Fluorescent Visualization of Nucleolar G-Quadruplex RNA and Dynamics of Cytoplasm and Intranuclear Viscosity

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Nov 2021Fluorescent Visualization of Nucleolar G-Quadruplex RNA and Dynamics of Cytoplasm and Intranuclear Viscosity Le Yu†, Peter Verwilst†, Inseob Shim, Yu-Qiang Zhao, Ying Zhou and Jong Seung Kim Le Yu† College of Chemical Science and Technology, Yunnan University, Kunming 650091 Department of Chemistry, Korea University, Seoul 02841 †L. Yu and P. Verwilst contributed equally to this work.Google Scholar More articles by this author , Peter Verwilst† KU Leuven, Rega Institute for Medical Research, Medicinal Chemistry, B-3000 Leuven †L. Yu and P. Verwilst contributed equally to this work.Google Scholar More articles by this author , Inseob Shim Department of Chemistry, Korea University, Seoul 02841 Google Scholar More articles by this author , Yu-Qiang Zhao College of Chemical Science and Technology, Yunnan University, Kunming 650091 Google Scholar More articles by this author , Ying Zhou *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] College of Chemical Science and Technology, Yunnan University, Kunming 650091 Google Scholar More articles by this author and Jong Seung Kim *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] Department of Chemistry, Korea University, Seoul 02841 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.020.202000479 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The nucleolus, the locus of ribosome biogenesis, was found to be the predominant intracellular target of a new fluorescent probe, V-P1. In solution, the probe demonstrated both a selectivity to RNA G-quadruplexes and a sensitivity to the viscosity, while G-quadruplex binding did not disturb the viscosity sensing. In cells, confocal and fluorescence lifetime imaging, combined with digestion and competition experiments, lent support to the hypothesis of an RNA-based G-quadruplex as the intracellular target, postulated to be nucleolar ribosomal RNA (rRNA). The probe demonstrated a high sensitivity to viscosity in both the cytoplasm and the nuclear compartment and was used to precisely interrogate the viscosity changes resulting from diverse stimuli, such as temperature, monensin treatment, and etoposide-induced apoptosis. Owing to the putative rRNA G-quadruplex binding in vitro and in vivo, and further combined with a relatively low degree of toxicity, the dye enabled the interrogation of cytoplasm and intranuclear viscosity changes under diverse conditions and found applications in studying the influence and significance of cytoplasm and intranuclear viscosity as well as in gaining insight into the native secondary structure of rRNA in nucleoli. Download figure Download PowerPoint Introduction Ribosome biogenesis in the nucleoli is controlled by RNA polymerase I, transcribing ribosomal RNA (rRNA) genes into pre-rRNAs, that are further processed into mature rRNA.1 As ribosomal availability is directly linked to a cell’s ability for de novo protein synthesis and thus regulates cell growth and proliferation,2 rRNA is a crucial component of the cellular machinery and an attractive, albeit underexplored, target for drug design.3 Despite the importance of this species, several key aspects of the regulation of rRNA biosynthesis and general structural properties of this polynucleotide are still being discovered. The GC-rich rDNA is known to have a propensity toward the formation of DNA G-quadruplexes, a noncanonical planarly orientated sheet of four guanines bonded in a Hoogsteen hydrogen-bonding arrangement, stacked, and further stabilized by interspersed cations such as Na+ and K+.4 These G-quadruplexes have been demonstrated to play an important role in the regulation and epigenetics of rRNA transcription.5 Apart from the well-known G-quadruplex DNA structures, which have been studied extensively by Teulade-Fichou et al.6–9, RNA has been shown to arrange in G-quadruplex structures as well, and may be even more prone to do so, as this single-stranded polynucleotide is not maintained in a (competitive) double-stranded architecture.10 While mature human rRNA contains several guanine rich regions, much less is known about its secondary structure and its potential function. Recent evidence regarding the presence of rRNA G-quadruplexes in vitro has come to light,11–14 and some tentative evidence pointing toward RNA G-quadruplexes in the nucleoli in vivo has been recently reported as well.15 rRNA plays an important role in cellular proliferation and, together with G-quadruplex rDNA, G-quadruplex rRNA is likely intricately involved in regulatory cellular processes. Thus, the design of fluorescent probes specific to either of these two secondary structures is highly important and could readily find applications. Currently, only a handful of fluorescent probes with in vitro interactions with RNA G-quadruplexes have been reported. Pyridostatin (PDS) has been shown to interact with both DNA and RNA G-quadruplexes,16 while a carboxylated analogue does demonstrate a higher selectivity for RNA G-quadruplexes.17–19 Furthermore, a decorated acridine, napthalene, and cyanine demonstrated the visualization of RNA G-quadruplex structures. However, all these studies were directed at human telomeric RNA structures or cytosolic targets.20–22 Thiazole orange (TO), a well-known sensor for the detection of DNA, exhibits an obvious enhancement of fluorescence upon binding to DNA and shows a high affinity with G-quadruplex DNA compared with other forms of DNA.9,23 Thioflavin T, a prototypical viscosity and protein aggregation sensor, was demonstrated to localize to the nucleoli and may have G-quadruplexes as the intracellular target.10,24–26 Sun et al.27 and Shivalingam et al.28 reported a G-quadruplex DNA sensor with a preferential localization to the nucleus in general and nucleoli, respectively. To design probes that can dynamically interact with biological species without the need to wash before imaging, probes must be equipped with a means to prevent fluorescence in the absence of analyte binding. Among the different mechanisms, molecular rotation has been found to fulfill this function well,29 and in the context of polynucleotide sensing has also been proven to be highly effective, for example, in ethidium bromide and propidium iodide. Furthermore, the presence of a molecular rotor generally also allows for the determination of the viscosity in the immediate vicinity of the probe, as the ease of molecular rotation corresponds directly to the probes excited-state behavior. In this work, we designed a small molecule probe for the elucidation of G-quadruplex RNA in living cells, based on several design aspects of polynucleotide targeting chemical probes. Our probe, V-P1, was equipped with a molecular rotor to enable bright fluorescence in the presence of the analyte, further allowing for the determination of dynamic changes is the viscosity of the cytosol, and to a lesser degree the nuclear compartment, by both fluorescence intensity and lifetime imaging (Scheme 1). Scheme 1 | (a and b) Structure and synthesis of V-P1. Reagents and conditions: (1) H2SO4, 100 °C and (2) ice water, 70% HClO4. Download figure Download PowerPoint Experimental Methods Materials and instrumentation All reagents were purchased from Shanghai Titan Scientific Co., Ltd. (Shanghai, China) and were used without further purification. Monensin was purchased from Shanghai Titan Scientific Co., Ltd. (Shanghai, China). All different types of DNA, RNA, and RNA and DNA G-quadruplex structures were purchased from Sangon Biotech (Shanghai) Co., Ltd. (Shanghai, China). Deionized water was used throughout all experiments. Flash chromatography was implemented on silica gel (200–300 mesh). 1H and 13C NMR spectra were recorded using a Bruker DRX 400 spectrometer (Bruker Corporation, Billerica, Massachusetts, USA); mass spectrometry was recorded with an Agilent 1100 LC-MSD TOF mass spectrometer (Agilent technologies inc., Santa Clara, California, USA). The fluorescence spectra were performed on a F97XP FL spectrophotometer (Lengguang technologies inc., Shanghai, China) with the 1 cm standard quartz cell. Excitation and emission slit widths were 5 nm × 5 nm. The UV–vis spectra were obtained using a UV-240IPC spectrophotometer (Shimadzu Corporation, Kyoto, Kyoto, Japan). Viscosity analysis Viscous solutions were prepared by mixing methanol and glycerol in different volume proportions. The viscosity of each sample was measured with an NDJ-5S rotational viscosimeter. About 50 μL of a V-P1 stock solution (2.0 mM in MeOH) was added to the methanol–glycerol mixture (4.95 mL) to give a final concentration of 20 μM. The resulting solutions were shaken for 30 min and then kept still for 30 min at 25 °C, after which the fluorescence was recorded. The quantitative relationship between fluorescence intensity and the viscosity of the solution was fitted by the Förster–Hoffmann eq 1 as follows30: log ( I f ) = C + x log η (1)where If is the fluorescence intensity; η stands for the viscosity of solution; C is a concentration- and temperature-dependent constant; and x is a sensor- and temperature-dependent constant. In addition, the fluorescence at 650 nm upon excitation at 590 nm of a solution of V-P1 (20 μM) in methanol/glycerol (7/3, v/v) was recorded between −5 and 40 °C at 5 °C increments. Fluorescence quantum yields measurements The relative fluorescence quantum yields were identified with Rhodamine B (Φs = 0.97) in pure ethanol as a reference and calculated utilizing the following equation31: Φ x = = Φ s ( F x / F s ) ( A s / A x ) ( λ exs / λ exx ) ( n x / n s ) 2 where Φ represents quantum yield; F is the integrated area under the corrected emission spectrum; A stands for absorbance at the excitation wavelength; n is the refractive index of the solvent [because of the low concentrations of the solution (10−6–10−7 mol/L), the change of refraction coefficient in solution can be ignored]; λex is the excitation wavelength; and the subscripts x and s represent the unknown and the reference, respectively. Fluorescence lifetime detection Solvents of variable viscosity were prepared as mentioned above. A fluorescence lifetime measuring apparatus (Shimadzu) was used to acquire the fluorescence lifetimes of V-P1, with the excitation wavelength at 580 nm and emission at 650 nm. An excellent straight fitting was obtained, and the quantitative relationship between the fluorescence lifetime of V-P1 and the viscosity of the solution is described by Förster–Hoffmann eq 230: log τ = C + x log η (2)where τ is the fluorescence lifetime; η stands for the viscosity of solution; C is a concentration- and temperature-dependent constant; and x is a sensor- and temperature-dependent constant. The investigation of selectivity Considering the complexity of the intracellular environment, potential interference by various ions and bio-analytes toward V-P1 was also monitored, including anions (ONOO−, NO2−, and ClO−), cations (Co2+, Fe2+, Hg2+, Ag+, and Ni2+), reactive oxygen species (H2O2 and TBHP), and thiols (GSH, Hcy, and Cys). The stock solutions of these biologically relevant analytes (1.0 mM each) were prepared in trice distilled water. Stock solutions of V-P1 (2.0 mM) were prepared in methanol. For the measurements of fluorescence spectra, the excitation wavelength was set at 590 nm with excitation and emission slit widths being 5 nm × 5 nm. The experiments were performed using 20 μM of V-P1 in phosphate-buffered saline (PBS) solutions (pH = 7.4, 0.01 M, 1% methanol, 25 °C) with 50 μM of each analyte. The ds-DNA and ss-RNA were formulated into a 0.4 mg/mL stock solution in 10 mM Tris–HCl (100 mM KCl, pH 7.4) buffer solution. ss-DNA and all types of DNA and RNA G-quadruplexes were formulated into a 400 μM stock solution in 10 mM Tris–HCl (100 mM KCl, pH 7.4) buffer solution. About 40 μL of a V-P1 stock solution (1.0 mM in MeOH) was added to the Tris–HCl (100 mM KCl, pH 7.4) solvent (4 mL) to give a final concentration of 10 μM. Then the fluorescence and absorbance titration spectra of these DNA, RNA, and G-quadruplexes were measured. Competition experiment between viscosity and G-quadruplex RNA Viscous solutions were prepared by mixing Tris–HCl buffer (100 mM KCl, pH 7.4) and glycerol in different volume proportions. The viscosity of each sample was measured with an NDJ-5S rotational viscosimeter. About 30 μL of a V-P1stock solution (1.0 mM in MeOH) was treated with 60 μL R-570NT (1.0 mM in Tris–HCl buffer, 100 mM KCl, pH 7.4) first, then was added to the Tris–HCl/glycerol mixture (3 mL) to give a final concentration of 10 μM. The resulting solutions were shaken for 30 min and then kept still for 30 min at 25 °C, after which the fluorescence and fluorescent lifetime were recorded. Circular dichroism spectra determination Circular dichroism (CD) was utilized to explore the structural deformations occurring in V-P1 in binary solvent mixtures with different viscosity and with G-quadruplexes. To binary solvent mixtures with different viscosities (different proportions of methanol–glycerol at 25 °C) a fixed concentration of V-P1 (10.0 μM) was added, and the CD absorbance spectrum was recorded in the 400–650 nm range. Similarly, to solutions of RNA G-quadruplexes [R-6960NT, 0–2.4 μM in Tris–HCl buffer solution (20 Mm, 100 mM KCl, pH 7.4)] a fixed concentration of V-P1 (10.0 μM) was added, and the CD absorbance spectrum was recorded in the 400–650 nm range. Density functional theory calculations A fully unrestrained ground-state optimization of V-P1 was performed using the Gaussian 16 software package,32 at the B3LYP/N07D level of theory using the integral equation formalism variant of the polarizable continuum model (IEFPCM) solvation model of methanol.33–35 A dihedral angle scanning was subsequently performed for the ground state and the first excited state, where the dihedral angle was set to the value indicated in the figures, but all other coordinates were otherwise unrestrained. For energy calculations, the ωB97XD functional with the same basis set was employed using the B3LYP/N07D optimized structures with a state-specific solvation correction.36–38 Orbital visualizations were performed using GaussView 6.0.39 Docking studies Docking studies were performed with the density functional theory (DFT)-optimized ground-state structure of V-P1, using AutoDock 4.2,40 using a grid with grid spacing of 0.375 Å. The oligonucleotide structures were built from the corresponding protein data bank (PDB) entries (see Docking studies with G-quadruplex RNA) and optimized using the Minimize Structure feature in Chimera 1.13.1.41 Oligonucleotide Gasteiger partial charges were assigned in AutoDockTools, with additional corrections for the K+ ion charges. Partial charges for the ligand were obtained using CM5 charge calculations from a B3LYP/6–31 + G(d) level calculation in methanol.42 The results from the docking studies were visualized using Chimera 1.13.1.41,43 Cytotoxicity test To investigate the cytotoxicity of V-P1, six cell lines were used: leukemic cells (HL-60), lung carcinoma cells (A549), liver hepatocarcinoma cells (SMMC-7721), breast adenocarcinoma cells (MCF-7), colonic cancer cells (SW480), and a normal cell line (BEAS-2B). IC50 was evaluated by utilizing the commercial 3-(4,5-dimethylthiazol-2-yl)-5(3-carboxymethoxyphenyl)-2-(4-sulfopheny)-2H-tetrazolium (MTS, a new derivative of MTT) viability indicator. All the details about the experiments of cell imaging are presented in the Supporting Information. Cell culture HeLa cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen Corporation, Carlsbad, California, USA) supplemented with 10% fetal bovine serum (FBS) in an atmosphere of 5% CO2 and 95% air at 37 °C. The cells were seeded onto 24-well flat-bottomed plates (1.0 × 104 each well). When the density of the cell reached 70–80% of confluence, the subculturing was conducted. The culture medium was replaced every 2–3 days after PBS washing. All the details about the experiments of cell imaging are presented in the Supporting Information. Results and Discussion Design and synthesis of V-P1 Our strategy to develop the multifunctional probe V-P1 was to combine a fluorescent cationic chromenylium ring that targetted the polynucleotides, with a N,N-dimethylaniline ring through a C–C bond to form a rotor that could produce fluorescence quenching in low-viscosity environments and demonstrate a fluorescence enhancement in highly viscous media. V-P1 was synthesized by a condensation reaction between 4-(diethylamino)salicylic aldehyde and 1-[4-(dimethylamino)phenyl]ethanone and was isolated in a moderate yield of 62% (Schemes 1a and 1b and Supporting Information Figures S1). V-P1 was fully characterized by NMR and high-resolution mass spectrometry (see Supporting Information Figures S1–S3). Spectral properties The absorbance and emission maxima of V-P1 in methanol were 603 and 648 nm, respectively ( Supporting Information Figure S4), indicating a Stokes shift that was sufficiently large to enable imaging with a high degree of signal to noise. The fluorescent response of V-P1 as a function of the solvent viscosity was studied in solutions of methanol with different proportions of glycerol. The fluorescence intensity of V-P1 clearly increased with increasing proportions of glycerol (Figure 1a). The viscosity of the samples tested increased from 1.3 cP (pure methanol) to 433.0 cP (90% glycerol). (The viscosities of the mixtures used are listed in Supporting information Table S1.) V-P1 exhibited a weak emission when it was excited at 590 nm in a low-viscosity solution (100% methanol, 1.3 cP), with a low quantum yield of 0.09 and a high molar extinction coefficient (ɛ = 0.7 × 105 L mol−1 cm−1). In 90% glycerol, a high quantum yield of 0.88 and molar extinction coefficient of 0.83 × 104 L mol−1 cm−1 were observed, while the absorbance of these V-P1 solutions remained largely unchanged ( Supporting Information Figure S5). Thus, a 10-fold fluorescence intensity increase was achieved for V-P1 in this viscosity range (Figure 1a). As shown in the inset, a good linear proportionality between the fluorescence intensity I650 (log I650) and viscosity (log η), with a correlation coefficient of 0.9903 was observed, demonstrating V-P1 could potentially be applied for quantitative viscosity measurements. Figure 1 | (a) The fluorescence spectra of V-P1 (20 μM) in mixed solvents with different proportions of methanol–glycerol at 25 °C. Inset: Linear relationship of log I650 and log η (R2 = 0.9903). (b) The fluorescence spectra of V-P1 (20 μM) at different temperatures in ethanol–glycerol (30/70, v/v), excited at 590 nm. Inset: Linear relationship of I650 and temperature (R2 = 0.9985). (c) Fluorescence lifetime of V-P1 (20 μM) in mixed solvents with different proportions of methanol–glycerol (λem = 650 nm). Inset: Linear relationship of log τ and log η (R2 = 0.9919);. (d) Fluorescence intensity (λem = 650 nm) of V-P1 (20 μM) in the presence of various analytes (50 μM n: MeOH/glycerol mixture with viscosity of 433 cP). Bars denote the average of n = 3 independent measurements, while error bars denote the standard deviation. Download figure Download PowerPoint As the viscosity is inversely related to the temperature, we tested the influence of the temperature toward the fluorescence emission of V-P1 (Figure 1b). In the 45 to −5 °C temperature range, the emission intensity at 650 nm increased about fourfold. Importantly, an excellent linear relationship existed between I650 and the temperature (R2 = 0.9985). Meanwhile, when the viscosity (η) in the mixed solvent increased from pure methanol (1.3 cP) to 90% glycerol (433 cP), the fluorescence lifetime (τ) gradually extended from 371 to 2200 ps with a good linear relationship between log τ and log η (R2 = 0.9919) and a steep slope (0.429), indicating that sensor V-P1 can also quantitatively detect the viscosity by measuring the fluorescence lifetime changes (Figure 1c). As viscosity sensors can also be sensitive to the solvent polarity, fluctuations in the polarity can affect the accuracy of microenvironment viscosity determinations. Thus, the influence of solvent polarity on the fluorescence of V-P1 was determined using virtually isoviscous 1,4-dioxane/water mixtures with large polarity differences. Upon polarity changes, V-P1 demonstrated only slight fluorescence fluctuations, meaning the polarity influence was negligible in the test system ( Supporting Information Figure S6 and Table S2). Thus V-P1 is endowed with the critical characteristic of detecting microenvironmental viscosity without disturbance from the polarity. Furthermore, only relatively small fluorescence changes could be observed in different organic solvents ( Supporting Information Figure S7), and the fluorescence intensity remained stable in the pH 3–8.5 range ( Supporting Information Figure S8), demonstrating the absence of pH effects under physiological pH conditions. To evaluate the possibility of V-P1 microviscosity sensing in complex cellular microenvironments, several potential interferents were tested, such as metal ions, anions, thiols, reactive oxygen species, and bioactive small molecules. The selectivity of V-P1 is exhibited in Figure 1d. All potential interference factors that were tested produced a negligible fluorescent effect, compared with the large fluorescent enhancement of the sensor at a viscosity of 433 cP. Fluorescence enhancements in the presence of oligonucleotides As shown in Figures 2a and 2b, in Tris–HCl buffer (20 mM, pH 7.4, containing 100 mM K+) the probe (10 μM) exhibited a specific absorption peak with a maximum at 570 nm and a relatively weak emission peak located at 637 nm. Upon addition of G4-RNA (R-6960NT, up to 10 μM) to the buffered V-P1 solution, the absorption peak gradually shifted to 605 nm with an obvious isosbestic point at 580 nm (Figure 1b), suggesting a strong interaction between V-P1 and R-6960NT. In accordance with the change in the absorption spectra, the fluorescence of V-P1 also displayed a remarkable change. The emission peak gradually shifted to 655 nm with an obvious enhancement of the intensity (Figure 1a). In addition, the reverse titration between V-P1 and R-6960NT also produced similar results, which further indicated an interaction of V-P1 with R-6960NT (Figure 2c). Figure 2 | (a and b) Fluorescence and absorption spectra of V-P1 (10 μM) with different concentrations of G4-RNA (R-6960NT) in Tris–HCl (20 mM, pH 7.4) buffer solutions containing K+ (100 mM). (c) Fluorescence spectra of G4-RNA (R-6960NT, 10 μM) with different concentrations of V-P1 in Tris–HCl (20 mM, pH 7.4) buffer solutions containing K+ (100 mM). Inset: Fluorescence intensities at 650 nm with different concentrations of V-P1. (d) Fluorescence intensity changes at 650 nm of V-P1 (10 μM) toward G4-RNA (R-6960NT, 10μM) after the addition of ss-DNA (mpu22, 10 μM), ss-RNA (10 μM), ds-DNA (10 μM), and G4-DNA (c-kit1, 10 μM) as potential competing analytes in Tris–HCl (20 Mm, pH 7.4) buffered solutions containing K+ (100 mM). Bars denote the average of n = 3 independent measurements, while error bars denote the standard deviation. Download figure Download PowerPoint As a potential fluorescent probe for oligonucleotides, the fluorescent responses of V-P1 toward different types of DNA and RNA structures were tested. As shown in Figure 2d, the fluorescence intensity of V-P1 at 650 nm in the presence of R-6960NT was approximately six times higher than other oligonucleotides, such as ds-DNA, ss-RNA, ss-DNA (mpu22), and G4-DNA (c-kit1). Clearly, these potential competing species did not interfere with the measured fluorescence sensitivity of V-P1 toward R-6960NT. These results indicate that V-P1 has the potential to bind with specific G4-RNA even in a complex biological environment. To further explore the selectivity of V-P1, different sequences of DNA, RNA, DNA and RNA G-quadruplex were used for the next experiments (Figure 3). As shown in Figures 3c and 3d, while the fluorescence enhancement in the presence of ss-DNA or ds-DNA and ss-RNA was negligible, obvious increases in fluorescence intensities were observed for G-quadruplex structures, especially for RNA G-quadruplexes (Figures 3b and 3d). Representative examples of binding isotherms to DNA G-quadruplex (c-kit2), RNA G-quadruplex (R-570NT), and ss-DNA (mpu22) are found in Figures 3a–3c, while the binding isotherms for all tested sequences listed in Figure 3d are found in Supporting Information Figures S9–S22, and the sequence identity of the tested oligonucleotides is listed in Supporting Information Table S3. Figure 3 | (a–c) Fluorescence spectra of V-P1 (10 μM) upon the addition of increasing concentrations of c-kit2, R-570NT, and mpu22 in Tris–HCl (20 mM, pH 7.4) buffer solutions containing K+ (100 mM). Inset: the Benesie–Hildebrand fitting of the titration curves and the fluorescence fitting curve of V-P1 with the oligonucleotides (for panels a and b) and fluorescence intensity at 650 nm for panel c. Bars denote the average of n = 3 independent measurements, while error bars denote the standard deviation. (d) Fluorescent response of V-P1 (10 μM) at the respective fluorescent maxima in the presence of various types of oligonucleotides (10 μM) in Tris–HCl (20 mM, pH 7.4) buffer solution with K+ (100 mM). Bars denote the average of n = 5 independent measurements, while error bars denote the standard deviation. (e) The extrapolated Kd values of V-P1 with various DNA/RNA G-quadruplexes. Download figure Download PowerPoint The obvious differences in these responses most likely arose from architectural differences in the quadruplex structures. A stronger fluorescence enhancement with RNA G-quadruplex may result from the more base-constrained RNA G-quadruplex as compared with DNA G-quadruplex structures. Fluorescence titrations of V-P1 with five RNA G-quadruplexes and seven DNA G-quadruplexes were tested, and the calculated binding constants are summarized in Figure 3e, with Kd values ranging from 1.82 to 7.66 μM and from 1.15 to 47.65 μM for DNA G-quadruplexes and RNA G-quadruplexes, respectively. It is well-known that some DNA/RNA binding ligands are optically inactive and achiral, and while interacting with DNA/RNA, an enhanced CD signal can be observed.44 Therefore, CD was utilized to explore the structural deformations occurring in V-P1 with G-quadruplexes. On the one hand, as expected, utilizing binary solvent mixtures with different viscosity did not lead to any appearance of CD signals ( Supporting information Figure S23a). This is because upon excitation of V-P1 in solution with a torsion angle of φ = 0°, the formation of a twisted shape can occur in both the left- and right-handed direction, as the 4-dimethylamino-phenyl group is fully symmetrical. However, by binding the G4 RNA, a degree of unidirectional molecular motion and axial chirality/atropisomerism in V-P1 was gained, which is also reflected in the CD spectrum ( Supporting information Figure S23b). From the results above, it was clear that there was a net CD absorbance effect upon the addition of G4 RNA, consistent with an expected groove binding mode between V-P1 and G4 RNA (see Docking studies with G-quadruplex RNA). As V-P1 exhibited fluorescence enhancements both in the presence of G-quadruplex oligonucleotides and increased viscosity, we next tested the potential influence of viscosity in the presence of oligonucleotides. In the presence of 10 μM R-570NT, a concentration in which fluorescence saturation for this oligonucleotide was reached, both the fluorescence intensity and lifetime were nonetheless clearly increased when the viscosity of the medium was increased (Figures 4a and 4b). When the viscosity was 433 cP, the lifetime reached 3.2 ns while that one in control was only 2.2 ns. Thus, clearly the binding with G4 RNA did not quench the response of V-P1 toward viscosity. Figure 4 | (a) Fluorescence spectra of V-P1 (10 μM) in the presence of 10 μM R-570NT (10 μM) in mixed solvents with different proportions of Tris–HCl (20 mM, pH 7.4) buffer solution with K+ (100 mM) and glycerol at 25 °C. I