Enantiomeric recognition of chiral L– and D–penicillamine Zinc(II) complexes: DNA binding behavior and cleavage studies

Abstract

Two designed L–/D–penicillamine based enantiomeric Zn(II) complexes 1a and 1b of 1,10–phenanthroline were synthesized and structurally characterized. The interactions of the complexes with CT DNA have been explored by absorption, fluorescence and CD measurements, revealing that both the complexes interact with DNA via electrostatic binding. All the corroborative results indicated the enantiopreferential selective binding of L–form of the complex over the D–form. A gel electrophoretic pictogram of the complexes 1a and 1b demonstrates their ability to cleave pBR322 DNA through hydrolytic process; validated by T4 religation assays; furthermore, the L–form of the complex exhibited more pronounced cleavage than the D–form. However, both complexes preferred the minor groove of the DNA double helix. Interaction studies with mononucleotides revealed that both the enantiomers possess high affinity towards the A–T base pairs of DNA, consistent with the previous reports on stereospecific selectivity of Zn(II) complexes. These studies were further supported by molecular docking studies and the resulting binding energy of docked metal complexes 1a and 1b were found to be −306.4 and −289.1 KJ mol⁻¹, respectively. The more negative relative binding energy of L–form of complex suggests greater propensity for DNA than the D–enantiomer.

---

1. Introduction

Penicillamine (d–3,3–dimethyl cysteine) is an active natural amino acid molecule demonstrating a diverse pharmacological profile. Two of the most common organic chelating agents, viz. BAL (2,3–dimercaptopropanol) and D–penicillamine have been widely used for the treatment of heavy metal toxicity.¹ However, in the case of penicillamine, only the pure D–form is of therapeutic importance as the racemate and L–isomer are toxic.² As a consequence, D–penicillamine has become increasingly important as a therapeutically beneficial molecule for treating rheumatoid arthritis, cystinuria, hepatitis and multiple schlerosis, in addition to its well known role for treating Wilson's disease.

Penicillamine, a chiral scaffold is interesting in that it: (i) contains an amide functionality capable of forming complexes which can serve as model systems to metalloenzymes of the living systems (ii) provides various binding sites viz., carboxylic, amino and sulfahydryl groups (Fig. 1). Therefore, it can be explored as a ligand for the design of metal-based cancer chemotherapeutics.

Fig. 1 Zwitterionic form of penicillamine with the three functional groups (–NH₂, –S⁻ and COO⁻).

DNA is an important component for many biochemical processes that occur in the cellular system. The different loci present in DNA are involved in various regulatory processes, such as gene expression, gene transcription, mutagenesis, carcinogenesis, etc.,³ which can be modified by the interaction of small organic molecules with specific regions in the DNA. Many molecules exert their anticancer activities by binding with DNA, thereby altering DNA replication and inhibiting the growth of tumor cells.⁴ Thus, DNA binding is a major criterion for the designing of novel anticancer agents.

As a part of our strategy of designing novel cancer chemotherapeutic agents, particularly metal-amino acid/metal-peptide complexes, the design of chemical entities bearing the phenanthroline moiety (recognition agent) and penicillamine (biologically active pharmacophore) was quite thought worthwhile. Although it is well-known that D–pen is therapeutically non-toxic and L–pen is toxic, in contrast to other L–forms of amino acids, it was interesting to study the enantiomeric disposition of ternary [Zn(II)(L–/D–pen)(phen)] complexes, 1a and 1b towards DNA. Zn complexes form the most propitious forms of Zn–metalloelement for the delivery to required cellular sites enabling Zn–dependent enzyme syntheses and facilitation of Zn–dependent biochemical processes.⁵ Zinc plays an important role in various biological systems; it is critical for numerous cell processes and is a major regulatory ion in the metabolism of cells.⁶ In the literature, diverse zinc complexes with biological activity are reported, but only zinc complexes with drugs used for the treatment of Alzheimer disease⁷ and others showing antibacterial,⁸ antidiabetic,⁹ anti-inflammatory¹⁰ and antiproliferative-antitumor¹¹ activity are structurally characterized. Herein we report the synthesis, characterization and in vitro DNA binding properties of ternary Zn(II) complexes of enantiomeric (L– and D–) penicillamine and 1,10–phenanthroline. The in vitro binding studies revealed that both the complexes 1a and 1b interacted with CT DNA via an electrostatic mode of interaction. Interestingly, the L–form of the complex, 1a exhibited a higher binding propensity for DNA compared to the D–form, 1b which could be attributed to the greater cytotoxic ability of L–enantiomer of penicillamine.

2. Materials and methods

2.1. Reagents

All reagents were of best commercial grade and were used without further purification. L–/D–penicillamine (Fluka), 1,10–phenanthroline monohydrate, Zinc acetate (Merck), H₂O₂, 3–mercaptopropionic acid, glutathione, methyl green, DAPI, tertiary butyl alcohol, sodium azide, superoxide dismutase (Sigma), 6X loading dye (Fermental Life Science), T4 DNA ligase (Calbiochem) and supercoiled plasmid DNA pBR322 (Genei) were utilized as received. Calf thymus DNA (CT DNA), guanosine–5′ monophosphate disodium salt (5′–GMP), thymine–5′ monophosphate disodium salt (5′–TMP) were purchased from Sigma Chemical Co. and Fluka, respectively.

2.2. Physical measurements

Elemental analyses were performed on an Elementar Vario EL III. Infrared spectra were recorded on an Interspec 2020 FTIR spectrometer in KBr pellets from 400–4000 cm⁻¹. The NMR spectra were obtained on a Bruker DRX–400 spectrometer operating at room temperature. Electrospray mass spectra were recorded on Micromass Quattro II triple quadrupol mass spectrometer. Molar conductance was measured at room temperature on a Digisun Electronic Conductivity Bridge. Electronic spectra, fluorescence measurements and CD spectra were recorded on a UV–1700 PharmaSpec UV–visible Spectrophotometer, Hitachi F–2500 Fluorescence Spectrophotometer and Applied Photophysics Chirascan Circular Dichroism Spectrometer with Stop flow, respectively.

2.3. DNA binding and cleavage experiments

DNA binding experiments which include absorption spectral traces, emission spectroscopy and circular dichroism conformed to the standard methods and practices previously adopted by our laboratory.^12–15 DNA cleavage experiments were performed by the standard protocol described in ref. 16, 17. While measuring the absorption spectra, an equal amount of DNA was added to both the compound solution and the reference solution to eliminate the absorbance of the CT DNA itself, and the CD contribution by the CT DNA and Tris buffer was subtracted through base line correction.

2.4. Molecular docking

The rigid molecular docking studies were performed using HEX 6.1 software,¹⁸ which is an interactive molecular graphics program to understand the drug–DNA interactions. The structure of the complex was sketched by CHEMSKETCH (http://www.acdlabs.com) and converted to pdb format from mol format by OPENBABEL (http://www.vcclab.org/lab/babel/). The crystal structure of the B–DNA dodecamer d(CGCGAATTCGCG)₂ (PDB ID: 1BNA) was downloaded from the protein data bank (http://www.rcsb.org./pdb). All calculations were carried out on an Intel pentium4, 2.4 GHz based machine running MS Windows XP SP2 as operating system. Visualization of the docked pose has been done using CHIMERA (http://www.cgl.ucsf.edu/chimera/) molecular graphics program.

2.5. Syntheses

Synthesis of [Zn(L–pen)(phen)]CH₃COO, 1a and [Zn(D–pen)(phen)]CH₃COO, 1b. To a 20 ml methanolic solution of Zn(CH₃COO)₂ (0.87 g, 4 mmol) was added simultaneously 10 ml methanolic solution of 1,10–phenanthroline monohydrate (0.79 g, 4 mmol) and L–/D–penicillamine (0.59 g, 4 mmol) drop wise and kept on reflux for 4–5 h. After refluxing, the reaction mixture was filtered to remove any precipitate which could be unreacted L–/D–penicillamine or [Zn(phen)₂] complex. The resulting solution was kept for slow evaporation at room temperature yielding a white product, which was filtered, washed with diethyl ether and dried in vacuo (Scheme 1).

Scheme 1 Synthetic route to complexes 1a and 1b.

1a: Yield: 52%, m.p.: 236 °C (d), Anal. Calc. for C₁₉H₂₁N₃O₄SZn (%): C, 50.40; H, 4.67; N, 9.28; Found: C, 50.89; H, 4.92; N, 9.98. Selected IR data (KBr, ν, cm⁻¹): 3282 ν_as(NH₂); 2595 ν(S–H)pen; 1622 ν_as(COO); 1566 δ(NH₂); 1345 ν_s(COO); 854 δ(S–H). ¹H–NMR (DMSO–d₆, 300 MHz, δ (ppm)): 8.65 (2H, phen aromatic H); 7.6–8.14 (6H, phen aromatic H + 2H, NH₂ pen); 3.5 (s, pen chiral –CH); 2.09 (s, 1H, –SH pen); 1.40 (s, 6H, –CH₃ pen). ¹³C–NMR (400 MHz, δ (ppm)): 181 (C=O pen); 148.3 (2C, aromatic C phen); 140 (2C, aromatic C, phen); 127.2 (4C, aromatic C phen); 125.25 (2C, aromatic C phen); 61.84 (1C, chiral C pen); 49.95 (1C, –C–SH pen); 26.75 (2C, –CH₃ C pen). Molar conductance (10⁻³ M, MeOH, 25 °C, Ω⁻¹cm²mol⁻¹): 87 (1 : 1 electrolyte). UV-vis (10⁻⁴ M, MeOH, nm): 271, 290. [α]_D (MeOH, 25 °C): −52. ESI-MS (m/z): 458 [C₁₉H₂₁N₃O₄SZn + 5H⁺].

1b: Yield: 58%, m.p.: 235 °C (d), Anal. Calc. for C₁₉H₂₁N₃O₄SZn (%): C, 50.40; H, 4.67; N, 9.28; Found: C, 50.79; H, 4.86; N, 9.77. Selected IR data (KBr, ν, cm⁻¹): 3292 ν_as(NH₂); 2591 ν(S–H)pen; 1620 ν_as(COO); 1516 δ(NH₂); 1375 ν_s(COO); 853 δ(S–H). ¹H–NMR (DMSO–d₆, 300 MHz, δ (ppm)): 8.68 (2H, phen aromatic H); 7.6–8.1 (6H, phen aromatic H + 2H, NH₂, pen); 3.19 (s, pen chiral –CH); 2.08 (s, 1H, –SH pen); 1.42 (s, 6H, –CH₃ pen). ¹³C–NMR (400 MHz, δ (ppm)): 181.48 (C=O pen); 148.46 (2C, aromatic C phen); 140.38 (2C, aromatic C, phen); 125.15 (6C, aromatic C phen); 62.91 (1C, chiral C pen); 49.95 (1C, –C–SH pen); 26.75 (2C, –CH₃ C pen). Molar conductance (10⁻³ M, MeOH, 25 °C, Ω⁻¹cm²mol⁻¹): 92 (1 : 1 electrolyte). UV-vis (10⁻⁴ M, MeOH, nm): 271, 291. [α]_D (MeOH, 25 °C): +31. ESI-MS (m/z): 455 [C₁₉H₂₁N₃O₄SZn + 3H⁺].

3. Results and discussion

3.1. Chemistry

Penicillamine-derived enantiomeric (L– and D–) Zn(II) complexes 1a and 1b complexes containing heterocyclic base, 1,10-phenanthroline are prepared in good yield from a general reaction in which a Zn(II) salt is reacted with the N,N–donor heterocyclic base, viz. phen and L–/D–pen. The complexes were formulated as [Zn(L–/D–pen)(phen)]CH₃COO and were characterized thoroughly from analytical and physicochemical data. The complexes showed good solubility in MeOH and DMSO, and were insoluble in hydrocarbons. They were stable in solid and solution phases. Because of the presence of chiral auxiliaries (L– and D–penicillamine) both the complexes 1a and 1b exhibited optical rotation, [α]_D values of −52 and +31, respectively indicating that the complexes are optically active.

3.2. Characterization

The elemental analyses of L– and D–penicillamine Zn(II) complexes are presented in Table 1. The IR spectra of complexes 1a and 1b were recorded in the region of 4000–400 cm⁻¹. The characteristic IR absorption frequencies (in cm⁻¹) and their assignments for the free ligand and its Zn(II) complexes are shown in Table 2.

Table 1 Elemental analysis data for complexes 1a and 1b

	Elemental Analyses (%)
Complex	Calculated			Found
	C	H	N	C	H	N
1a	50.40	4.67	9.28	50.89	4.92	9.98
1b	50.40	4.67	9.28	50.79	4.86	9.77

---

Table 2 Characteristic IR frequencies (in cm⁻¹) of free L–/D–pen and its Zn(II) complexes 1a and 1b

IR bands (cm^−1)	L–/D–penicillamine	Complex 1a	Complex 1b
ν as(NH2)	3250	3282	3292
ν(S–H)	2600–2400	2595	2591
ν as(COO)	1590	1622	1620
ν s(COO)	1397	1345	1375
δ(NH2)	1650	1566	1516
δ(S–H)	870	859	853
ν(C–H) phen	853, 737	843, 724	842, 723

---

Coordination by amino group. The IR –NH₂ stretching frequencies were used to distinguish coordinated from non-coordinated amino groups. The position and intensity of the ν(N–H) bands are influenced by hydrogen bonding and by coordination of nitrogen to metal. The most important bands in the spectra of free L–/D–pen are assigned to ν(NH₃⁺) at 3174–2967 cm⁻¹ and ν_as(COO⁻) at 1595 cm⁻¹ corresponding to the Zwitterionic form of penicillamine.¹⁹ The free L–/D–pen exhibit absorption frequency band at 3250 cm⁻¹ corresponding to ν_as(NH₂) shifted towards higher frequencies in the complexes, indicating coordination to metal through the NH₂ group.²⁰ Also confirmed by the changes in the δ(NH₂) bands from 1650 cm⁻¹ in free L–/D–pen to 1566 and 1516 cm⁻¹ in complexes 1a and 1b, respectively.²¹

Coordination by carboxylate group. The IR –O–C=O stretching frequencies distinguish coordinated from the uncoordinated carboxyl groups. The absorption bands of ν_as(COO) exhibited by the complexes (1622 and 1620 cm⁻¹ for 1a and 1b, respectively) was subsequently higher than free L–/D–pen (1590 cm⁻¹).²² The ν_s(COO) frequency in complexes 1a and 1b were shifted to lower frequencies (1345 cm⁻¹ and 1375 cm⁻¹ for 1a and 1b, respectively) as compared to the free pen ligand (1397 cm⁻¹) indicating that the terminal coordination mode of the carboxylate group of penicillamine to metal ion is via deprotonation.²³ The Δv value which represents the coordination modes of the carboxylate groups was greater than 200 cm⁻¹ indicating unidentate carboxylate ligation.

Furthermore, the ν(S–H) absorption bands at 2600–2400 cm⁻¹ in free L–/D–pen¹⁹ did not undergo any significant shift in the IR frequency (2595 and 2591 cm⁻¹ in 1a and 1b, respectively) which was suggestive of non-involvement of the –SH group in coordination to the metal ion. The bands assigned to ν(C–H) of phenanthroline moiety at 843 and 724 cm⁻¹ in 1a and 842 and 723 cm⁻¹ in 1b were shifted to lower frequencies compared to the free phen (853 and 737 cm⁻¹) indicating the formation of a coordinate covalent bond between the metal and phen N atoms.²⁴

¹H NMR. The ¹H NMR spectra of the complexes 1a and 1b were recorded in DMSO–d₆ (Fig. S1, ESI†). The –NH₂ resonance of free L–/D–pen (δ 6.10 ppm²⁵) was shifted downfield (δ ∼7.6 ppm) which merged with the aromatic proton resonances in the complexes 1a and 1b upon coordination of the –NH₂ group to the metal, however, there was no chemical shift in the resonance signals corresponding to the –SH proton (δ 1.0–2.0 ppm in free L–/D–pen) in the complexes (δ∼2.09 ppm) indicating non-involvement of –SH group in coordination. The –COOH resonance of free L–/D–pen (δ 12.0–13.0 ppm) was absent in the complexes, suggesting the replacement of the carboxylic proton and coordination through Zn metal. While the other peaks at 1.4 and 3.5 ppm correspond to the methyl protons and the proton attached to the chiral carbon of L–/D–pen moiety, respectively, confirming the presence of L–/D–pen moiety in the Zn(II) complexes. The protons in the aromatic region of the phen moiety exhibited resonance peaks in the 7.60–8.14 ppm region in the complexes, confirming the presence of phen moiety.

¹³C NMR. The characteristic resonance peaks in ¹³C NMR spectra of complexes 1a and 1b were recorded in DMSO–d₆ (Fig. S2, ESI†). The resonance of carboxylic carbon (–COOH) was shifted downfield (δ 180 ppm) in the complexes 1a and 1b compared to the ligand (δ 169 ppm),²⁵ suggesting the coordination of pen to zinc through the carboxylic moiety via deprotonation.²⁶ Other resonance peaks at around δ 61.8, 49.9 and 26.7 ppm in complexes 1a and 1b, correspond to the chiral carbon, C–SH and methyl carbons of pen moiety, respectively, confirming its presence in the complexes. The signals corresponding to the chiral C of pen (–CH–NH₂) also shifted downfield as a consequence of zinc coordination to the NH₂ group. The phenyl carbons (δ 127–147 ppm) of phen appeared at δ 125–140 ppm in the complexes.²⁷

The formation of metal complexes and the speciation of various ionic forms in a DMSO solution were studied with ESI–MS. ESI–MS spectra of the complexes 1a and 1b displayed prominent peaks corresponding to the molecular ion fragment at m/z at 458 and 455 corresponding to [C₁₉H₂₁N₃O₄SZn + 5H⁺] and [C₁₉H₂₁N₃O₄SZn + 3H⁺], respectively. Other peaks observed in complexes 1a and 1b at m/z 319 correspond to fragments obtained by the expulsion of the –CH₃COO and –C(CH₃)₂SH moieties of penicillamine, at m/z 244 after removal of a full pen moiety and at m/z 180 by the successive expulsion of the zinc metal corresponding to the phen moiety.

Both the complexes exhibited strong bands in the UV region at ∼270 nm which may be attributed to ligand based charge transfer transitions and other low energy bands at around 290 nm corresponding to π–π* transitions of the coordinated phen ligand.^28,29 The electronic spectra of complexes 1a and 1b displayed transitions at around 270 nm, characteristic of a tetrahedral environment around the Zn(II) metal center.³⁰

3.3. DNA binding and cleavage studies

Electronic absorption spectroscopy is one of the most common ways to investigate the interaction of compounds with DNA. Upon adding increasing amounts of CT DNA (0.00–3.33 × 10⁻⁵ M) to complexes 1a and 1b (6.66 × 10⁻⁶ M), the ligand based charge transfer spectral band at ∼273 nm exhibited hyperchromism with a significant blue shift of ∼4–7 nm in the band position (Fig. 2). This concomitant blue shift in the spectral band rules out the groove binding nature of the complexes, suggesting an intimate association of the complexes with CT DNA and is likely that these complexes bind to CT DNA electrostatically via external contact (surface binding) with the DNA duplex.³¹ Although there are subtle differences in binding ability of different enantiomers we can nevertheless discriminate between the two enantiomeric forms by observing their spectral titration profiles. The differences in binding of the two enantiomeric L– and D– forms are quite evident as there is greater increase in the molar extinction coefficient values attributed to hyperchromism, 39% in the case of the L–enantiomer, complex 1a with a blue shift of 7 nm in comparison to the D–enantiomer complex 1b which exhibited a relatively lower hyperchromism of 26% and a blue shift of 4 nm. In order to quantitatively compare the binding strength of complexes 1a and 1b with CT DNA, the intrinsic binding constant K_b values were calculated and were found to be 6.7 × 10⁴ and 2.6 × 10⁴ M⁻¹, for complexes 1a and 1b, respectively. The observed values revealed that the L–enantiomer complex possess a higher propensity for DNA binding in comparison with the D–enantiomer. Furthermore, it implies that the different conformational features are processed differently by cellular machinery. The L–form of the Zn(II) complex interacts with right handed B–DNA and forms a two–pole complementary complex compatible to its molecular symmetry and therefore exhibits strong electrostatic interactions via hydrogen bonding.³² Nevertheless, due to the presence of the aromatic phen moiety, a partial intercalative mode of binding cannot be ruled out.³³ The binding mode needs to be proved through some more experiments.

Fig. 2 UV-vis absorption spectra of complexes (a) 1a and (b) 1b in Tris–HCl buffer (0.01 M, pH = 7.2) upon addition of CT–DNA. Arrows indicates the change in absorbance upon increasing DNA concentration. Inset: Plots of [DNA] vs. [DNA]/ε_a–ε_f for the titration of CT–DNA with complexes.

Interaction with 5′–GMP and 5′–TMP. To further examine the mode of binding and to provide concrete evidence for electrostatic binding, the interaction of complexes 1a and 1b were carried out with 5′–GMP and 5′–TMP. The absorption spectra of the complexes 1a and 1b in presence of 5′–GMP and 5′–TMP are shown in Fig. S3, S4 (ESI†). Upon addition of the nucleotides, the absorption spectra of the complexes exhibited hyperchromism without any shift in the absorption band indicating an electrostatic mode of binding with the oxygen of the negatively charged surface phosphate group of the DNA helix.³⁴K_b values calculated for complex 1a and 1b were found to be 2.11 × 10⁴, 1.18 × 10⁴ M⁻¹ with 5′–GMP and 3.29 × 10⁴, 2.82 × 10⁴ M⁻¹, with 5′–TMP, respectively. The K_b values calculated suggested greater binding of Zn(II) complexes with 5′–TMP rather than 5′–GMP clearly indicating that zinc specifically binds to TMP through its N3 nitrogen. A literature survey reveals that Zn(II) complexes, in contrast to Cu(II), break the hydrogen bonds of A–T base pairs of the DNA helix, binding selectively to thymine, altering the local structure of the A–T regions of the DNA. The complex shows preference for thymine due to two effects—when thymine binds, hydrogen bonds are formed between NH₂ groups of the ligand and secondly a significantly stronger MO interaction was identified for thymine in comparison to guanine. The presence of two electron withdrawing oxo groups at the C₂ and C₃ positions of the thymine ring lower the energy of the lone pair orbital at N3 of thymine base.³⁵ Since the L–enantiomeric form of the complex exhibited greater binding propensity, the interaction of complex 1a was also validated by ¹H and ³¹P NMR spectroscopy. Relative shifts in the ¹H and ³¹P NMR signal of 5′–GMP and 5′–TMP in the presence of complex 1a provide information about the binding mode of this complex. The free 5′–GMP exhibited H8 signal at 8.07 ppm and H₁–H_5′ ribose proton signals at 3.85–5.80 ppm. Complex 1a + 5′–GMP (Fig. S5a, ESI†) showed no significant shifts in the H8 signal of the purine base, strongly suggesting a negligible interaction of the Zn(II) atom with the N7 of the nucleotide.³⁶ On the other hand, upon addition of 5′–TMP to 1a, the signal of the N3 atom of free 5′–TMP (7.64 ppm) undergoes a substantial shift to 7.80 ppm (Fig. S5b, ESI†) while other resonances at 6.2 and 1.98 ppm (C6–H and CH₃ of thymine group T–CH₃, respectively) in free 5′–TMP shifted to 6.16 and 1.76 ppm, respectively, suggestive of complex interaction with 5′–TMP. The ³¹P NMR spectra of complex 1a in presence of 5′–GMP revealed a shift of 3.55 (3.76 ppm in free 5′–GMP); in the presence of 5′–TMP, the complete disappearance of the signal (3.18 ppm in free 5′–TMP) was observed which was consistent with the electrostatic interaction modes through the phosphate backbone of the DNA helix. These results strongly suggest an electrostatic mode of interaction between phosphate groups (O6 of phosphate group) with the Zn(II) centre (Fig. S6 a and b, ESI†).

To further confirm the interaction between the complexes 1a and 1b and CT DNA, emission experiments were carried out. Complexes 1a and 1b exhibited luminescence in Tris–HCl buffer with a maximum wavelength at 370 nm (excited at 260 nm). The results of emission titration for complexes 1a and 1b with CT DNA are illustrated in Fig. 3. An increase in DNA concentration increased the emission intensity of the complexes 1a and 1b. The observed enhancement could be due to the relatively non-polar environment of the bound metal complex in the presence of DNA, such that the complexes were less deeply inserted inside the hydrophobic pockets or grooves of CT DNA.³⁷ The binding constant value, K for the complexes 1a and 1b determined from the Scatchard equation were calculated to be 5.9 × 10⁴ and 3.1 × 10⁴ M⁻¹, respectively. On the basis of the K values, we can evaluate the extent and mode of binding of different enantiomers showing enantiospecific discrimination.

Fig. 3 Emission enhancement spectra of the complexes (a) 1a and (b) 1b with increasing concentration of DNA in Tris–HCl buffer (0.01 M, pH = 7.2). The arrows indicate the change in intensity upon increasing concentration of the complexes.

To evaluate the interacting strength of enantiomer complexes 1a and 1b, emission quenching experiments using [Fe(CN)₆]⁴⁻ as a quencher were also performed. [Fe(CN)₆]⁴⁻ poorly quenches the fluorescence of complexes that are strongly bound to DNA whereas the complexes that are free in solution are quenched efficiently due to ion pairing.³⁸ The quenching efficiency was evaluated from the Stern–Volmer equation. The emission intensity of the complexes 1a and 1b was greatly affected by the addition of anionic quencher. The decrease in emission intensity of the complex was due to the repulsion of highly negatively charged [Fe(CN)₆]⁴⁻ from the DNA polyanion backbone, which hinders access of [Fe(CN)₆]⁴⁻ to the DNA–bound complexes.³⁹ The plot of free complexes 1a and 1b gave a K_sv value of 8.23 × 10⁴ and 6.96 × 10⁴ M⁻¹, respectively (Fig. S7, ESI†). In the presence of DNA the quenching curve was depressed, reflecting the protection of complex by the DNA helix and the K_sv value of 1a and 1b decreased to 4.22 × 10⁴ and 3.76 × 10⁴ M⁻¹, respectively.

The above results are consistent with our UV-vis titration observations that the L–enantiomeric form of the complex binds to DNA more avidly compared to the D–enantiomeric form. Such quenching behavior is regarded as non competitive quenching pertaining to electrostatic binding of the complexes with DNA.

The circular dichroism (CD) spectrum in the UV range is sensitive to the conformational changes of the helix and provides detailed information about the binding of the complex with DNA.⁴⁰ The CD of unbound CT DNA is characterized by a positive band at 275 nm and a negative band near 240 nm with a zero-crossover around 258 nm. These two bands are the net result of exciton coupling interactions of the bases which depend on the skewed orientation on the DNA backbone.⁴¹Fig. 4 displays the CD spectra of CT DNA in absence and presence of complexes 1a and 1b. Upon the addition of the L–form of complex 1a to CT DNA, both the positive and negative bands exhibited a decrease in intensity in comparison with free CT DNA in the CD spectrum. While the D–form of the complex 1b presented an inverse CD spectra in which the intensities of both positive and negative bands increased, suggesting that the complex may interact in an electrostatic mode. The nucleobases in the deep major groove of DNA double helix are preferably accessible to the positively charged transition metal complexes and neutralize the two closely associated negatively charged sugar phosphate backbones along the major groove and results in B→A conformational transitions.⁴² Both the enantiomers showed opposite spectral changes in CD spectrum upon binding to CT DNA, attributed to the different matching between the enantiomers and the DNA or to the different binding sites of the enantiomers.

Fig. 4 CD spectrum of CT–DNA alone (green), in presence of complex 1a (blue) and 1b (red).

3.4. Gel electrophoretic assay

The DNA cleavage ability of enantiomers of complexes 1a and 1b was studied by agarose gel electrophoresis using supercoiled plasmid pBR322 DNA as a substrate. The activity of 1a and 1b was assessed by the conversion of DNA from Form I to Form II or Form III. A concentration-dependent DNA cleavage by 1a and 1b was first performed. Briefly, pBR322 DNA was mixed with different concentration of 1a and 1b and the mixture was incubated at 310 K for 45 min as shown in Fig. S8a and b (ESI†). With the increase of concentration of 1a and 1b (20–35 μM) intensified nicked form (Form II, NC) was observed and the amounts of Form I diminishes gradually whereas those of Form II progressively increased (Lane 2–5) without the formation of Form III, suggesting single strand DNA cleavage. This distinct pattern of gel electrophoresis discriminates clearly DNA cleavage activity by L– and D– enantiomers of the complex; D–enantiomer reveals less efficient DNA cleavage whereas L–enantiomer cleaves DNA with higher efficiency to convert SC form into NC form completely at higher concentration.

The nuclease efficiency of complexes is usually dependent on activators. Therefore, DNA cleavage activity of 1a and 1b in presence of H₂O₂, ascorbate, 3–mercaptopropionic acid and glutathione was evaluated (Fig. 5). The cleavage activity of 1a and 1b were significantly enhanced by these activators and follows the order MPA > H₂O₂ > Asc > GSH for 1a, while 1b follows the order MPA > Asc > GSH > H₂O₂.

Fig. 5 Agarose gel electrophoresis pattern for the cleavage pattern of pBR322 plasmid DNA (300 ng) in the presence of different activating agents, ROS and groove binders at 310 K after 45 min of incubation in buffer (5 mM Tris–HCl/50 mM NaCl, pH = 7.2) by (a) 1a (35 μM) and (b) 1b (35 μM); Lane 1, DNA control; Lane 2, DNA + complex + H₂O₂ (0.4 M); Lane 3, DNA + complex + MPA (0.4 M); Lane 4, DNA + complex + GSH (0.4 M); Lane 5, DNA + complex + ASc (0.4 M); Lane 6, DNA + complex + DMSO (0.4 μM); Lane 7, DNA + complex + EtOH (0.4 μM); Lane 8, DNA + complex + NaN₃ (0.4 μM); Lane 9, DNA + complex + SOD (15 Units); Lane 10, DNA + complex + DAPI (8 μM); Lane 11: DNA + complex + methyl green (2.5 μL of a 0.01 mg ml⁻¹ solution).

To explore the mechanistic pathway of the cleavage activity, comparative DNA cleavage experiments of complexes 1a and 1b were carried out in the presence of some standard radical scavengers such as DMSO and ethyl alcohol as a hydroxyl radical scavenger (OH^•), sodium azide (NaN₃) as a singlet oxygen (¹O₂) quencher and superoxide dismutase as a superoxide anion radical (O₂^•−) scavenger and were used prior to the addition of metal complexes to DNA solution (Fig. 5a and b). Upon addition of DMSO and ethyl alcohol (Lane 6 and 7) to 1a and 1b, DNA cleavage was inhibited suggesting the possibility of hydroxyl radical as one of the active species. Thus, free radicals participate in the oxidation of the deoxyribose moiety, followed by hydrolytic cleavage of the sugar phosphate back bone in the absence of the scavengers.⁴³ On the other hand, the addition of NaN₃ (Lane 8) did not attenuate the DNA strand scission indicative of non-involvement of singlet oxygen in DNA cleavage. However, superoxide dismutase SOD (lane 9) enhances the cleavage efficiency in both the complexes, and displayed concomitant conversion of Form II to Form III. Therefore, an oxidative cleavage pathway of DNA by complex 1a has been excluded and evidently the cleavage proceeds by a hydrolytic mechanism.

The potential interacting site of complexes 1a and 1b was further explored in the presence of the minor groove binder, DAPI and the major groove binder, methyl green. The DNA cleavage activity of 1a and 1b was inhibited in the presence of DAPI while it remained unaffected in the presence of methyl green (Fig. 5a and b, Lanes 10 and 11), indicating minor groove-binding preference of the complexes.

3.5. T4 DNA ligase assay

To ascertain the discernible hydrolytic DNA cleavage pathway mediated by complexes 1a and 1b, a DNA religation experiment was performed in which supercoiled pBR322 DNA was treated with a T4 DNA ligase enzyme and subjected to gel electrophoresis.⁴⁴ The complexes 1a and 1b yielded linearized DNA which was religated using T4 DNA ligase enzyme. In our case, the nicked form (Form II) was relegated to a large extent in comparison to control DNA alone in the supercoiled form (Fig. 6), providing a direct evidence in favor of hydrolytic mechanism.

Fig. 6 Agarose gel electrophoresis pattern for the ligation of pBR322 plasmid DNA linearized by complexes 1a and 1b: Lane 1, DNA control; Lane 2, pBR322 plasmid DNA cleaved by complex 1a; Lane 3, ligation of linearized pBR322 plasmid DNA by T4 DNA ligase, Lane 4, pBR322 plasmid DNA cleaved by complex 1b, Lane 5, ligation of linearized pBR322 plasmid DNA by T4 DNA ligase.

3.6. Molecular docking

Molecular docking techniques are well-documented computational tools to understand the Drug–DNA interactions for structure-based drug design and discovery, as well as mechanistic study by placing a small molecule into the binding site of the target specific region of the DNA mainly in a non-covalent fashion.⁴⁵ Different structural properties lead to different binding modes; in fact, one of the most important factors governing the binding mode is the molecular shape. The enantiomers of chiral metal complexes 1a and 1b have different complementary shapes which complement the groove curvature (isohelicity) and functional group placement and H–bonding groups act as critical components for binding affinity and molecular recognition.⁴⁶ Herein, molecular docking studies of complexes 1a and 1b with DNA duplex of sequence d(CGCGAATTCGCG)₂ dodecamer (PDB ID:1BNA) were performed in order to predict the chosen binding site along with preferred orientation of the molecules inside the DNA minor groove. The complexes adopt a characteristic shape and are flexible enough to adopt a conformation which is complementary to the minor groove. The energetically most favorable conformation of the docked pose (Fig. 7a and b) revealed that 1a and 1b located within the central A–T (∼10.8 Å) rich regions of DNA in the minor groove compared to peripheral G–C (∼13.2 Å) ones, due to enantioselectivity and site-specificity of complexes towards A–T region and leads to van der Waals interaction and hydrophobic contacts with DNA functional groups which define the stability of groove.⁴⁷ It is well known that the interactions of chemical species with the minor groove of B–DNA differ from those occurring in the major groove, both in terms of electrostatic potential and steric effects, because of the narrow shape of the former. In contrast to the major groove, small molecules preferentially interact with the minor groove due to little steric interference.⁴⁸ Moreover, minor groove-binding molecules generally have aromatic rings connected by single bonds that allow for torsional rotation in order to fit into the helical curvature of the groove.⁴⁹ Changes in the accessible surface area of the interacting residues show a preferential binding of complexes between A–T base pairs, and bends the DNA slightly in such a way that a part of the phenanthroline ring makes favorable stacking interactions between the ring systems of the DNA bases. In addition, there are H–bonds (2.8–3.0 Å) between the –SH groups of penicillamine and C–2 carbonyl oxygen of T8 at a distance of 3.71 Å.

Fig. 7 Molecular docked model of complex (a) 1a and (b) 1b with DNA dodecamer duplex of sequence d(CGCGAATTCGCG)₂ (PDB ID: 1BNA).

The resulting binding energy of docked metal complexes 1a and 1b were found to be −306.4, and −289.1 KJ mol⁻¹, respectively, correlating well with the experimental DNA binding values. The more negative the relative binding energy of L–form of complex possess greatest propensity for DNA than D–form. Because right handed chiral DNA being complimentary to the L–form of the complex exhibits strong interactions via hydrogen bonding complimentary to its molecular symmetry.³² Thus, we can conclude that there is a mutual complement between spectroscopic techniques and molecularly docked model, which can substantiate our spectroscopic results and at the same time provides further evidence of groove binding.

4. Conclusion

Two enantiomeric pairs of Zn(II) complexes of L–/D–penicillamine and 1,10–phenanthroline 1a and 1b were synthesised and thoroughly characterized. Their in vitro DNA binding abilities were investigated by absorption, fluorescence and CD measurements. The results supported the fact that both the enantiomeric forms of complexes 1a and 1b bind to CT DNA via electrostatic mode of interaction. The binding constant values clearly indicated the enantiopreferential binding of the L– form of the complex, 1a over the D–form, 1b. The interaction studies of both the pairs of the complexes with mononucleotides suggested that the complexes prefer relatively facile A–T sequences to the G–C one exhibiting slight base pair specificity in binding to DNA. Both the complexes 1a and 1b were able to promote DNA cleavage via hydrolytic mechanistic pathway with greater cleavage ability of L–form than the D–form in which the free radicals participate in the oxidation of the deoxyribose moiety, followed by hydrolytic cleavage of the sugar phosphate back bone in the absence of the scavengers; further supported by T4 religation assay. Both the enantiomeric pairs of Zn(II) complexes preferred the minor groove of the DNA helix; which was further supported by molecular docking experiments.

Abbreviations

Pen	Penicillamine

Phen	1,10–phenanthroline

UV–vis

Ultraviolet–visible spectroscopy

5′–GMP

Guanosine 5′–monophosphate

5′–TMP

Thymidine 5′–monophosphate

Acknowledgements

The authors are grateful to Regional Sophisticated Instrumentation Center (RSIC), Central Drug Research Institute, Lucknow, Sophisticated Analytical Instrumentation Facility, Panjab University, Chandigarh, Sophisticated Test and Instrumentation Centre (STIC), Cochin University, Cochin and Advanced Instrumentation Research Facility (AIRF), Jawaharlal Nehru University, Delhi, India for providing ESI-MS, NMR, elemental analysis and CD facility, respectively.

References

1. A. Gergely and I. Sovago, Metal Ions in Biology, ed. H. Sigel, Vol. 9, M. Decker, New York, 1980.
2. W. M. Weigert, H. Offermanns and P. Scherberick, Angew. Chem., Int. Ed. Engl., 1975, 14, 330–336.
3. S. Sreelatha, P. R. Padma and M. Umadevi, Food Chem. Toxicol., 2009, 47, 702–708.
4. D. Gholamreza, E. N. D. Jafar, J. Abolghasem, A. Z. Karim, M. A. Seyed and K. Soheila, DNA Cell Biol., 2011, 30, 195–201.
5. D. M. Boghaei and M. Gharagozlou, Spectrochim. Acta, Part A, 2007, 67, 944–949.
6. G. K. Walkup, S. C. Burdette, S. J. Lippard and R. Y. Tsien, J. Am. Chem. Soc., 2000, 122, 5644–5645.
7. M. Di Vaira, C. Bazzicalupi, P. Orioli, L. Messori, B. Bruni and P. Zatta, Inorg. Chem., 2004, 43, 3795–3797.
8. Z. Q. Li, F. J. Wu, Y. Gong, C. W. Hu, Y. H. Zhang and M. Y. Gan, Chin. J. Chem., 2007, 25, 809–1814.
9. H. Sakurai, Y. Kojima, Y. Yoshikawa, K. Kawabe and H. Yasui, Coord. Chem. Rev., 2002, 226, 187–198.
10. Q. Zhou, T. W. Hambley, B. J. Kennedy, P. A. Lay, P. Turner, B. Warwick, J. R. Biffin and H. L. Regtop, Inorg. Chem., 2000, 39, 3742–3748.
11. Z. Travnicek, V. Krystof and M. Sipl, J. Inorg. Biochem., 2006, 100, 214–225.
12. J. Marmur, J. Mol. Biol., 1961, 3, 208–218.
13. M. E. Reicmann, S. A. Rice, C. A. Thomas and P. Doty, J. Am. Chem. Soc., 1954, 76, 3047–3053.
14. A. Wolfe, G. H. Shimer and T. Meehan, Biochemistry, 1987, 26, 6392–6396.
15. J. R. Lakowiez and G. Webber, Biochemistry, 1973, 12, 4161–4170.
16. F. Arjmand and M. Aziz, Eur. J. Med. Chem., 2009, 44, 834–844.
17. S. Parveen and F. Arjmand, Spectrochim. Acta, Part A, 2012, 85, 53–60.
18. D. Mustard and D. W. Ritchie, Proteins, Proteins: Struct., Funct., Bioinf., 2005, 60, 269–274.
19. G. Cervantes, V. Moreno, E. Molins and M. Quiros, Polyhedron, 1998, 17, 3343–3350.
20. S. T. Chow, C. A. McAuliffe and B. J. Sayle, J. Inorg. Nucl. Chem., 1973, 35, 4349–4354.
21. Y. Hojo, Y. Sugiura and H. Tanaka, J. Inorg. Nucl. Chem., 1976, 38, 641–644.
22. M. Nath, S. Pokharia, G. Engg, X. Song and A. Kumar, J. Organomet. Chem., 2003, 669, 109–123.
23. C. T. Yang, B. Moubaraki, K. S. J. Murray and J. Vittal, Dalton Trans., 2003, 880–889.
24. L. Jin and P. Yang, Polyhedron, 1997, 16, 3395–3398.
25. C. J. Pouchert and J. Behnke, The Aldrich Library of ¹³C and ¹H FTNMR Spectra, vol. I, Aldrich Chemical Company, Wisconsin, 1993, p. 273, 892.
26. G. Domazetis, R. J. Magee and B. D. James, J. Organomet. Chem., 1978, 162, 239–252.
27. (a) M. Nath and R. Yadav, Bull. Chem. Soc. Jpn., 1997, 70, 1331–1337; (b) Bull. Chem. Soc. Jpn., 1998, 71, 1355–1362.
28. Q. Zhang, F. Zhang, W. Wang and X. Wang, J. Inorg. Biochem., 2006, 100, 1344–1352.
29. R. Rao, A. K. Patra and P. R. Chetana, Polyhedron, 2007, 26, 5331–5338.
30. F. Arjmand and M. Muddassir, Chirality, 2011, 23, 250–259.
31. G. M. Blackburn and M. J. Gait (ed.)Nucleic Acids in Chemistry and Biology, second ed., Oxford University Press, New York, 1996, p 329.
32. P. Yang and M. Guo, Met.-Based Drugs, 1998, 5, 41–58.
33. P. J. Cox, G. Psomas and C. A. Bolos, Bioorg. Med. Chem., 2009, 17, 6054–6062.
34. E. P. Cassus, S. P. Machado and J. L. Wardell, Appl. Organomet. Chem., 2007, 21, 203–208.
35. S. Aoki and E. Kimura, J. Am. Chem. Soc., 2000, 122, 4542–4548.
36. W. Curtis–John, K. Nakanishi and N. Berova, R. W. Woody (ed.)VHS, New York, 1994, pp. 523–540 (Chapter 19.
37. S. Kashanian, M. B. Gholivand, F. Ahmadi, A. Taravati and A. H. Colagar, Spectrochim. Acta, Part A, 2007, 67, 472–478.
38. A. E. Friedman, C. V. Kumar, N. J. Turro and J. K. Barton, Nucleic Acids Res., 1991, 19, 2595–2602.
39. F. Liu, K. Wang, G. Bai, Y. Zhang and L. Gao, Inorg. Chem., 2004, 43, 1799–1806.
40. N. Berova, K. Nakanishi and R. W. Woody, (ed.) Circular Dichroism: Principles and applications, Second Ed., 2000, John Wiley and Son, Inc..
41. A. Rodger and B. Nord'en, Circular Dichroism and Linear Dichroism, 1997, Oxford Chemistry Press, UK.
42. P. Lincoln, E. Tuite and B. Norden, J. Am. Chem. Soc., 1997, 119, 1454–1455.
43. W. K. Pogozelski and T. D. Tullius, Chem. Rev., 1998, 98, 1089–1107.
44. D. Mandal, M. Chauhan, F. Arjmand, G. Aromi and D. Ray, Dalton Trans., 2009, 9183–9191.
45. R. Rohs, I. Bloch, H. Sklenar and Z. Shakked, Nucl. Acids Res., 2005, 33, 7048–7057.
46. S. Neidle, Nat. Prod. Rep., 2001, 18, 291–309.
47. R. Filosa, A. Peduto, S. Di Micco, P. de Caprariis, M. Festa, A. Petrella, G. Capranico and G. Bifulco, Bioorg. Med. Chem., 2009, 17, 13–24.
48. R. Corradini, S. Sforza, T. Tedeshi and R. Marchelli, Chirality, 2007, 19, 269–294.
49. R. B. Silverman, The Organic Chemistry of Drug Design and Drug Action, Second ed., Elsevier, New York, 2004.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c2ra20660a/

---