Structural Characterization of Zinc(II)/Cobalt(II) Complexes of Chiral N-(Anthracen-9-yl)methyl-N,N-bis(2-picolyl)amine and Evaluation of DNA Photocleavage Activity

Abstract

We designed and synthesized a chiral ligand N-(anthracen-9-ylmethyl)-1-(pyridin-2-yl)-N-(pyridin-2-ylmethyl)ethanamine (APPE) DNA photocleavage agent to investigate the effects of chirality of bis(2-picolyl)amine on the DNA photocleavage activity of metal complexes. The structures of Zn^II and Co^II complexes in APPE were analyzed via X-ray crystallography and fluorometric titration. APPE formed metal complexes with a 1 : 1 stoichiometry in both the crystalline and solution states. Fluorometric titration was used to show that the Zn^II and Co^II association constants of these complexes (log K_as) were 4.95 and 5.39, respectively. The synthesized complexes were found to cleave pUC19 plasmid DNA when irradiated at 370 nm. The DNA photocleavage activity of the Zn^II complex was higher than that of the Co^II complex. The absolute configuration of the methyl-attached carbon did not affect DNA cleavage activity and, unfortunately, an achiral APPE derivative without the methyl group (ABPM) was found to perform DNA photocleavage more effectively than APPE. One reason for this may be that the methyl group suppressed the structural flexibility of the photosensitizer. These results will be useful for the design of new photoreactive reagents.

Introduction

DNA photocleavage metal complexes have unique structures that have formed the basis of many studies.^1,2) Although complexes that inhibit the growth of cancer cells by light irradiation have been reported,^3,4) continuing research seeks to improve their effectiveness. The application of DNA photocleavage metal complexes to biological tools requires the development of plural ligands responding to various wavelengths. We focus on the structural extensibility of bis(2-picolyl)amine (BPA). BPA is a tridentate 3N ligand capable of binding various metals, including non-redox active metals⁵⁾ and transition metals.⁶⁾ Furthermore, secondary amine substitutions in BPA can be used to increase the types of metals to which the complex can coordinate and/or to introduce bioactive structures.^7,8) Several studies have reported metal-complexed BPA derivatives for DNA cleavage activity,⁹⁾ including photocleavage.¹⁰⁾ Since the metal atom itself exhibits poor directivity toward DNA, target-directivity in vivo is ligand-dependent. The design of artificial ligands with desired sequence specificity is necessary to achieve practical application of DNA-cleaving metal-complexed BPA derivatives. The introduction of an anthracene moiety is among the promising strategies proposed for the design of sequence-specific compounds.¹¹⁾ Anthracene moieties also function as photosensitizers¹²⁾ and have potential for controlling biological activity. Furthermore, the control of ligand chirality may be advantageous for DNA cleavage.¹³⁾ Although the effectiveness of N-substituents in BPA derivatives has been frequently examined,^9,14) the effectiveness of introducing chirality has received less attention. Accordingly, we synthesized N-(9-anthracenyl)methyl BPA derivatives (Fig. 1) and evaluated the DNA photocleavage activity of metal complexes using the synthesized ligands. The main purposes of this study were to investigate the steric hindrance of ligand molecules on DNA photocleavage activity. Zn^II and Co^II were used as the central metals of these complexes because they have shown promise in previous studies.¹⁰⁾ To evaluate steric hindrance, we used a methyl group to evaluate the small substituent effect.

Fig. 1. Chemical Structures of Synthesized Ligands

Results and Discussion

Synthesis

We designed a chiral metal-ligand termed N-(anthracen-9-ylmethyl)-1-(pyridin-2-yl)-N-(pyridin-2-ylmethyl)ethanamine (APPE). The synthetic route for this ligand is shown in Chart 1. 1-Pyridin-2-ylethylamine was used as the starting material. The absolute configuration of APPE corresponds to that of 1-pyridin-2-ylethylamine, a racemic mixture of APPE, rac-APPE (2), and the pure S configuration of APPE, S-APPE (3), were synthesized. The overall yields of (2) and (3) from 1-pyridin-2-ylethylamine were 28% and 6.8%, respectively. For comparative purposes, an achiral metal-ligand, 1-(anthracen-9-yl)-N,N-bis(pyridin-2-ylmethyl)methanamine (ABPM, 1), was also synthesized according to previously described methods.¹⁵⁾

Chart 1. Synthetic Scheme of APPE

X-Ray Crystal Structure of Zn^II/Co^II Complexes

X-Ray crystallographic analysis was undertaken to evaluate the structure of Zn^II/Co^II complexes of the synthesized ligands. Crystals of these complexes were prepared using Zn(NO₃)₂·6H₂O and Co(NO₃)₂·6H₂O. Briefly, rac-APPE and either Zn(NO₃)₂·6H₂O or Co(NO₃)₂·6H₂O were mixed with methanol and stirred at 40 °C. The mixture was allowed to stand at room temperature, after which crystals suitable for X-ray crystallography were obtained. An S-form absolute configuration of APPE was obtained in the Zn–APPE crystals (CCDC-2237481). Interestingly, the reactants of Co and rac-APPE yielded, respectively, Co–S-APPE (CCDC-2237486) and Co–R-APPE (CCDC-2237782) crystals from the same reaction vessel; spontaneous resolution was observed. ORTEP drawings of [Zn(S-APPE)(NO₃)₂]·CH₃OH, [Co(S-APPE)(NO₃)₂]·CH₃OH, and [Zn(ABPM)(NO₃)₂] are shown in Fig. 2A. The crystal structure of [Zn(ABPM)(NO₃)₂] has previously been reported by Nugent (CCDC-1019945).¹⁶⁾ In the crystalline state, the binding stoichiometry between Zn^II/Co^II and APPE was found to be 1 : 1, with tridentate ligands coordinated to the central metal with a mer (meridional) coordination. In addition, two nitrate ions were coordinated horizontally and vertically against the plane formed by the tridentate nitrogen atoms. Therefore, the metal coordination structures had a distorted square pyramidal form, irrespective of whether Zn^II or Co^II was used as the central metal. In terms of metal–nitrogen distances, that between the tertiary amine (3°N) and metal was longer than that between the pyridine nitrogen and the metal (Fig. 2B). A similar tendency has been observed in Zn^II complexes of BPA derivatives.^17–19) Due to the bond distance imbalance, the anthracenylmethyl moiety and vertically coordinated nitrate were stabilized in the s-trans conformation. A set of torsion angles defined as M–N–C–C, which are metal, 3°N, methylene C, and anthracene C, respectively, are shown as solid lines in Fig. 2C. The torsion angles on [Zn(ABPM)(NO₃)₂], [Zn(S-APPE)(NO₃)₂]·CH₃OH, and [Co(S-APPE)(NO₃)₂]·CH₃OH were 175.5, 43.3, and 39.8°, respectively. Another set of torsion angles defined as C*–N–C–C, which are chiral C, 3°N, methylene C, and anthracene C, are shown as dotted lines in Fig. 2C. The torsion angles on [Zn(S-APPE)(NO₃)₂]·CH₃OH and [Co(S-APPE)(NO₃)₂]·CH₃OH were 160.9 and 156.2°, respectively. The corresponding torsion angle of [Zn(ABPM)(NO₃)₂] was 60.1°. These angles indicate that Zn^II and the anthracene moiety adopt an antiperiplanar conformation in Zn–ABPM complexes, whereas the methyl group and anthracene adopt an antiperiplanar conformation in Zn–S-APPE and Co–S-APPE complexes. Thus, the methyl group introducing BPA is capable of controlling the orientation of the N-substituted moiety. These structural features, characteristic of Co–S-APPE, were also observed in Co–R-APPE.

Fig. 2. Crystal Structures of [Zn(ABPM)(NO₃)₂], [Zn(S-APPE)(NO₃)₂], and [Co(S-APPE)(NO₃)₂]

(A) ORTEP drawing showing [Zn(ABPM)(NO₃)₂] (CCDC 1019945),¹⁶⁾ [Zn(S-APPE)(NO₃)₂]·CH₃OH (CCDC-2237481), and [Co(S-APPE)(NO₃)₂]·CH₃OH (CCDC- 2237486). Thermal ellipsoids are drawn at 50% probability. In [Zn(S-APPE)(NO₃)₂]·CH₃OH and [Co(S-APPE)(NO₃)₂]·CH₃OH, the solvent molecule (CH₃OH) is omitted for clarity. (B) Schematic structure and atomic distances from central metal cations. (C) Solid lines indicate the torsion angles between defined as M–N–C–C, which are metal, 3°N, methylene C, and anthracene C. Dotted lines indicate the torsion angles defined as C*–N–C–C, which are chiral C, 3°N, methylene C, and anthracene C. Nitrate ions and the solvent molecule (CH₃OH) are omitted for clarity.

Determination of Complex Association Equilibria in Aqueous Solution

Fluorescence titration was performed to clarity complex association behaviors in aqueous solution. Fluorescence spectra of S-APPE (3) recorded in the presence of increasing amounts of divalent cations (Zn^II and Co^II) are shown in Fig. 3. Upon addition of Zn^II, the fluorescence intensity increased by 9.0 times at 415 nm (Fig. 3A), but only by 1.3 times upon addition of Co^II (Fig. 3C). APPE shows a strong increase in fluorescence intensity due to Zn^II coordination, whereas Co^II show strong quenching, as expected from its redox mechanism.^16,20) Both Zn^II and Co^II exhibited complexation with APPE with a 1 : 1 stoichiometry. An equivalent complex association stoichiometry was observed in both the crystalline and solution states. The complex association constant log K_a of Zn–S-APPE was 4.95 ± 0.07. This value is comparable to the reported log K_a of Zn–ABPM (5.3).²¹⁾ Thus, the introduction of an asymmetric methyl group does not affect the Zn^II affinity of the ligands. The log K_a value of Co–S-APPE was 5.39 ± 0.11, i.e., little difference was observed in the affinity between Zn^II and Co^II complex associations with S-APPE. On the other hand, the reported log K_a values of Zn^II and Co^II complexes of BPA were 6.8 and 5.2, respectively.²²⁾ When in complex association of BPA, Zn^II has been shown to exhibit a tenfold higher affinity than Co^II. The anthracen-9-ylmethyl group has been shown to exert no influence over the affinity between the BPA moiety and Co^II; instead, it adversely affects the affinity between the BPA moiety and Zn^II. The binding constants corresponding to Zn^II (log K_a (Zn^IIL)) and Co^II (log K_a (Co^IIL)) for each ligand are shown in Fig. 4.

Fig. 3. The Fluorescence Titration Curves of S-APPE

(A) Fluorescence–intensity change profile of S-APPE (10 µM in 80% CH₃OH–water at room temperature.; λ_ex = 350 nm) with addition of Zn(CH₃COO)₂ (0–22 µM). (B) Fluorescence titration curve of S-APPE against changing Zn(CH₃COO)₂ concentrations at an intensity of 415 nm. (C) Fluorescence–intensity change profile of S-APPE (10 µM) with the addition of Co(CH₃COO)₂·4H₂O (0–22 µM). (D) Fluorescence titration curve of S-APPE against changing Co(CH₃COO)₂·4H₂O concentrations at an intensity at 415 nm. Solid lines represent the nonlinear least-squares fit of the data using eq. (5).

Fig. 4. Association Constants Log K_as of Bis(2-picolyl)amine Derivatives

DNA Photocleavage Activity

To determine DNA cleaving activity and the photoreactivity of the synthesized ligands (1–3), their DNA cleavage activities toward pUC19 plasmid DNA were assessed by quantitating the conversion of supercoiled pUC19 (Form I) to open circular pUC19 (Form II) in a 50 mM Tris-borate buffer (pH 8.1) at 20 °C under atmospheric conditions. Reaction mixtures were prepared by mixing a metal acetate salt (200 µM), ligands (420 µM), and plasmid DNA in a plastic cell. The metal to ligands ratio was determined with reference to previously reported DNA cleavage experiment using ABPM complexed with Co^II.¹⁰⁾ Reactions were performed under conditions of continuous light irradiation (wavelength = 370 nm) using a Hitachi F-4500 fluorescence spectrometer. Ethylenediaminetetraacetic acid (EDTA) and the loading buffer were added to inhibit reactions. Aliquots were electrophoresed in 1% agarose gels containing ethidium bromide (0.5 mg/mL) at 100 V for 60 min. The fluorescence intensity of ethidium bromide was measured using a densitometer (AE-6900M, ATTO Co., Ltd., Tokyo, Japan).

First, the effects of chelated metal type and photoirradiation on DNA cleavage activity were investigated using ABPM (1) as the ligand. The results of these 30 min reactions are shown in Fig. 5. Under lightproof conditions, no DNA cleavage activity of ABPM was detected in the presence of Co^II or Zn^II (lanes 2–4). By contrast, Form II bands were observed with light irradiation and ABPM (lanes 5–7), i.e., ABPM cleaved DNA plasmid under the experimental light conditions. DNA cleavage was not observed in any reaction without ABPM (lanes 8–11). To summarize, the most efficient DNA-cleaving condition of ABPM included ligands and Zn^II under light exposure (lane 6). Bhattacharya and Mandal reported that a mixture of ABPM and Co(CH₃COO)₂·4H₂O led to photochemical cleaving of pTZ19R plasmid DNA. The differences between their results and those shown herein may be explained by the different irradiation wavelengths used. Bhattacharya and Mandal irradiated their reaction mixtures using a wide range of light, either visible light or sunlight, which is expected to induce DNA cleavage activity of their Co^II complex.¹⁰⁾ In the present study, we showed that Zn^II might improve ABPM DNA photocleavage activity. ABPM includes an anthracene moiety, which is expected to intercalate between base pairs in DNA and act as a photosensitizer, such that the photoactivated anthracene moiety activates ambient O₂ through energy transfer. In photoactivated ABPM complexed with Co^II, the energy of the photosensitizer may be transferred not only to ambient O₂ but also to Co^II. Energy transfer to Co^II interferes with energy transfer to reactive oxygen species, thereby inhibiting DNA oxidation. By contrast, energy transfer from photoactivated ABPM to coordinated Zn^II might be weaker than that resulting from Co^II because the d-orbital of Zn^II is full. This hypothesis is supported by the results of fluorometric titration showing that the fluorescence intensity of Co–APPE was significantly weaker than that of Zn–APPE. We speculate that this difference in energy transfer efficiency from photoactivated ABPM to Co^II relative to Zn^II might explain the observed differences in DNA photocleavage activity of the corresponding ABPM complexes. Based on these results, we used Zn^II as the chelated metal for further experiments.

Fig. 5. The DNA Photocleavage Activity of ABPM and Metal Complex

Reactants; pUC19 plasmid DNA strand (2.0 µg/50 µL), metals (Co(CH₃COO)₂⸱4H₂O or Zn(CH₃COO)₂; final 200 µM), and ABPM (1) (final 420 µM). Reaction conditions; The reactants were dissolved in 50 mM Tris-borate buffer (pH 8.1) at 20 °C under atmospheric conditions. For photoreactive DNA cleavage, the light (wavelength = 370 nm) was continuously irradiated cell using the Hitachi F-4500 fluorescence spectrometer for 30 min (lanes 5–7, 10, and 11).

Second, we investigated the effects of the chemical structure of the ligand on DNA cleavage activity using Zn^II as the chelated metal and a reaction time of 30 min (Fig. 6). Under lightproof conditions, no ligands exhibited significant DNA cleavage activity (lanes 2–4), whereas light irradiation evoked DNA cleavage activity in the Zn^II complex of each ligand (lanes 5–7). ABPM (1) displayed slightly stronger DNA cleavage activity than APPE (2 and 3), suggesting that branched chain amine structures might be unfavorable to the DNA photocleavage activity of the Zn^II complex of BPA. DNA cleavage by BPA derivatives with bulkier moieties has been reported.^23,24) Thus, the bulkiness of a methyl group is unlikely to affect the DNA binding affinity of APPE. Instead, the methyl group of APPE may limit free rotation of the anthracenylmethyl moiety, as suggested by X-ray crystallographic results. We thus speculate that suppression of free rotation interferes with the fit of DNA binding moieties to the DNA double strand. Significant differences in the DNA photocleavage activity were not observed between lanes 6 and 7 (Fig. 6), i.e., the absolute configuration of the chiral center in APPE did not affect DNA photocleavage activity. When the ligand structures fit to DNA strands by strong suppression of free rotation of the anthracenylmethyl moiety due to the introduction of substituents bulkier than a methyl group, the metal complex may acquire sequence-selective targeting ability.

Fig. 6. The DNA Photocleavage Activity of Synthesized Ligands Chelated Zn^II

Reactants: pUC19 plasmid DNA strand (2.0 µg/50 µL), Zn(CH₃COO)₂ (final 200 µM), and ligands (final 420 µM). Reaction conditions: reactants were dissolved in 50 mM Tris-borate buffer (pH 8.1) at 20 °C under atmospheric conditions. For photoreactive DNA cleavage, light (wavelength = 370 nm) continuously irradiated the cell using a Hitachi F-4500 fluorescence spectrometer for 30 min.

Finally, we investigated the irradiation time-dependence of DNA photocleavage activity using ABPM-chelated Zn^II (Fig. 7). Form I fluorescence intensity decreased immediately upon irradiation, and almost completely disappeared within 60 min (lanes 1–6). Form II fluorescence intensity increased with decreasing Form I intensity (lanes 1–6). Form II reduced under continued light irradiation (lanes 7 and 8) due to further cleavage and fragmentation. Although the cleavage rate of Zn–ABPM was slower than that previously reported for Co–ABPM,¹⁰⁾ it is possible that time-dependence may itself depend on the irradiation wavelength.

Fig. 7. The UV Irradiation Time Dependency against DNA Photocleavage Activity of Zn-ABPM Complex

Reactants: pUC19 plasmid DNA strand (2.0 µg/50 µL), Zn(CH₃COO)₂ (final 200 µM), and ABPM (final 420 µM). Reaction conditions: reactants were dissolved in 50 mM Tris-borate buffer (pH 8.1) at 20 °C under atmospheric conditions. For photoreactive DNA cleavage, light (wavelength = 370 nm) continuously irradiated the cell using a Hitachi F-4500 fluorescence spectrometer throughout the period of the reaction.

Conclusion

We synthesized a BPA-based chiral ligand termed APPE as DNA photocleavage reagent with DNA sequence specificity derived from the chiral center. By introducing an anthracene moiety, the ligand functioned as a DNA intercalator and a photosensitizer. APPE formed Zn^II and Co^II complexes with a 1 : 1 ratio in both the crystalline and solution states. The affinities of APPE to Zn^II and Co^II were comparable: The association constants of complexes (log K_a values) were 4.95 and 5.39, respectively. The synthesized complexes cleaved pUC19 plasmid DNA when irradiated with 370 nm UV light. The DNA photocleavage activity of the Zn^II complex was higher than that of the Co^II complex. We speculate that differences between the energy transfer efficiency of Zn^II and Co^II from the photoactivated ligand to its bound metal might explain the difference in DNA photocleavage activity of APPE. Unfortunately, we noted that Zn^II complexes of achiral APPE derivatives without a methyl group (Zn–ABPM) showed better DNA photocleavage activity than Zn–APPE. This may be due to methyl groups preventing the optimal orientation of the anthracene moiety for DNA photocleavage activity or intercalation into DNA. These negative factors could be mitigated by improvements to the photosensitizer structure and linker. A possible strategy would be to design a ligand with a high affinity for metals based on tetradentate picolylamine derivatives, such as tris(2-picolyl)amine.²⁵⁾ However, the example of Cu^II complexes shows that the steric bulkiness of the ligand may be detrimental to interactions with DNA.²⁶⁾ As also shown in this paper, BPA coordinates to metals in the mer fashion and is less susceptible to steric hindrance. In future study, we intend to develop novel ligands based on BPA by introducing photosensitive moieties other than anthracene, such as quinolones, as well as by using flexible linker moieties. Recently, the biological activities, such as anticancer^27,28) and antibacterial,²⁸⁾ of metal complexes with DNA cleavage activity have been revealed. The application of our research to give DNA sequence specificity to metal complexes and to control their activity by photoirradiation would be useful for the development of new drug candidates.

Experimental

General

Unless otherwise noted, regents and solvents were obtained from commercial suppliers and used without further purification. TLC and column chromatography were carried out using a glass TLC plate (silica gel 60 F₂₅₄, Merck, Germany) and Wakogel C-300 (silica gel, FUJIFILM Wako, Osaka, Japan), respectively. ¹H-NMR spectra were recorded using a JEOL ECA-500 (500 Hz) spectrometer in CDCl₃ with tetramethylsilane (TMS) as the internal standard (0 ppm). High-resolution mass spectra of the final products were obtained using an JEOL Accu TOF mass spectrometer. The purities of the final products were determined to be 95% using HPLC chromatograms acquired using a Shimadzu prominence HPLC system. Analyses were conducted using TSKgel ODS-100 V (particle size 5 µm, 4.6 mm I.D. × 25.0 cm, Tosoh, Tokyo, Japan) and water − CH₃OH (1 : 1) containing 0.1% formic acid. Detection was performed at 254 nm, and the averaged peak area was used to determine purity.

Chemistry

(RS)-1-(Pyridin-2-yl)-N-(pyridin-2-ylmethyl)ethanamine (5a)

A solution of picolinaldehyde (2.01 g, 18.8 mmol) and (RS)-1-pyridin-2-ylethylamine (2.30 g, 18.8 mmol) in CH₃CN (20 mL) was allowed to stand at room temperature for 18 h in the presence of molecular sieves (4A). The mixture was filtered, and the filtrate was concentrated under reduced pressure to afford compound 4a, which was used without further purification. 10% Pd/C (382 mg) was added to a solution of compound 4a in ethanol (30 mL), and the mixture was stirred at room temperature under 1 atm of H₂ for 32 h. The resulting mixture was filtered, and the filtrate was concentrated under reduced pressure to yield compound 5a, which was directly used in the next step without further purification.

(RS)-N-(Anthracen-9-ylmethyl)-1-(pyridin-2-yl)-N-(pyridin-2-ylmethyl)ethanamine (rac-APPE) (2)

SOCl₂ (1.45 mL, 22.0 mmol) was added to a solution of anthracen-9-ylmethanol (2.81 g, 13.5 mmol) in CHCl₃ (50 mL) and the mixture was stirred under reflux conditions for 18 h. The mixture was concentrated under reduced pressure to yield compound 6, which was used without further purification. A solution comprising unpurified compound 5a (2.88 g, 13.5 mmol) and triethylamine (1.93 mL, 13.8 mmol) in CHCl₃ (60 mL) was added to compound 6 in CHCl₃ (50 mL) in an ice bath. The mixture was stirred under reflux conditions for 18 h and concentrated under reduced pressure. The resulting residue was purified using column chromatography on a silica gel (eluent, CHCl₃ to CHCl₃/CH₃OH = 96 : 4) to yield compound 2 as a yellow solid (2.15 g, total yield: 28% from (RS)-1-pyridin-2-ylethylamine). ¹H-NMR (CDCl₃, 500 MHz) δ: 1.71 (3H, d, J = 6.9 Hz), 3.84 (2H, q, J = 14.2 Hz), 4.04 (1H, q, J = 6.8 Hz), 4.60 (1H, d, J = 13.1 Hz), 4.85 (1H, d, J = 13.5 Hz), 6.87–7.03 (1H, m), 7.05 (1H, d, J = 8.0 Hz), 7.16–7.20 (1H, m), 7.23 (1H, d, J = 8.0 Hz), 7.38–7.45 (5H, m), 7.63 (1H, td, J = 7.7 and 1.7 Hz), 7.91–7.97 (2H, m), 8.24–8.30 (2H, m), 8.33 (1H, s), 8.36 (1H, d, J = 4.0 Hz), 8.60 (1H, d, J = 4.0 Hz). HR–MS (DART); m/z Calcd for C₂₈H₂₆N₃ [M + H]⁺ 404.21267; Found, 404.21499.

(S)-N-(Anthracen-9-ylmethyl)-1-(pyridin-2-yl)-N-(pyridin-2-ylmethyl)ethanamine (S-APPE) (3)

Compounds 4b and 5b were prepared following the procedure described above for compound 5a using picolinaldehyde (1.33 g, 12.4 mmol) and (S)-1-(pyridin-2-yl)-N-(pyridin-2-ylmethyl)ethanamine (1.52 g, 12.4 mmol). Compound 3 was obtained by reacting 5b with 6 according the same method as in the synthesis of compound 2. The resulting residue was purified using column chromatography on a silica gel (eluent, CHCl₃ to CHCl₃/CH₃OH = 96 : 4) to yield compound 3 as a yellow solid (0.85 g, total yield: 6.8% from (S)-1-pyridin-2-ylethylamine). ¹H-NMR (CDCl₃, 500 MHz) δ: 1.71 (3H, d, J = 6.9 Hz), 3.84 (2H, q, J = 14.3 Hz), 4.04 (1H, q, J = 6.8 Hz), 4.60 (1H, d, J = 13.5 Hz), 4.85 (1H, d, J = 13.5 Hz), 6.98–7.03 (1H, m), 7.05 (1H, d, J = 7.7 Hz), 7.15–7.20 (1H, m), 7.23 (1H, d, J = 8.0 Hz), 7.38–7.45 (5H, m), 7.63 (1H, td, J = 7.7 and 1.8 Hz), 7.91–7.97 (2H, m), 8.24–8.30 (2H, m), 8.33 (1H, s), 8.36 (1H, d, J = 4.9 Hz), 8.61 (1H, d, J = 4.9 Hz). HR–MS (DART); m/z Calcd for C₂₈H₂₆N₃ [M + H]⁺ 404.21267; Found, 404.21644.

X-Ray Crystallography

Crystal data were collected on a Rigaku XtaLAB Synergy-i diffractometer using graphite-monochromate Cu-Kα radiation (λ = 1.54184 Å) at 93.15 K. The initial structure was solved using an intrinsic phasing method with SHELXT-2018/3 software.²⁹⁾ Non-hydrogen atoms were refined using the full-matrix least-squares method on F² of SHELXL-2018/3.³⁰⁾ Hydrogen atoms were located using a geometrical calculation and were not refined. All calculations were performed using Olex2 crystallographic software.³¹⁾ Crystallographic data reported in this manuscript have been deposited at the Cambridge Crystallographic Data Center (CCDC) under the following supplementary publication numbers: CCDC-2237481 for [Zn(S-APPE)(NO₃)₂]·CH₃OH, CCDC-2237486 for [Co(S-APPE)(NO₃)₂]·CH₃OH, and CCDC-2237782 for [Co(R-APPE)(NO₃)₂]·CH₃OH. Copies of the data can be obtained free of charge via the CCDC website.

Preparation and Crystal Data of Zn^II Complex [Zn(APPE)(NO₃)₂]·CH₃OH

Compound 2 (96.7 mg, 0.240 mmol) and Zn(NO₃)₂·6H₂O (72.0 mg, 0.240 mmol) were mixed with methanol (9 mL) and stirred at 40 °C, after which a few drops of water were added to obtain a clear homogeneous solution. The mixture was allowed to stand for 2 weeks at room temperature. [Zn(APPE)(NO₃)₂]·CH₃OH was obtained as colorless crystals with a yield of 47% (70 mg).

Crystal data for [Zn(APPE)(NO₃)₂]·CH₃OH: C₂₉H₂₉N₅O₇Zn, FW = 624.94, colorless block, monoclinic, Space group: P2₁/n, a = 10.51960(10) Å, b = 20.8846(3) Å, c = 12.12560(10) Å, β = 92.6060(10)°, V = 2661.21(5) ų, Z = 4, D_calcd = 1.560 g cm⁻³, No. of reflections collected = 14903, No. of independent reflections = 4849, Restraints = 0, Parameters = 385, 2θ range for data collection: 8.438 to 137.386°, R indices [I > 2σ(I)]: R₁ = 0.0349, wR₂ = 0.0912, R indices (all data): R₁ = 0.0378, wR₂ = 0.0938, Goodness-of-fit on F² = 1.061.

Preparation and Crystal Data of Co^II Complex [Co(APPE)(NO₃)₂]·CH₃OH

Compound 2 (45.2 mg, 0.112 mmol) and Co(NO₃)₂·6H₂O (72.0 mg, 0.112 mmol) were mixed with methanol (9 mL) and stirred at 40 °C. The mixture was allowed to stand for 2 d at room temperature. [Co(APPE)(NO₃)₂]·CH₃OH was obtained as dark red crystals with a yield of 61% (90 mg). Crystals of [Co(S-APPE)(NO₃)₂]·CH₃OH and [Co(R-APPE)(NO₃)₂]·CH₃OH were obtained from the same vessel by spontaneous resolution.

Crystal data for [Co(S-APPE)(NO₃)₂]·CH₃OH: C₂₉H₂₉CoN₅O₇, FW = 618.50, colorless block, monoclinic, Space group: P2₁/n, a = 10.14090(10) Å, b = 21.7845(2) Å, c = 12.05620(10) Å, β = 97.8230(10)°, V = 2638.60(4) ų, Z = 4, D_calcd = 1.557 g cm⁻³, No. of reflections collected = 29160, No. of independent reflections = 4824, Restraints = 0, Parameters = 382, 2θ range for data collection: 8.118 to 136.722°, R indices [I > 2σ(I)]: R₁ = 0.0298, wR₂ = 0.0730, R indices (all data): R₁ = 0.0313, wR₂ = 0.0797, Goodness-of-fit on F² = 1.078.

Crystal data for [Co(R-APPE)(NO₃)₂]·CH₃OH: C₂₉H₂₉CoN₅O₇, FW = 618.50, colorless block, monoclinic, Space group: P2₁/n, a = 10.13670(10) Å, b = 21.7802(3) Å, c = 12.05380(10) Å, β = 97.8200(10)°, V = 2636.48(5) ų, Z = 4, D_calcd = 1.558 g cm⁻³, No. of reflections collected = 31504, No. of independent reflections = 4825, Restraints = 0, Parameters = 382, 2θ range for data collection: 8.118 to 136.792°, R indices [I > 2σ(I)]: R₁ = 0.0341, wR₂ = 0.0783, R indices (all data): R₁ = 0.0364, wR₂ = 0.0793, Goodness-of-fit on F² = 1.035.

Fluorometric Titrations

Ten micromolar of compound 3 (S-APPE) was titrated with increasing concentrations (0–22 µM) of Zn(CH₃COO)₂ or Co(CH₃COO)₂·4H₂O in 80% CH₃OH–water at room temperature. Excitation was conducted at 350 nm, and emission spectra were recorded between 370 and 550 nm using a JASCO FP-6200 spectrofluorometer set at 10 nm slit widths.

Determination of Association Constants, K_as, of S-APPE (3) with Zn(CH₃COO)₂ or Co(CH₃COO)₂·4H₂O

The complex association constants (K_as) were calculated as the inverse of the apparent dissociation constants (K_ds).

The dissociation equilibrium is defined as

  

(1)

and with dissociation constant

  

(2)

  

(3)

The overall molar fluorescence intensity F is defined by

  

(4)

where a is the maximum fluorescence intensity and b is the fluorescence intensity without M^II, respectively.

Eqs. (2)–(3) are substituted in eq. (4), to obtain eq. (5).

  

(5)

The apparent dissociation constants, K_ds, were obtained by fitting F to eq. (5) using a non-linear least-square method (KaleidaGraph software).³²⁾

Biology

We followed a general protocol for DNA cleavage, as follows. Reaction mixtures were produced containing 0.48 µg of pUC19 DNA in a mixture of 50 mM Tris-borate buffer (pH 8.1), metal salts (final concentration; 200 µM) and ligands (final concentration; 420 µM) in CH₃OH. Reactions were performed under continuous light irradiation (wavelength = 370 nm) using a Hitachi F-4500 fluorescence spectrometer. EDTA and the loading buffer were added to stop reactions. Following the reaction, the loading buffer was added and aliquots were electrophoresed in a 1% agarose gel (containing ethidium bromide 0.5 mg /mL) at 100 V for 60 min. The fluorescence intensity of ethidium bromide was determined using a densitometer (AE-6900M, ATTO Co., Ltd., Tokyo, Japan).

Acknowledgments

We are grateful to Ms. Emiko Sakata (Kumamoto University) for her great contribution to the experimental work.

Conflict of Interest

The authors declare no conflict of interest.

References

• 1) Armitage B., Chem. Rev., 98, 1171–1200 (1998).
• 2) Da Ros T., Spalluto G., Boutorine A. S., Bensasson R. V., Prato M., Curr. Pharm. Des., 7, 1781–1821 (2001).
• 3) Vizzotto B. S., Dias R. S., Iglesias B. A., Krause L. F., Viana A. R., Schuch A. P., J. Photochem. Photobiol. B, 209, 111922 (2020).
• 4) Liu X.-W., Liu N.-Y., Deng Y.-Q., Wang S., Liu T., Tang Y.-C., Chen Y.-D., Lu J.-L., J. Radiat. Res. Appl. Sci., 13, 331–342 (2020).
• 5) Cui S., Liu G., Pu S., Chen B., Dyes Pigm., 99, 950–956 (2013).
• 6) Juraj N. P., Muratović S., Perić B., Vujičić N. Š., Vianello R., Žilić D., Jagličić Z., Kirin S. I., Cryst. Growth Des., 20, 2440–2453 (2020).
• 7) González-Montiel S., Valdez-Calderón A., Vásquez-Pérez J. M., Torres-Valencia Martín J., Martínez-Otero D., López J. A., Cruz-Borbolla J., J. Mol. Struct., 1145, 112–120 (2017).
• 8) dos Santos N. A. C., Natali M., Badetti E., Wurst K., Licini G., Zonta C., Dalton Trans., 46, 16455–16464 (2017).
• 9) Sudarga Tjakraatmadja A. A. J., Kulak C. L. N., Inorg. Chim. Acta, 452, 159–169 (2016).
• 10) Bhattacharya S., Mandal S. S., Chem. Commun., 13, 1515–1516 (1996).
• 11) Ichimaru Y., Kato K., Okuno Y., Yamaguchi Y., Jin W., Fujita M., Otsuka M., Imai M., Kurosaki H., Bioorg. Med. Chem. Lett., 35, 127782 (2021).
• 12) Chejara D. R., Kondaveeti S., Siddhanta A. K., Polym. Bull., 72, 35–48 (2015).
• 13) Kato K., Ichimaru Y., Okuno Y., Yamaguchi Y., Jin W., Fujita M., Otsuka M., Imai M., Kurosaki H., Bioorg. Med. Chem. Lett., 36, 127834 (2021).
• 14) Antonioli B., Büchner B., Clegg J. K., Gloe K., Gloe K., Götzke L., Heine A., Jäger A., Jolliffe K. A., Kataeva O., Kataev V., Klingeler R., Krause T., Lindoy L. F., Popa A., Seichtere W., Wenzel M., Dalton Trans., 24, 4795–4805 (2009).
• 15) de Silva S. A., Whitener M. A., Baron D. E., Zavaleta A., Isidor E. V., Allam O., Acta Crystallogr. C, 54, 1117–1119 (1998).
• 16) Nugent J. W., Lee H., Lee H.-S., Reibenspies J. H., Hancock R. D., Inorg. Chem., 53, 9014–9026 (2014).
• 17) Kurosaki H., Yoshida H., Yamaguchi Y., Goto M. X-ray Struct., Anal. Online, 20, x145–x146 (2004).
• 18) Škalamera Đ., Sanders E., Vianello R., Maršavelski A., Pevec A., Turelb I., Kirin S. I., Dalton Trans., 45, 2845–2858 (2016).
• 19) Götzke L., Gloe K., Jolliffe K. A., Lindoy L. F., Heine A., Doert T., Jäger A., Gloe K., Polyhedron, 30, 708–714 (2011).
• 20) Lee J. H., Jeong A. R., Shin I.-S., Kim H.-J., Hong J.-I., Org. Lett., 12, 764–767 (2010).
• 21) de Silva S. A., Zavaleta A., Baron D. E., Allam O., Isidor E. V., Kashimura N., Percarpio J. M., Tetrahedron Lett., 38, 2237–2240 (1997).
• 22) Gruenwedel D. W., Inorg. Chem., 7, 495–501 (1968).
• 23) Humphreys K. J., Karlin K. D., Rokita S. E., J. Am. Chem. Soc., 124, 6009–6019 (2002).
• 24) Carreira-Barral I., Riopedre-Fernández M., de Blas A., Mosquerac J., Vázquezd M. E., Platas-Iglesias C., Esteban-Gómez D., J. Inorg. Biochem., 205, 110995 (2020).
• 25) Mareque-Rivas J. C., Prabaharana R., de Rosales R. T. M., Chem. Commun., 1, 76–77 (2004).
• 26) Liu C., Zhou J., Li Q., Wang L., Liao Z., Xu H., J. Inorg. Biochem., 75, 233–240 (1999).
• 27) Rodríguez M. R., Lavecchia M. J., Parajón-Costa B. S., González-Baró A. C., González-Baró M. R., Cattáneo E. R., Biochimie, 186, 43–50 (2021).
• 28) Sen S., Chowdhury N., Kim T.-W., Paul M., Debnath D., Jeon S., Bagchi A., Im J., Biswas G., Bioinorg. Chem. Appl., 2022, 8453159 (2022).
• 29) Sheldrick G. M., Acta Crystallogr. A, 71, 3–8 (2015).
• 30) Sheldrick G. M., Acta Crystallogr. C, 71, 3–8 (2015).
• 31) Dolomanov O. V., Bourhis L. J., Gildea R. J., Howard J. A. K., Puschmann H., J. Appl. Cryst., 42, 339–341 (2009).
• 32) KaleidaGraph software Version 5.0. Synergy Software, Synergy Software 2457 Perkiomen Avenue Reading, PA 19606–2049, U.S.A.