higher temperature, CT efficiency increased due to thermal motions for all DNA duplexes, and the matched DNA had the highest CT efficiency.²⁵ Despite these efforts being devoted to understand the CT of metal-mediated DNA base pairing, there are still fundamental issues that have not been explained.

Here, based on a DNA mediated electrochemical process, the CT kinetics were compared in standard dsDNA, dsDNA with mismatched T–T base pairs, and dsDNA with T–Hg(II)–T base pairs. Double strands of 25 base pairs were adsorbed onto a gold surface to form a monolayer, whereas at the middle of DNA monolayer was (TT)_n mismatch pairs. In the presence of Hg(II) ions, T–Hg(II)–T was formed in the DNA duplexes. At a modified electrode surface, methylene blue (MB) was attached to the end of the dsDNA. Under a negative potential, from the electrochemical reduction currents of MB, we could determine CT properties of the DNA duplexes.

Gel electrophoresis

Fig. 1 is the gel electrophoresis analysis of the abovementioned samples. From the gel electrophoresis image of the fluorescence signal, it is shown that matched DNA, dsDNA with (TT)₁, dsDNA with (TT)₃, and dsDNA with (TT)₆ can be hybridized to form duplexes. In comparison, the dsDNA with a (TT)₁₂ sequence was less stable than the other dsDNAs, indicating that the DNA sequence with (TT)₁₂ is not able to hybridize.

Fig. 1 Gel electrophoresis image for assay of the DNA duplexes. Lanes 1 and 6: mark, matched DNA, dsDNA with (TT)₁, dsDNA with (TT)₃, dsDNA with (TT)₆, and dsDNA with (TT)₁₂, respectively.

Surface density and coverage of adsorbed DNA monolayer

Cationic redox molecules of Ru(NH₃)₆³⁺ were electrostatically associated with the anionic DNA backbone and provided charge compensation for the backbone at a low ionic strength of electrolyte. The surface density of DNA probes on the surface was calculated from the number of cationic redox (Ru(NH₃)₆³⁺) and determined by applying the integrated Cottrell equation,²⁶where n is the number of electrons per molecule for reduction, F is the Faraday constant (C per equiv.), A is the electrode area (cm²), D₀ is the diffusion coefficient (cm² s⁻¹), C^*₀ is the bulk concentration (mol cm⁻²), Q_dl is the capacitive charge (C), and nFΓ₀ is the charge from the reduction of Γ₀ (mol cm⁻²) of the adsorbed Ru(NH₃)₆³⁺. The term Γ₀ designates the surface excess and represents the amount of Ru(NH₃)₆³⁺ confined near the electrode surface. The chronocoulometric intercept at t = 0 is then the sum of the double layer charging and the surface excess terms. The surface excess is determined from the difference in chronocoulometric intercepts for the identical potential step experiment in the presence and absence of Ru(NH₃)₆³⁺ (Fig. 2).

Fig. 2 Chronocoulometric response curves for the dsDNA probe modified electrode in 10 mM Tris buffer (pH 7.4) in the presence and absence of 50 μM Ru(NH₃)₆³⁺ (a) and interacting with Hg(II) (b).

The DNA probe was adsorbed onto the Au (111) facet of the Au electrode surface via the formation of the Au–S complex. In the presence of Hg(II) ions, if the thiol groups of the DNA probe react with the Hg(II) ions, the DNA will be removed from the Au surface, resulting in a lower surface density. It was found that the surface density of the well-matched DNA monolayer measured was about 1 × 10¹³ molecules per cm² and did not change evidently after the DNA modified electrode was immersed in 10 nM HgCl₂ solution for 30 min. The self-assembled DNA monolayer on the Au surface used in this work was stable in the presence of Hg(II) ions. Taking the diameter of the DNA helix as 2 nm, the surface coverage of the DNA monolayer over the Au surface was about 40%, indicating a highly packed monolayer.

AFM imaging

From the AFM image of Au (111), it is shown that the surface is an atomically flat multi-stage facets and the roughness of the facet is 0.15 nm (Fig. 3a). DNA monolayers were examined to visualize the DNA surface coverage and distribution of individual helices within the monolayer. A typical AFM image of the matched dsDNA monolayer is shown in Fig. 3b and reveals no large-domain clustering (packed into cluster). Notably, the images show some monolayer stratification, which is consistent with a compact film structure, and small clusters of DNA that are remarkably uniform in size (20 nm) and shape. The number of DNA helices present in a microcluster is approximately 6.25 × 10¹² molecules per cm², which is consistent with that obtained from chronocoulometry.

Fig. 3 AFM image of DNA monolayer on Au (111): (a) fresh Au (111), (b) matched dsDNA, (c) dsDNA with (TT)₃, (d) dsDNA with (T–Hg(II)–T)₃, (e) dsDNA with (TT)₆, (f) dsDNA with (T–Hg(II)–T)₆, (g) dsDNA with (TT)₁₂, and (h) dsDNA with (T–Hg(II)–T)₁₂.

The DNA monolayers with (TT)₃ (Fig. 3c) and (TT)₆ (Fig. 3e) mismatches also showed uniform surface coverage and distribution of microdomains. For the (TT)₃ duplexes, after being treated with Hg(II) ions (Fig. 3d), the heights of the DNA duplexes increased slightly, as observed from the height distribution, indicating that the existence of the T–Hg(II)–T complexes in the DNA duplexes increased their rigidity. However, for DNA duplexes with (TT)₆ mismatches, after the formation of the T–Hg(II)–T complexes (Fig. 3f), the height of the DNA monolayer decreased, so maybe the DNA duplexes were bent after formation of the (T–Hg(II)–T)₆ complexes. This means that a (T–Hg(II)–T)₆ structure may adopt a less rigid straight conformation. However, the SAM still showed a compact monolayer coverage distribution.

The DNA monolayer with (TT)₁₂ mismatches (Fig. 3g) showed quite a different morphology from the matched DNA SAM and the monolayers containing (TT)₃ or (TT)₆ mismatches. DNA molecules containing (TT)₁₂ easily tilted or even lay upon the Au (111), thus leading to measured height values compatible with the axial width of the molecule instead of its end-to-end length, yet still showed a compact coverage. DNA molecules containing (TT)₁₂ may not be able to hybridize well, and the (TT)₁₂ sequence might split, causing the DNA molecules to lie down. Upon the addition of Hg(II) ions (Fig. 3h), DNA molecules lying upon the Au (111) shrank and appeared as an aggregated DNA wire at the Au (111) surface. The height of the DNA wire was about that of the axial width, and the Au (111) substrate was exposed, indicating that T–Hg(II)–T complexes may form between different DNA molecules and cause aggregation into larger DNA clusters (Scheme 1). In this case, DNA molecules containing (TT)₁₂ and (T–Hg(II)–T)₁₂ were not suitable for DNA duplex mediated electrochemical CT studies.

Scheme 1 CT process of dsDNA with (T–Hg(II)–T) sequence: (a) matched dsDNA, (b) dsDNA with (T–Hg(II)–T)₁, (c) dsDNA with (T–Hg(II)–T)₃, (d) dsDNA with (T–Hg(II)–T)₆, and (e) dsDNA with (T–Hg(II)–T)₁₂. (T–Hg(II)–T); matched DNA bases; C₆ linker; methylene blue.

The electron-transfer rate of the DNA self-assembled monolayers

The CT rates were determined by applying Laviron analysis to CV data (Fig. 4 and 5) acquired at scan rates ranging from 50 mV s⁻¹ to 15 V s⁻¹:

log k = α log(1 − α) + (1 − α)log α − log(RT/nFv) − α(1 − α)nFΔE_p/2.3RT (2)

where k is the rate constant of the electron transfer (s⁻¹), n is the number of electrons per molecule for reduction, F is the Faraday constant (C per equiv.), T is the thermodynamic temperature (K), and R is the gas constant (J mol⁻¹ K⁻¹).

Fig. 4 Cyclic voltammetry of dsDNA with (T–T)₁ (a) and (T–Hg(II)–T)₁ (b) in 10 mM PBS (pH 7.4) with different scan rates.

Fig. 5 The fitting line of E_pa − E_1/2 vs. log(V) with different samples: matched dsDNA (a1) treated with Hg(II) ions (a2), dsDNA with (TT)₁ (b1) and (T–Hg(II)–T)₁ (b2), dsDNA with (TT)₃ (c1) and (T–Hg(II)–T)₃ (c2), and dsDNA with (TT)₆ (d1) and with (T–Hg(II)–T)₆ (d2).

For DNA molecules containing no TT mismatches, the addition of Hg²⁺ ions increased CT kinetics. Matched DNA molecules containing AT base pairs, which will form A–Hg–T complex.²¹ TT mismatches increased CT kinetics, and the T–Hg(II)–T complex increased the kinetic further (Table 1). However, DNA molecules containing (TT)₁₂ and (T–Hg(II)–T)₁₂ were not suitable for DNA duplex mediated electrochemical CT studies (Scheme 1).

Table 1 The charge transfer rates (k/s⁻¹) of dsDNA with different (TT)_n or (T–(Hg)–T)_n

	(TT)0	(T–(Hg)–T)0	(TT)1	(T–(Hg)–T)1	(TT)3	(T–(Hg)–T)3	(TT)6	(T–(Hg)–T)6
k/s^−1	17.63	29.75	20.03	37.13	21.57	34.54	32.89	37.58

---

Evidently, (TT)_n mismatches or (T–Hg(II)–T)_n complexes increased CT kinetics. DNA base pair mediated CT process is conformation-dependent, and the redox probe molecules are also capable of being reduced directly by the surface of the electrode (direct CT process).^27,28 The extent to which this mechanism contributes to the observed signal was found to be directly influenced by assembly conditions, the conformation of the SAM or the DNA packing density. In this work, from AFM, DNA molecules containing (TT)_n≤6 form compact DNA SAMs. Therefore, the CT kinetics may be regarded as dependent on the base pair or T–Hg(II)–T complex mediated process and not on the direct electrochemical action of the attached redox probe molecules. From AFM, the DNA SAM containing (TT)_n or (T–Hg(II)–T)_n is bent and the height is decreased slightly, so the direct CT process may contribute to the faster CT kinetics. However, this can be avoided because high scan rates were used in the CV study, and it has been reported that increasing scan rates can minimize contributions from the direct CT process.²⁹ The (TT)_n may have more π-overlapping compared with matched base pairs, and the intercalation of Hg(II) into TT may further increase this overlapping. It has been reported that an ionic complex adsorbed onto the DNA backbone facilitates CT processes.²⁸ Hg(II) ions form T–Hg(II)–T complexes and do not adsorb onto the backbone, so the CT mechanism may originate from the mediation of π-overlapping of the T–Hg(II)–T complex.

In summary, dsDNA monolayers with different (TT)_n mismatched sequences were assembled onto Au (111) facet on a Au electrode. The conformation or morphology and kinetics of the CT process showed a (TT)_n sequence-dependence. When n ≤ 6, dsDNA with (TT)_n can hybridize well to form a duplex structure. The (TT)_n might have more π-overlapping than the corresponding matched base pairs, and the intercalation of Hg(II) into TT may further increase this overlapping, causing a faster CT kinetics. When n ≥ 12, the dsDNA with (TT)₁₂ sequence cannot hybridize to form a duplex. In the presence of Hg(II), DNA strands hybridize via (T–Hg(II)–T)_n formation between different DNA molecules, and in this case, dsDNA with (TT)₁₂ was not suitable for the study of CT processes via π-stacking within the dsDNA.

Acknowledgements

J.L. thanks the Natural Science Foundation of China for Funding (20110706) and Open Project Program of Shandong Provincial Key Lab of Test Technology on Food Quality and Safety.

Notes and references

1. J. C. Genereux and J. K. Barton, Chem. Rev., 2010, 110, 1642.
2. S. Delaney and J. K. Barton, J. Org. Chem., 2003, 68, 6475.
3. J. C. Genereux, A. K. Boal and J. K. Barton, J. Am. Chem. Soc., 2010, 132, 891.
4. S. O. Kelley, J. K. Barton, N. M. Jackson and M. G. Hill, Bioconjugate Chem., 1997, 8, 31.
5. S. O. Kelley, N. M. Jackson, M. G. Hill and J. K. Barton, Angew. Chem., Int. Ed., 1999, 38, 941.
6. E. M. Boon, N. M. Jackson, M. D. Wightman, S. O. Kelley, M. G. Hill and J. K. Barton, J. Phys. Chem. B, 2003, 107, 11805.
7. C. J. Murphy, M. R. Arkin, Y. Jenkins, N. D. Ghatlia, S. H. Bossmann, N. J. Turro and J. K. Barton, Science, 1993, 262, 1025.
8. C. J. Murphy, M. R. Arkin, N. D. Ghatlia, S. H. Bossman, N. J. Turro and J. K. Barton, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 5315.
9. D. B. Hall, R. E. Holmlin and J. K. Barton, Nature, 1996, 382, 731.
10. K. Kawai, H. Kodera and T. Majima, J. Am. Chem. Soc., 2010, 132, 14216.
11. E. M. Boon and J. K. Barton, Bioconjugate Chem., 2003, 14, 1140.
12. S. O. Kelley, E. M. Boon, J. K. Barton, N. M. Jackson and M. G. Hill, Nucleic Acids Res., 1999, 27, 4830.
13. A. K. Boal and J. K. Barton, Bioconjugate Chem., 2005, 16, 312.
14. E. M. Boon, J. E. Salas and J. K. Barton, Nat. Biotechnol., 2002, 20, 282.
15. E. L. S. Wong and J. J. Gooding, J. Am. Chem. Soc., 2007, 129, 8950.
16. F. Boussicault and M. Robert, Chem. Rev., 2008, 108, 2622.
17. B. Halliwell and O. I. Aruoma, FEBS Lett., 1991, 281, 9.
18. J. A. Imlay and S. Linn, Science, 1988, 240, 1302.
19. M. S. Cooke, M. D. Evans, M. Dizdaroglu and J. Lunec, FASEB J., 2003, 17, 1195.
20. Q. Guo, Q. Yue, J. Zhao, L. Wang, H. Wang, X. Wei, J. Liu and J. Jia, Chem. Commun., 2011, 47, 11906.
21. Y. Takezawa and M. Shionoya, Acc. Chem. Res., 2012, 45, 2066.
22. S. Katz, Biochim. Biophys. Acta, 1963, 68, 240.
23. Y. Miyake, H. Togashi, M. Tashiro, H. Yamaguchi, S. Oda, M. Kudo, Y. Tanaka, Y. Kondo, R. Sawa, T. Fujimoto, T. Machinami and A. Ono, J. Am. Chem. Soc., 2006, 128, 2172.
24. S. Liu, G. H. Clever, Y. Takezawa, M. Kaneko, K. Tanaka, X. Guo and M. Shionoya, Angew. Chem., Int. Ed., 2011, 50, 8886.
25. I. Kratochvílová, M. Golan, M. Vala, M. Špérová, M. Weiter, O. Páv, J. Šebera, I. Rosenberg, V. Sychrovský, Y. Tanaka and F. M. Bickelhaupt, J. Phys. Chem. B, 2014, 118, 5374.
26. A. B. Steel, T. M. Herne and M. Tarlov, J. Anal. Chem., 1998, 70, 4670.
27. C. G. Pheeney and J. K. Barton, Langmuir, 2012, 28, 7063.
28. A. Abi and E. E. Ferapontova, J. Am. Chem. Soc., 2012, 134, 14499.
29. W. Yang and R. Y. Lai, Langmuir, 2011, 27, 14669.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra06238a

---