Chiroptical property enhancement of chiral Eu(III) complex upon association with DNA-CTMA

DNA-based materials have attracted much attention due to their unique photo-functional properties and potential applications in various fields such as luminescent and biological systems, nanodevices, etc. In this study, the photophysical properties of a chiral Eu(III) complex, namely (Eu(D-facam)3), within DNA films were extensively investigated. The enhancement of photoluminescence (more than 25-folds increase of luminescence quantum yield) and degree of circularly polarization in luminescence (glum = − 0.6) was observed upon interaction with DNA. Various photophysical analyses suggested that the emission enhancement was mainly due to an increase of the sensitization efficiency (high ηsens) from the ligands to Eu(III) and suppression of the vibrational deactivation upon immobilization onto the DNA molecule. From CD and VCD measurements, it was suggested that the coordination structure of Eu(D-facam)3 was affected by the interaction with DNA, suggesting that the structural change of Eu(D-facam)3 contributed to the improvement of its luminescent properties.


Scientific Reports
| (2020) 10:18917 | https://doi.org/10.1038/s41598-020-75808-w www.nature.com/scientificreports/ the luminescence deriving from Eu(III) complexes can be attributed to the forbidden f-f transitions and their low |μ| results in a high g lum . Therefore, the simultaneous achievement of a strong emission intensity and high g lum seems challenging. On the other hand, the emission enhancement of luminophores due to their association with DNA was widely reported [36][37][38][39] . In our previous study, emission enhancement and induced CPL were achieved by associating an achiral Eu(III) complex with DNA-CTMA 40 . Therefore, a more distinctive enhancement of the optical properties can be expected by adding chiral sites to the Eu(III) complex, which interacts with DNA.
In this study, we investigated the luminescence properties of a chiral Eu(III) complex within a DNA film. To this aim, we selected Eu(D-facam) 3 (europium tris[3-(trifluoromethylhydroxymethylene)-(+)-camphorate]) as chiral Eu(III) complex, which is known for its use as NMR-shift reagent and biological sensing probe 41,42 . Interestingly, a higher luminescence intensity and |g lum | of CPL were achieved from Eu(D-facam) 3 compared with the conventional polymer upon interaction with DNA.

Results and discussion
Interaction between the chiral Eu(III) complex and DNA-CTMA. First, we introduced Eu(Dfacam) 3 into DNA backbone to observe their optical properties. Since DNA is soluble only in water whereas Eu(D-facam) 3 is insoluble in water, DNA is modified with CTMA, which is one of the most typical surfactants utilized for biomolecules. By utilizing DNA-CTMA, we successfully fabricated DNA-CTMA/Eu(D-facam) 3 films. The absorption and CD spectra of the DNA-CTMA/Eu(D-facam) 3 films at various Eu(D-facam) 3 :DNA-CTMA molar ratios are shown in Fig. 1. For comparison, the optical analysis of a PMMA/Eu(D-facam) 3 film was also conducted; PMMA has no chirality in its structure, while DNA has a well-known axisymmetric helical structure 43 . In the PMMA/Eu(D-facam) 3 film, an absorption band assignable to the π-π* transition of the β-diketonate 44 in the D-facam was observed around 305 nm. On the other hand, the absorption peaks of the DNA-CTMA/Eu(D-facam) 3 films were red-shifted by about 10 nm compared to that of the PMMA/Eu(Dfacam) 3 film, suggesting an interaction between DNA and Eu(D-facam) 3 . It is possible that Eu(D-facam) 3 electrostatically approach the anionic phosphate groups in the DNA backbone, and such absorption change suggests the intercalation or semi-intercalation between base pairs subsequent by electrostatic interaction 45,46 . The CD spectrum indicates ellipticity at each wavelength 29 . Ellipticity is proportional to the difference of absorbance against the left-handed and right-handed circularly polarized light, thus a change in ellipticity indicates the structural chirality of the molecules corresponding to its absorption. In the PMMA/Eu(D-facam) 3 film, a positive Cotton effect corresponding to the absorption band of D-facam was observed. This was attributed to the fact that the D-facam ligand of Eu(D-facam) 3 possesses a chiral structure. For the DNA-CTMA/Eu(D-facam) 3 films, typical exciton-splitting CD signals with positive (330 nm) and negative (300 nm) Cotton effects centered at the absorption peak of the ligand were observed. This indicated that the exciton coupling of the D-facam ligands in Eu(D-facam) 3 occurred upon interaction with DNA. In addition, the signal intensity of the exciton coupling increased relative to the increase of the DNA ratio, strongly indicating that the structural chirality of Eu(D-facam) 3 was enhanced.
Vibrational circular dichroism (VCD) spectroscopy experiments were carried out to determine more detailed structural change of Eu(D-facam) 3 in the presence of DNA. Figure 2 shows the infrared absorption (IR) and VCD spectra of the Eu(D-facam) 3 3 film, emission peaks were observed at 579, 585-600 and 613 nm; they are assignable to the 5 D 0 → 7 F 0 , 5 D 0 → 7 F 1 and 5 D 0 → 7 F 2 transition of Eu(III) ions, respectively 48 . Interestingly, for the DNA-CTMA/Eu(D-facam) 3 film, the emission peak assignable to the 5 D 0 → 7 F 1 transition split into two peaks (586 and 595 nm). This change of the emission peak obviously indicated that the interaction with DNA affected the crystal field around the Eu(III) ion (Fig. 3b inset). In comparison to the PMMA/Eu(D-facam) 3 film, the DNA-CTMA/Eu(D-facam) 3 films showed a stronger emission. Their emission intensity was enhanced with increasing ratios of DNA-CTMA, suggesting that a non-radiative deactivation caused by a molecular vibration was suppressed due to the immobilization onto the DNA backbone 49,50 .
In regard to luminescence intensity of each emission band, radiative rate of the emission band is mainly determined by electric dipole transition which is sensitive to the ligand field (i.e. electric field around Eu(III) ion). Because the emission peaks assignable to the 5 D 0 → 7 F 1 transition derives from mainly magnetic dipole (MD) transition, its radiative rate is not considerably affected by the ligand field. Therefore, the symmetry of Eu(D-facam) 3 can be discussed based on the ratio of the emission intensities obtained from the MD moment (I MD ) and ED moment (I ED ) 51 . The emission ratio (I rel = I ED /I MD ) of the PMMA/Eu(D-facam) 3 and DNA-CTMA/ Eu(D-facam) 3 films are shown in Table 1. I MD and I ED were calculated by integrating the emission intensities at 582-600 nm and 605-630 nm, respectively. Since the values of I rel for the DNA-CTMA/Eu(D-facam) 3 films (5.14-5.60) were lower than that of the PMMA/Eu(D-facam) 3 film (7.64), it was estimated that the coordination structure around the Eu(III) ion in Eu(D-facam) 3 changed to a higher symmetric structure in the presence of DNA. The changes in the structure of Eu(D-facam) 3 were also confirmed by CD and VCD measurements. These results clearly supported that Eu(D-facam) 3 and DNA interacted with each other and led to a stronger emission, as shown in Fig. 3. 3 films. Generally, Eu(III) ions with a highly symmetrical structure hardly show any strong emission since the high symmetry of the Eu(III) ion results in a low radiative rate 52 . Therefore, we determined the factors that contributed to the emission enhancement of Eu(D-facam) 3 in DNA-CTMA films. The total quantum yield (Φ tot ) and luminescence lifetime (τ) of the PMMA/Eu(D-facam) 3 and DNA-CTMA/Eu(D-facam) 3 films were obtained ( Table 1). The luminescent quantum yield was 0.5 and 11.9-13.7% for the PMMA/Eu(D-facam) 3 and DNA-CTMA/Eu(Dfacam) 3 films, respectively; the luminescent quantum yield of Eu(D-facam) 3 was significantly increased in the case of DNA-CTMA compared to PMMA. The radiative rate constants of Eu(III) ions (k r ) was also calculated using the relation:

Detailed analyses of photophysical parameters of the DNA-CTMA/Eu(D-facam)
where A MD,0 is the spontaneous emission coefficient of the 5 D 0 → 7 F 1 transition (= 14.65 s −1 ), n is the refractive index of the medium and I tot /I MD is the ratio of the integrated radiation corresponding to the 5 D 0 → 7 F j transition (j = 0-6) to the peak area corresponding to the 5 D 0 → 7 F 1 transition 51,53 . Here, the value of n was determined to be 1.52 and 1.49 for the DNA-CTMA 54 and PMMA solid film 55 , respectively. The non-radiative rate constant (k nr ) can be calculated from the relations: while the intrinsic quantum yield (Φ Ln ) and efficiency of sensitization of the lanthanide luminescence by the ligand (η sens ) can be calculated from the relations 56 : Table 1. Ratio of the emission intensity of the MD and ED moment (I rel ), luminescence lifetime (τ), radiative rate (k r ), non-radiative rate (k nr ), intrinsic quantum yield of Eu(III) ion (Φ Ln ), total quantum yield (Φ tot ), and efficiency of sensitization (η sens ) of the DNA-CTMA/Eu(D-facam) 3 and PMMA/Eu(D-facam) 3 films.  3 for the DNA-CTMA films (316-340 s −1 ) was lower than that of the PMMA film (419 s −1 ). This implies that the probability of light emission from the excited state decreased in the presence of DNA, which is consistent with the structural change around the Eu(III) ion towards a higher symmetry structure, as discussed above. On the other hand, k nr decreased and η sens increased for the DNA-CTMA films compared to the PMMA film. The k nr value of PMMA was approximately 8400 s −1 ; it decreased to 1300 s −1 upon mixing with DNA. It is known that the molecular vibration can be suppressed when the molecules are immobilized on a DNA structure 49,50 . Therefore, the decrease in k nr clearly indicated a suppressed vibrational deactivation of the excited states of the Eu(III) ion due to the immobilization on a DNA molecule. The decreased k nr contributed to improving the intrinsic quantum yield (Φ Ln ); the calculated Φ Ln increased from 4.77% (in PMMA) to 20.5% (in DNA-CTMA). In addition, the η sens value of Eu(D-facam) 3 was significantly improved upon interaction with DNA; it was almost 6 times higher compared to that of PMMA. This improvement of η sens is believed to contribute to the emission enhancement of Eu(D-facam) 3 .
It is known that the η sens value of lanthanide complexes significantly depends on the relationship between the T 1 level of the ligands and the accepting 4f level of the central metal ion 56 . An adequate energy gap between the T 1 and 4f levels facilitates the energy transfer from the ligands to the metal ion. However, a close match between the T 1 and 4f levels should be avoided as it induces back energy transfer from the metal ion to the ligands. To investigate the change of the T 1 level of the D-facam ligand in the different polymers, the phosphorescence of Gd(D-facam) 3 of the PMMA and DNA-CTMA film was measured at 77 K (Fig. 4) 57 . It is known that the first excitation energy of Gd(III) is much higher than the triplet energy of general complex ligands 58 . Thus, the energy transfer from ligands to the Gd(III) ion hardly occur. Only the phosphorescence from the ligands should be observed. The T 1 level can be calculated from the onset wavelength of the phosphorescence spectrum. Broad phosphorescence bands of D-facam were observed around 460-650 nm for both films. The onset wavelength of the phosphorescence spectrum of each film almost matched (462 nm), indicating that the T 1 level of D-facam was unperturbed (calculated as ca. 21,600 cm −1 ). Thus, the improvement of η sens was caused by factors other than the change in the T 1 level of the ligands. Therefore, the changes of the distance between Eu(III) and ligands as well as the angle between them, as evidenced by the CD and VCD spectra and I rel values, contributed to the significant improvement of η sens . In addition, the decrease of k nr discussed above obviously indicates that Eu(Dfacam) 3 was tightly immobilized onto the DNA structure. The immobilization of these molecules might suppress the vibrational deactivation of the T 1 states, and allow to improve η sens , which represents the energy transfer efficiency from the ligands to the central metal ion.
Circularly polarized luminescence induced by DNA-CTMA. As discussed above, it was demonstrated that the structural chirality of Eu(D-facam) 3 significantly changed upon its interaction with DNA-CTMA. Therefore, it was assumed that CPL, which reflects the chiral luminescence, might also be greatly enhanced 59 . Therefore, we measured the CPL spectra of the PMMA/Eu(D-facam) 3 and DNA-CTMA/Eu(D-facam) 3 films and their g lum values are shown in Fig. 5. An emission dissymmetry factor [g lum = 2(I L − I R )/(I L + I R )] was utilized to quantitatively evaluate the magnitude of CPL, where I L and I R represent the emission intensity of left-handed and right-handed circular polarized luminescence, respectively 35 . For the PMMA/Eu(D-facam) 3 film, the CPL intensity was very weak, and the g lum was calculated to be − 0.02 at 5 D 0 → 7 F 1 (598 nm, MD transition). On the other hand, in the case of the DNA-CTMA/Eu(D-facam) 3 films, clear CPL signals were observed at 5 D 0 → 7 F 1 (598 nm, MD transition) and 5 D 0 → 7 F 2 (615 nm, ED transition). The g lum at 5 D 0 → 7 F 1 (598 nm) was determined to be − 0.62, which resulted approximately 30 times enhanced compared to the PMMA/Eu(D-facam) 3 film. Such chirality enhancement of the luminescence was supposed to be due to the change of the ligand field of Eu(Dfacam) 3
Preparation of the DNA-CTMA complex. DNA-CTMA was prepared by precipitating DNA with a cationic surfactant complex of CTMA in water through an ion exchange reaction that replaced the sodium cations of the DNA. The DNA complex with CTMA (DNA-CTMA) was prepared by the addition of a 10 mM aqueous solution of DNA (based on the concentration of the phosphate groups) to a 10 mM CTMA solution. The precipitate was filtered and thoroughly washed with ultrapure water and then dried in vacuo. The resulting DNA-CTMA was more water insoluble and more mechanically stable than the DNA itself due to the long alkyl chain of the CTMA. Through the formation of the CTMA complex, DNA-CTMA was soluble in solvents more compatible with device fabrication, such as chloroform, ethanol, methanol, butanol, or a chloroform/alcohol blend. Measurements of the optical properties. The absorbance and CD spectra of the DNA-CTMA/Eu(Dfacam) 3 and PMMA/Eu(D-facam) 3 films were acquired using a photonic multichannel analyzer (J-1100, JASCO Corporation, Japan). The emission spectra were acquired using a spectrofluorometer (FP-6600, JASCO Corporation, Japan). The emission quantum yields were calculated from the data obtained from an absolute PL quantum yield spectrometer (Quantaurus-QY C11347-01, Hamamatsu photonics K. K., Japan). The emission lifetimes were determined using a time-resolved fluorescence spectrometer (Quantaurus-Tau C11367-21, Hamamatsu photonics K. K., Japan). CPL measurements were conducted using a previously reported system 40,60 , which consisted of the following components: 375 nm LED (M365L2, Thorlabs Japan Inc., Japan), LED driver (DC2100, Thorlabs Japan Inc., Japan), photoelastic modulator (PEM-90, Hinds instruments, Inc. United States), photo- Scientific Reports | (2020) 10:18917 | https://doi.org/10.1038/s41598-020-75808-w www.nature.com/scientificreports/ multiplier tube (H7732-10, Hamamatsu photonics K. K., Japan), linearly polarized cubic prism (200,000:1), photomultiplier tube (H7732-10, Hamamatsu photonics K. K., Japan), and dual phase DSP lock-in amplifier (7265, Signal Recovery Ltd., United Kingdom). The appropriate detection wavelength of the monochromator and PEM was controlled by a PC.