The influence of loops on the binding of the [Ru(phen)₂dppz]²⁺ light-switch compound to i-motif DNA structures revealed by time-resolved spectroscopy†

Frederico R. Baptista ^a, Stephen J. Devereux ^a, Sarah P. Gurung ^bc, James P. Hall ^bd, Igor V. Sazanovich ^e, Michael Towrie ^e, Christine J. Cardin ^b, John A. Brazier ^d, John M. Kelly ^f and Susan J. Quinn *^a
aSchool of Chemistry, University College Dublin, Dublin 4, Ireland. E-mail: susan.quinn@ucd.ie
^bDepartment of Chemistry, University of Reading, Whiteknights, Reading, Berks RG6 6AD, UK
^cDiamond Light Source, Chilton, Didcot, Oxfordshire, UK
^dSchool of Pharmacy, University of Reading, Whiteknights, Reading, Berks RG6 6AD, UK
^eCentral Laser Facility, Rutherford Appleton Laboratory, STFC, Harwell Campus, OX11 0QX, UK
^fSchool of Chemistry, Trinity College Dublin, Dublin 2, Ireland

First published on 16th July 2020

---

---

In 2018 confocal measurements confirmed the formation of i-motif DNA structures in the nuclei of human cells.¹ I-motif forming DNA is known to play an important role in the binding, recognition and regulation of transcriptional proteins in cells and is present in promoter regions of oncogenes.^2,3 The i-motif structure is characterized by intercalated hemi-protonated cytosine base-pairs (Fig. 1a and b)^4–7 and is typically formed in solution under slightly acidic conditions, though sequences prevalent in the human genome can form i-motifs at physiological pH.⁸ This structural response to pH has been used to develop an i-motif based aptamer for imaging of the weakly acidic microenvironment in tumours.⁹ In contrast to the quadruplex structure, there are relatively fewer reports of i-motif targeting ligands in the literature.^10–15 Significantly, i-motif binding compounds have been found to alter gene expression,² and to change telomerase activity.¹¹ The cationic porphyrin TMPyP4 (5,10,15,20-tetrakis-(N-methyl-4-pyridyl)) was one of the first molecules reported to bind to the i-motif structure.¹² Notably, the ability to distinguish TMPyP4 external groove binding from intercalation was demonstrated by transient spectroscopy.¹³ Fluorescent displacement assays have been used to screen new i-motif binding ligands¹⁴ and have revealed how the drug mitoxantrone can induce folding of i-motif forming DNA sequences.¹⁵ The recent discovery of the role of i-motif structures in the modulation of viral transcription has raised the possibility to selectively recognize the HIV-1LTR i-motif in drug design.¹⁶ The biological importance of i-motif DNA warrants the development of new molecular probes.

Ruthenium(II) polypyridyls and especially the ‘light-switch’ complex [Ru(phen)₂(dppz)]²⁺ (1), (dppz = dipyrido[3,2-a:2′,3′-c]phenazine) and related structures are very effective DNA probes due to their structural and photophysical properties.¹⁷ In aqueous solution solvent hydrogen bonds to the dppz pyrazine nitrogens renders 1 non-luminescent. However, 1 is highly luminescent in aprotic organic solvents and when bound to DNA, which shields the dppz ligand from water.^18,191 has been shown to bind to the human telomere i-motif sequence 5′-(CCCTAA)₃CCCT-3′.²⁰ A recent study also showed that the enantiomers of 1 can stabilize long loop i-motif structures, with increased emission with increasing loop length.²¹ Time-resolved infrared (TRIR) spectroscopy is a powerful technique for studying ultrafast processes in nucleic acid systems²² and has previously been used to characterize the excited state dynamics of cytosine-rich DNA,²³ including i-motif structures.²⁴ TRIR can reveal information about the binding site of 1 due to the changes in the characteristic ground state vibrations of DNA bases in the presence of the excited state, formed by visible light excitation.²⁵ These changes can be considered as a ‘Stark-like’ effect, which reports the response of bases to the redistribution of electron density in the excited state.²⁶ In this way it provides insights that transient visible absorption (TA) and steady-state or time resolved emission spectroscopy alone cannot. TRIR also reports on the distinct transient IR marker bands for the ‘bright’ (1396 cm⁻¹) and ‘dark’ (1290 cm⁻¹, 1321 cm⁻¹ and 1345 cm⁻¹) states of 1, which arise respectively, due to the localization of electron density on the phen (A, B & C) and phenazine (B, D & E) parts of the dppz rings, Fig. 1.²⁵

Here we first use TA and TRIR to study the excited state dynamics of 1 bound to unimolecular i-motif structures comprising a central run of three cytosine bases with different thymine loop lengths d(CCCT_n)₃CCC (n = 5 to 8, C₃T₅, C₃T₆, C₃T₇ and C₃T₈), see Fig. 1. The i-motif structures were annealed prior to addition of 1 and the formation of the i-motif was confirmed using circular dichroism (CD).²⁷ CD also indicates that the presence of the bound racemic complex is not found to significantly perturb the i-motif structure.²¹ TA and TRIR experiments were performed with 0.5 mM of rac-[Ru(phen)₂dppz]²⁺ (1) in the presence of 0.625 mM of the i-motif structures in aerated 50 mM potassium phosphate buffer at pH 5 in D₂O (Fig. S1, ESI†). This ratio ensures that the i-motif is present in slight excess and under pH conditions where it is known to form.^21,27

The ns-TA spectra recorded between 1 and 2000 ns upon 400 nm excitation of 1 in the presence of i-C₃T₈ at pH 5 are shown in Fig. 2a. Excitation results in the loss of the ground state MLCT absorption in the region of 440 nm (negative bleached band). The appearance of a broad transient absorbance feature with a maximum at 598 nm is attributed to an MLCT based excited state, which is reportedly dominated by low-lying π–π* transitions of the anionic radicals of the phen and dppz ligands.²⁸ The transient band underwent fully reversible decay, with concomitant recovery of the bleached ground state absorbance at 438 nm, without any photodamage observed. The ground state bleach recovery and transient decay was best fit by bi-exponential fits and yielded average lifetimes of 85 ± 10 ns and 603 ± 66 ns (Fig. 2b), which are more than two orders of magnitude longer than the value of 600 ps observed for the unbound 1 in aerated D₂O.¹⁹

The presence of a shorter component, on the tens of nanoseconds, and a longer component, on the hundreds of nanoseconds timescale was similarly obtained for i-C₃T₅, i-C₃T₆, and i-C₃T₇ (Fig. S2–S4, ESI†). Importantly, the longer component excited state lifetime of 1 increased as the loop length increased, from 190 ± 13 ns for i-C₃T₅ to 603 ± 44 ns for i-C₃T₈ (Table 1 and Fig. 3a). It is notable that there is not much difference from T5 to T6 and then large (and rather similar-sized) changes from T6 to T7 and T7 to T8 (Fig. 3a). The increased lifetime is related to the degree of protection from solvent afforded by the binding site²⁹ with the latter value approaching the value of 720 ns observed for 1 when intercalated in double-stranded natural DNA.³⁰ Notably, these lifetimes are significantly longer than those observed for 1 in the presence of the systems at pH 8, where the single-stranded form dominates²³ (Fig. 3b and Fig. S5, Table S1, ESI†). The different rates of excited state decay are highlighted in Fig. 3.

---

The TRIR obtained upon 400 nm excitation of i-C₃T₈ yields a highly structured spectrum with several transient and bleach bands, which fully decay/recover over 2 μs (Fig. 4). Below 1600 cm⁻¹ the spectrum is dominated by contributions from 1. A notable feature is the broad transient band at 1395 cm⁻¹ (yellow box) of the “bright” excited state of the DNA bound complex.²⁵ There are several bleached bands characteristic of cytosine and thymine in the DNA region of the TRIR spectrum (1600–1700 cm⁻¹). The bleach at 1644 cm⁻¹ (Thymine ν_ring), indicates the presence of thymine bases in the immediate vicinity of the excited state. The bleaches at 1667 cm⁻¹ and 1705 cm⁻¹ are attributed to a combination of thymine carbonyl bleaches 1655–1660 cm⁻¹ (ν_C4O4) and 1696 cm⁻¹ (ν_C2O2) and the carbonyl vibrations of the hemi-protonated cytosine bases, at 1666 cm⁻¹ and 1697 cm⁻¹ (shifted from 1650 cm⁻¹ at neutral pH) (Fig. 4).²⁴

Similar TRIR spectra were obtained for 1 in presence of i-C₃T₅, i-C₃T₆, and i-C₃T₇ (Fig. S6, ESI†). Notably, the intensity of the “bright” (1396 cm⁻¹) excited-state was found to decrease as the loop length decreased. This supports the observation that loop length in these i-motif structures is an important factor for the stabilization of the “bright” excited-state.²¹ Examination of the DNA region (green box) reveals similar bleach structure for all four systems (Fig. S7, ESI†). The intensity of the bleach bands is found to grow as the length of the thymine loop increased (Fig. S8, ESI†), which suggests that there is a greater interaction with the excited state of 1 as the length of the loop increases. This is attributed to the better accommodation of the dppz ligand in the i-motif structure with the improved interaction resulting in a greater perturbation of the bases and the stronger signal. Notably, the position of the bleach bands does not change for the different i-motifs but a greater intensity of the thymine ring bleach at 1644 cm⁻¹ is observed. Kinetic analysis of the ‘bright’ state band at 1395 cm⁻¹ reveals similar kinetics to those obtained from the TA measurements with an increase in the life-time as the loop length increases (Fig. S9, ESI†). A summary of the lifetimes of the complex measured are in Table S2 (ESI†).

A recent study on these i-motif structures revealed that binding of 1 results in an increase in the thermal melting temperature as the loop length increases.²¹ This stabilization effect indicates increased interaction with the i-motif structure, which correlates well with the observed increase in the excited state life-time found in our time resolved experiments. The binding mode is known to greatly influence the lifetime of 1.²⁹ The observed biexponential kinetics may arise due to the presence of more than one site of binding in the system (possibly due to different binding of the two enantiomers),²¹ which may differ in the access to water molecules. The TRIR indicates that 1 is bound in the vicinity of both cytosine and thymine. However, the observation of increasing lifetime and a stronger thymine ring bleach signal in the series i-C₃T₆, i-C₃T₇, and i-C₃T₈ suggests that the complex is bound more effectively to thymine in the longer loops, leading to a more protected dppz ligand. Interestingly, analysis of the data reveals the importance of the extension from i-C₃T₆ to i-C₃T₇, which may be related to the stability of the inherent i-motif structure, as reflected in the T_m decreasing from 43 to 39.5 °C. This may imply that better binding is obtained in the less stable i-motif structure that is better able to accommodate the complex. The complex may bind to either the central loop by stacking as a capping element at the open side of the structure (Fig. 5a) or to the two lateral loops (Fig. 5b), which is expected to provide good protection from the solvent molecules. The long thymine loops may well generate thymine–thymine base pairs, which have previously been simulated in i-motif structures^31a and shown to be stabilized by the dppz ligand of the related [Ru(TAP)₂(dppz)]²⁺ complex in quadruplex DNA (Fig. 5c).^31b

In the final part of the study TRIR was used to probe the interaction of 1 with the human telomeric i-motif sequence (Fig. S10, ESI†). This biologically relevant sequence (CCCTAA)₃CCC contains shorter loops comprising a thymine and two adenine bases. The TRIR spectrum recorded at 20 ps again shows an interaction of 1 with the loops. However, the presence of a strong bleach band at 1620 cm⁻¹ indicates an interaction with the adenine base while a relatively weaker bleach is observed for thymine (Fig. S11, ESI†) than is found with the i-C₃T_n systems. It is also notable that the “bright state” transient is weaker (relative to the bleach band at 1360 cm⁻¹). A lifetime of 180 ± 23 ns determined for the longer component (Fig. S12, ESI†) also confirms a weaker interaction with this sequence, which may be attributed to its shorter loop length.²⁰

In summary this study uses time-resolved techniques to profile the binding between 1 and i-motif structure containing different end loops. The TA experiments reveal that all four i-C₃T_n systems can bind to 1 resulting in strong enhancement of its excited state lifetime that was seen to trend with the length of the loop. By TRIR the marker band of the complex “bright” excited-state at 1395 cm⁻¹ was detected for all i-motif structures. The DNA region of the TRIR spectra demonstrates that 1 binds to the loops. For the i-C₃T_n systems the bleach band structure is characteristic of both thymine and cytosine bases for all samples with an increased contribution of the thymine ring bleach for the longer loop lengths consistent with improved accommodation of 1 in the binding site. The extent of the perturbation of the ground state vibrations of the DNA is expected to be sensitive to the proximity/overlap with the excited state. Thus, the increase in thymine signal is taken to represent better overlap in the site and demonstrates the important role of the loops in the binding of DNA intercalating compounds.

Information on the binding sites to i-motif structures is limited. One of the few examples in the literature used modified bases capable of FRET signalling to distinguish binding between the central and lateral loops.² The TRIR site effect has the advantage of distinguishing the presence of different bases in the loops (as was shown for the human telomer sequence), which demonstrates the potential to identify the binding for biologically relevant sequences. Crystallography has proved a powerful resource in resolving the interactions of 1 and related complexes to DNA. However, flexible looped structures, such as those in the systems here are not readily crystallized. Thus, in the absence of X-ray, computational and NMR structural studies, the TRIR measurements reported here allow the interactions of 1 with i-motif structures to be resolved.

This work was supported by BBSRC grants BB/K019279/1 and BB/M004635/1, STFC for program access to the CLF (App 13230047), UCD School of Chemistry, (FRB), the Irish Research Council (SJD: GOIPG/2016/805). We thank Fergus Poynton and Thorfinnur Gunnlaugsson for the generous gift of 1.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. M. Zeraati, D. B. Langley, P. Schofield, A. L. Moye, R. Rouet, W. E. Hughes, T. M. Bryan, M. E. Dinger and D. Christ, Nat. Chem., 2018, 10, 631–637 CrossRef CAS PubMed .
2. S. Kendrick, H. J. Kang, M. P. Alam, M. M. Madathil, P. Agrawal, V. Gokhale, D. Yang, S. M. Hecht and L. H. Hurley, J. Am. Chem. Soc., 2014, 136, 4161–4171 CrossRef CAS PubMed .
3. H. J. Kang, S. Kendrick, S. M. Hecht and L. H. Hurley, J. Am. Chem. Soc., 2014, 136, 4172–4185 CrossRef CAS PubMed .
4. S. Benabou, S. Avinó, R. Eritja, C. González and R. Gargallo, RSC Adv., 2014, 4, 26956–26980 RSC .
5. (a) K. Gehring, J. L. Leroy and M. Gueron, Nature, 1993, 363, 561–565 CrossRef CAS PubMed ; (b) J. L. Leroy, K. Gehring, A. Kettani and M. Guéron, Biochemistry, 1993, 32, 6019–6031 CrossRef CAS PubMed .
6. (a) A. M. Fleming, Y. Ding, R. A. Rogers, J. Zhu, J. Zhu, A. D. Burton, C. B. Carlisle and C. J. Burrows, J. Am. Chem. Soc., 2017, 139, 4682–4689 CrossRef CAS PubMed ; (b) R. A. Rogers, A. M. Fleming and C. J. Burrows, ACS Omega, 2018, 3, 9630–9635 CrossRef CAS PubMed .
7. J. A. Brazier, A. Shah and G. D. Brown, Chem. Commun., 2012, 48, 10739–10741 RSC .
8. B. M. Mir, I. Serrano, D. Buitrago, M. Orozco, N. Escaja and C. González, J. Am. Chem. Soc., 2017, 139, 13985–13988 CrossRef CAS PubMed .
9. Y. Lei, X. He, J. Tang, H. Shi, D. He, L. Yan, J. Liu, Y. Zeng and K. Wang, Chem. Commun., 2018, 54, 10288–10291 RSC .
10. (a) H. A. Day, P. Pavlou and Z. A. E. Waller, Bioorg. Med. Chem., 2014, 22, 4407–4418 CrossRef CAS PubMed ; (b) S. Satpathi, S. Sappati, K. Das and P. Hazra, Org. Biomol. Chem., 2019, 17, 5392–5399 RSC ; (c) B. Shu, J. Cao, G. Kuang, J. Qiu, M. Zhang, Y. Zhang, M. Wang, X. Li, S. Kang, T.-M. Ou, J.-H. Tan, Z.-S. Huang and D. Li, Chem. Commun., 2018, 54, 2036–2039 RSC ; (d) L. Xu, J. Wang, N. Sun, M. Liu, Y. Cao, Z. Wang and R. Pei, Chem. Commun., 2016, 52, 14330–14333 RSC ; (e) S. N. Journey, S. L. Alden, W. M. Hewitt, M. L. Peach, M. C. Nicklaus and J. S. Schneekloth Jr, Med. Chem. Commun., 2018, 9, 2000–2007 RSC .
11. Y. Chen, K. Qu, C. Zhao, L. Wu, J. Ren, J. Wang and X. Qu, Nat. Commun., 2012, 3, 1074–1087 CrossRef PubMed .
12. O. Y. Rangan, V. V. Chemeris and L. H. Hurley, Biochemistry, 2000, 39, 15083 CrossRef PubMed .
13. T. Qin, K. Liu, D. Song, C. Yang and H. Su, Chem. – Asian J., 2017, 12, 1578–1586 CrossRef CAS PubMed .
14. Q. Sheng, J. C. Neaverson, T. Mahmoud, C. E. M. Stevenson, S. E. Matthews and Z. A. E. Waller, Org. Biomol. Chem., 2017, 15, 5669 RSC .
15. E. P. Wright, H. A. Day, A. M. Ibrahim, J. Kumar, L. J. E. Boswell, C. Huguin, C. E. M. Stevenson, K. Pors and Z. A. E. Waller, Sci. Rep., 2016, 6, 39456 CrossRef CAS PubMed .
16. E. Ruggiero, S. Lago, P. Šket, M. Nadai, I. Frasson, J. Plavec and S. N. Richter, Nucleic Acids Res., 2019, 47, 11057–11068 CrossRef CAS PubMed .
17. C. J. Cardin, J. M. Kelly and S. J. Quinn, Chem. Sci., 2017, 8, 4705–4723 RSC .
18. A. E. Friedman, J. C. Chambron, J. P. Sauvage, N. J. Turro and J. K. Barton, J. Am. Chem. Soc., 1990, 112, 4960–4962 CrossRef CAS .
19. E. J. C. Olson, D. Hu, A. Hormann, A. M. Jonkman, M. R. Arkin, E. D. A. Stemp, J. K. Barton and P. F. Barbara, J. Am. Chem. Soc., 1997, 119, 11458–11467 CrossRef CAS .
20. S. Shi, J. Zhao, X. Geng, T. Yao, H. Huang, T. Liu, L. Zheng, Z. Li, D. Yang and L. Ji, Dalton Trans., 2010, 39, 2490–2493 RSC .
21. B. J. Pages, S. P. Gurung, K. McQuaid, J. P. Hall, C. J. Cardin and J. A. Brazier, Front. Chem., 2019, 7, 744 CrossRef CAS PubMed .
22. (a) M. Towrie, G. W. Doorley, M. W. George, A. W. Parker, S. J. Quinn and J. M. Kelly, Analyst, 2009, 134, 1265–1273 RSC ; (b) D. B. Bucher, B. M. Pilles, T. Carell and W. Zinth, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 4369–4374 CrossRef CAS PubMed .
23. P. M. Keane, M. Wojdyla, G. W. Doorley, J. M. Kelly, I. P. Clark, A. W. Parker, G. M. Greetham, M. Towrie, L. M. Magno and S. J. Quinn, Phys. Chem. Chem. Phys., 2012, 14, 6307–6311 RSC .
24. (a) P. M. Keane, M. Wojdyla, G. W. Doorley, J. M. Kelly, A. W. Parker, I. P. Clark, G. M. Greetham, M. Towrie, L. M. Magno and S. J. Quinn, Chem. Commun., 2014, 21, 2990–2992 RSC ; (b) P. M. Keane, F. R. Baptista, S. P. Gurung, S. J. Devereux, I. V. Sazanovich, M. Towrie, J. A. Brazier, C. J. Cardin, J. M. Kelly and S. J. Quinn, ChemPhysChem, 2016, 17, 1281–1287 CrossRef CAS PubMed .
25. F. E. Poynton, J. P. Hall, P. M. Keane, C. Schwarz, I. V. Sazanovich, M. Towrie, T. Gunnlaugsson, C. J. Cardin, D. J. Cardin, S. J. Quinn, C. Long and J. M. Kelly, Chem. Sci., 2016, 7, 3075–3084 RSC .
26. S. D. Fried and B. S. G. Boxer, Acc. Chem. Res., 2015, 48, 998–1006 CrossRef CAS PubMed .
27. S. P. Gurung, C. Schwarz, J. P. Hall, C. J. Cardin and J. A. Brazier, Chem. Commun., 2015, 51, 5630–5632 RSC .
28. C. G. Coates, P. Callaghan, J. J. McGarvey, J. M. Kelly, L. Jacquet and A. Kirsch-De Mesmaeker, J. Mol. Struct., 2001, 598, 15 CrossRef CAS .
29. A. W. McKinley, P. Lincoln and E. M. Tuite, Dalton Trans., 2013, 42, 4081–4090 RSC .
30. Y. Jenkins, A. E. Friedman, N. T. Turro and J. K. Barton, Biochemistry, 1992, 31, 10809–10816 CrossRef CAS PubMed .
31. (a) R. P. Singh, R. Blossey and F. Cleri, Biophys. J., 2013, 105, 2820–2831 CrossRef CAS PubMed ; (b) K. McQuaid, J. P. Hall, L. Baumgaertner, D. J. Cardin and C. J. Cardin, Chem. Commun., 2019, 55, 9116–9119 RSC .

---

Footnote

† Electronic supplementary information (ESI) available: Further details, complex synthesis and data analysis methods. See DOI: 10.1039/d0cc03702h

This journal is © The Royal Society of Chemistry 2020