Heteroleptic Copper(II) Complexes Containing an Anthraquinone and a Phenanthroline as Synthetic Nucleases and Potential Anticancer Agents

Abstract

Two ternary copper(II) complexes with an anthraquinone and a N,N-heterocyclic donor, [Cu(dmp)(L)(H₂O)](ClO₄) (1), [Cu(bpy)(L)(dmso)](ClO₄) (2), in which dmp = 2,9-dimethyl-1,10-phenanthroline, bpy = 2,2′-bipyridine, and HL = 1-hydroxyanthracene-9,10-dione were synthesized and fully characterized by conductivity, elemental, and spectral analyses (FTIR and UV-Vis; EPR and ESI-MS). The structure of 1 reveals that Cu(II) is bound to two oxygens of L, two nitrogens of dmp, and a molecule of water in the fifth position. In complex 2.1, Cu(II) is also pentacoordinated with an O-bonded dmso in the axial position. The presence of the heteroleptic complexes in solution was evidenced by ESI-MS, EPR in dmso solution and UV-Vis spectrophotometry. All complexes bind to CT-DNA with affinity constants of approximately 10⁴. Complex 2 can nick plasmid DNA but no cleavage was performed by complex 1. The investigation of DNA interactions by spectrofluorimetry using ethidium bromide (EB) showed that it was displaced from DNA sites by the addition of the complexes. The complexes inhibited the growth of chronic myelogenous leukemia and human squamous carcinoma cells with low IC₅₀ values, complex 1 being the most effective.

1. Introduction

Cancer is the second cause of death worldwide [1]. Unfortunately, the emergence of cellular resistance to chemotherapy and side effects to the current available drugs make the investigation of new synthetic derivatives an important goal [2]. Redox active metal compounds are attractive candidates as they can act by modulating redox processes and producing reactive oxygen species that can disrupt cellular components, including proteins and DNA. Recently reported positive results encourage the design of copper complexes as anticancer agents [3,4,5,6]. The hypoxic characteristic of cancer cells may induce the reduction of Cu(II) to Cu(I), allowing a selective targeting of cancer cells [6]. Intracellular copper trafficking is tightly regulated because low concentrations of the free metal ion can be toxic [7,8].

The mechanism of the anticancer activity of copper complexes is still under investigation and the following different hypotheses have been proposed: (i) in the reduced state, copper can generate ROS with the consequent degradation of vital biomolecules, (ii) copper can directly bind to biomolecules causing conformational distortions and stopping tumor growth, and (iii) copper can act by a specific mechanism called cuproptosis.

On the one hand, the ability of synthetic molecules to catalyze DNA cleavage in physiological conditions has encouraged the search for novel therapeutic agents. Therefore, several copper complexes have been envisioned as potential anticancer agents by binding to/cleaving DNA [9,10,11,12,13,14,15,16,17,18]. Copper toxicity can also occur by a mechanism different from all other pathways of cell death, called cuproptosis [19,20]. The anticancer drug elesclomol, a copper ionophore, induces cuproptosis in cancer cells [21]. Bis[(2-oxindol-3-ylimino)-2-(2-aminoethyl)pyridine-N,N′]copper(II) is a delocalized lipophilic cation that accumulates in mitochondria, causing energy depletion and activating AMP-activated protein kinase (AMPK), which induces cellular apoptosis [22].

On the other hand, antitumor antibiotics, especially anthracyclines that are considered one of the strongest anticancer drugs, appear to selectively target cancer cells over normal cells due to their high clinical efficacy. Several anticancer compounds contain an anthraquinone moiety in their chemical structure, due to its central role in the control of cellular proliferation [23]. This work describes three ternary copper(II) complexes containing a dihydroxyanthraquinone and an N,N-heterocyclic donor as potential anticancer agents.

2. Results

Two heteroleptic copper(II) complexes were synthesized by the reaction between the metal ion and a N,N-heterocyclic ligand in ethanol, followed by the addition of 1-hydroxyanthracene-9,10-dione, according to Scheme 1. The complexes were characterized by elemental and conductivity analyses, FTIR and UV-Vis, EPR and ESI-MS spectrometries. Crystals suitable for X-ray diffraction analysis were obtained from a dmso/acetone solution.

2.1. Structures

The results of the elemental analyses are in agreement with the proposed formula. The conductivities measured in 1 × 10^–3 mol L^–1 solutions of 1 and 2 were in the typical range of a 1:1 electrolyte [24], indicating that perchlorates act as counterions. The ESI spectra confirmed the formula proposed for complexes 1 and 2 (Figures S1 and S2). The isotopic distribution was calculated with the program Qual Browser (version 2.0.7 copyright^® from the Thermo Fischer Scientific Inc., Waltham, USA). In Figures S1b and S2b, one can observe a good agreement between the calculated and experimental spectra.

Crystals grown from a solution of complex 2 in a mixture of dmso:acetone yielded a species with the presence of a coordinated dmso molecule, named complex 2.1. Complexes 1 and 2.1 adopted a distorted square pyramidal geometry, in which the copper ion is bound to two oxygens of 1-hydroxyanthracene-9,10-dione, to a N,N-heterocyclic ligand via two nitrogen and to an oxygen of the solvent in the axial position (either water or dmso), in a (CuN₂O₃) coordination sphere (Figure 1a,b). The Cu-Cu distance is greater in complex 1 (5.582 Å) than in 2.1 (5.046 Å) because of the steric hindrance of the of two methyl substituents in dmp. In complex 1, the coordinated water molecule forms hydrogen bonds (OH···O) with the L ligand and ClO₄⁻ anions, forming a dimeric arrangement (Figure 1c). Figure 1d shows the tridimensional hydrogen-bonding network of type CH···O between the complex and ClO₄^– anions, which contributes to stabilizing the crystal structure of complex 2.1. In both complexes, are observed π-stacking interactions among the aromatic rings. In complex 1, these interactions occur between aromatic rings of 1-hydroxyanthracene-9,10-dione, having a centroid-centroid distance of 3.738 Å. On the other hand, in complex 2.1, π-stacking interactions are observed among bipyridine and 1-hydroxyanthracene-9,10-dione. The centroid–centroid distances are 3.657 and 3.950 Å. All these distances are shorter than 4.0 Å, indicating that these interactions are significant in the crystal packing stabilization. Figure S3 (Supplementary Materials) displays the spatial distribution of these interactions in the complexes.

The crystallographic study of complex 3 was already reported in a previous paper [10]. In complex 3, two oxygens of L and two nitrogens of phen are bonded to the copper ion. Another oxygen of a second anthracene molecule makes a long-distance bond with copper forming a [Cu₂(L)₂(phen)₂]²⁺ dimer. In the present work, we discuss some of its features in comparison to the new complexes 1 and 2.1. Complexes 1 and 2.1 do not exhibit dimeric structures since the ligand L coordinates exclusively to a single metal site in each complex, in contrast to complex 3. The near Cu···O distance is 4.516 and 4.393 Å, respectively, for complexes 1 and 2.1. When comparing these three structures, it becomes evident that, in complexes 1 and 2.1, the coordination of water and dmso molecules contributes more significantly to the solid’s stabilization compared to the π-stacking between the anthraquinone ligand and the bidentate coordination observed in complex 3. Crystal data and selected geometrical parameters are displayed in

Table 1. Crystallographic data for complexes 1 and 2.1.

Compound	Complex 1	Complex 2.1
Formula	C28H21N2O8ClCu	C26H21N2O8.5SClCu
MM/g mol−1	612.46	628.50
Crystal system	Triclinic	Triclinic
Space group	P-1	P-1
a/Å	9.2886(1)	9.0562(2)
b/Å	11.7516(2)	12.6827(2)
c/Å	12.9607(2)	13.4040(1)
α, β, γ/°	104.007(1), 104.096(1), 102.293(1)	76.097(1), 71.503(1), 76.447(1)
V/Å3	1274.57(4)	1396.18(4)
Temperature/K	301(2)	300(2)
Z	2	2
Dcalc/g cm−3	1.596	1.495
Crystal size/mm	0.06 × 0.09 × 0.19	0.04 × 0.15 × 0.23
µ(Mo Kα)/cm−1	2.671	3.146
Measured/unique reflections	37,678/5437	41,284/5981
Rint	0.0363	0.0574
Observed refletions [Fo2 > 2σ(Fo2)]	4815	5277
Refined parameters	364	372
Robs [Fo > 2σ(Fo)]/Rall	0.0401/0.0440	0.0821/0.0868
wRobs [Fo2 > 2σ(Fo)2]/wRall	0.1104/0.1135	0.2544/0.2630
S	1.041	1.077
RMS/e Å−3	0.051	0.151

and

Table 2. Selected geometric parameters.

Bond Distance/Å
	1	2		1	2
Cu-O1	1.8931(17)	1.943(3)	Cu-N1	2.035(2)	1.989(3)
Cu-O2	1.9581(14)	1.887(3)	Cu-N2	2.0467(19)	1.996(3)
Cu-O4	2.2432(16)	2.288(3)	Cu···Cu	5.582	5.046
Bond angle/°
	1	2		1	2
O1-Cu-O2	89.08(7)	91.75(11)	O2-Cu-N1	90.82(7)	167.34(13)
O1-Cu-O4	98.81(7)	93.93(12)	O2-Cu-N2	159.90(8)	90.98(12)
O1-Cu-N1	165.34(8)	93.71(11)	N1-Cu-O4	95.80(7)	95.97(12)
O1-Cu-N2	93.42(8)	169.54(12)	N1-Cu-N2	81.79(9)	81.69(12)
O2-Cu-O4	95.25(6)	95.05(13)	N2-Cu-O4	104.05(7)	95.89(13)
Hydrogen bond
D-H···A	D-H/H···A/Å	D···A/Å		D-H···A/°
1
O4-H4A···O3 i	0.85/1.94	2.782(3)		168.0
O4-H4B···O5	0.85/2.05	2.806(3)		147.5
2
C5-H5···O5	0.93/2.58	3.371(13)		142.9
C17-H17···O3 i	0.93/2.48	3.359(5)		158.3
C20-H20···O3 i	0.93/2.45	3.359(7)		167.2
C21-H21···O5 i	0.93/2.56	3.466(11)		164.0
C23-H23···O8 ii	0.93/2.48	3.287(13)		145.6
C25-H25A···O8 iii	0.96/2.68	3.512(15)		145.5
C26-H26C···O7 iv	0.96/2.62	3.476(15)		148.7

Symmetry code: complex 1: i (−x, 1 − y, 1 − z); complex 2.1: i (1 + x, y, −1 + z), ii (−x, 2 − y, 1 − z), iii (1 − x, 2 − y, 1 − z), iv (1 + x, y, z).

. Comparing the N-ligands in complexes 1, 2.1, and 3, it is possible to note that in 1, the aromatic rings of the dmp molecule do not form a plane (torsion angles of 6.76, 4.07, and 10.75°), and in complexes 2.1 and 3, these rings in bpy and phen are almost plane (4.90° in 2.1 and 0.83, 2.03, and 2.08 in 3). Similar results are observed in Cu-N bond distances that are very similar in complexes 2.1 (1.988(3) and 1.996(3) Å) and 3 (1.982(6) and 1.996(5) Å) and are greater in complex 1 (2.035(2) and 2.047(2) Å), indicating that, in complex 1, the N-ligand is weakly bonded to the metal site.

2.2. Spectral Studies

2.2.1. IR Spectral Study

The infrared spectra of the copper complexes, shown in Figures S4 and S5, confirm the coordination by the oxygens of the anthraquinone and the nitrogens of the N,N–donor ligand (bpy or dmp). Two intense bands, attributed to the stretching of carbonyl groups, shifted to lower frequencies in relation to the free ligand appearing at 1605 cm⁻¹ and 1586 cm⁻¹, in the case of complex 1, and at 1608 cm⁻¹ and 1586 cm⁻¹ for complex 2. The displacement, in relation to the free ligand, was greater for the carbonyl at C9 due to the weakening of the C=O bond and shifting the stretching to a lower frequency. The ν₃ frequency of ClO₄⁻ appeared at 1088 and 1096 cm⁻¹ in the spectra of 1 and 2, respectively.

2.2.2. EPR Spectra

The EPR spectra of complexes 1 and 2, measured in solid and dmso solutions, at 77 K and 298 K, are shown in Figures S6 and S7, respectively, and the EPR parameters are indicated in

Table 3. EPR parameters for complexes 1 and 2 in solid state and in dmso solution.

		77 K							298 K
Complex		giso	g⊥	g//	A//	A///g// *	gms2	Ams2	giso	g⊥	g//	gms2
[Cu(dmp)(L)](ClO4) 1	solid	-	2.071	2.242	-	-	-	-	-	2.076	2.239	-
[Cu(dmp)(L)](ClO4) 1	dmso	2.156	2.083	2.301	155 G	138	-	-	2.124	-	-	-
[Cu(bpy)(L)](ClO4) 2	solid	2.060	-	-	-	-	-	-	2.063	-	-	-
[Cu(bpy)(L)](ClO4) 2	dmso	2.083	2.069	2.111	75 G	-	4.172	87.0	2.069	-	-	-
		2.104	g1 = 2.316 g2 = 2.069 g3 = 1.926		A1 = 103	-			2.107	-	-	-

* A_//(cm⁻¹) = 0.4668 × 10⁻⁴.g_//. A_//(G).

. The EPR spectrum of complex 1 is indicative of axial symmetry elongated in the solid, with the parameters g_// = 2.242 and g_⊥ = 2.071, at 77 K (Figure 2), and g_// = 2.239 and g_⊥ = 2.076, at 298 K. In the dmso solution, only one isotropic value is observed at room temperature attesting to the presence of only one species with a copper center. In the spectrum of the solution at 77 K, an axial signal was observed with parameter values g_// = 2.301; A_// = 155 G and g_⊥ = 2.083. The ratio g_///A_// indicates a tetragonal or square planar geometry, and no coupling was observed between two adjacent copper centers.

For complex 2, both at low and room temperatures, there is only one isotropic signal in the solid, characteristic of octahedral complexes, with the parameters of g_iso = 2.060, at 77 K, and g_iso = 2.063, at 298 K. In the dmso solution, at room temperature, two isotropic signals are observed, indicating the presence of two chemical species with different copper centers. This probably refers to dmso coordinating to the metal center, substituting a water molecule, and causing a consequent change in the coordination sphere:

[Cu(bpy)(L)(H₂O)]⁺ + dmso ⇄ [Cu(bpy)(L)(dmso)]⁺ + H₂O

g = 2.069    g = 2.107

In accordance with this result, an O-bound dmso is present in the crystal structure of complex 2.1, obtained by slow evaporation of a solution made in a dmso:acetone mixture.

In frozen dmso solution (77 K) of complex 2, magnetic coupling between two copper centers was observed, with a typical 7-line signal, indicating a very weak intermolecular interaction between the structural units, compatible with π-stacking-type interactions (Figure 1d and Figure S3b). This signal can be assigned to a forbidden ∆ms = ±2 transition usually indicative of the presence of species with S > 1/2, confirming that some magnetic interaction occurs between two copper centres [25]. The observed signals seem to correspond to a species with axial symmetry and another one with rhombic symmetry, whose determined parameters are species 1 (axial) g_// = 2.111; A_// = 75 G and g_⊥ = 2.069 species 2 (rhombic) g_x1 = 2.329; A_x1 = 103 G; g_x2 = 2.069 and g_x3 = 1.926 (Figure 3). Therefore, it can be inferred that species 2 changes its geometry from the solid state to the frozen dmso solution, leaving an isotropic system for a rhombic complex (x ≠ y ≠ z), while species 1 is also isotropic in the solid state, but changes to an axial geometry in frozen dmso solution.

These data are corroborated by the crystallographic results, as seen in Figure 1d and Figure S3b, where the distance between the layers in π-stacking is a slightly greater (3.657 angstrom). In complex 1, this distance is even longer (3.738), and the EPR signal around 1500 G (Ms + 1) was not detectable.

2.2.3. Electronic Spectral Study

To verify the permanence of the ternary complexes in solution, spectrophotometric studies were conducted in the ultraviolet and visible region. The absorption spectra recorded for the complexes and their ligands, at 1.0 × 10^–5 mol L^–1, are represented in Figure 4.

The heterocyclic ligands (bpy, dmp, and phen) exhibit intense bands at around 270 nm, attributed to π–π* transitions [18,26] with high molar absorptivity coefficients as they are completely permitted by selection rules (Laporte and spin). Anthraquinones have four absorption bands in the 250–500 nm region due to π–π* and n–π* type transitions [27]. These bands are very sensitive either to the deprotonation of the phenolic group or to coordination with metal ions [28]. In the complexes, the free hydroxyanthraquinone band underwent a bathochromic shift, appearing at around 492 nm. This displacement evidences the deprotonation of the ligand and can be justified considering that the formation of the complex reduces the energy necessary for the electronic transition to occur. A broad band at around 290 nm is a consequence of the overlap of the N,N–donor ligands and the hydroxyanthraquinone present in the copper(II) coordination sphere.

2.3. DNA Interactions

2.3.1. DNA Binding

The binding of the complexes to calf tymus DNA (CT-DNA) was investigated by molecular absorption spectrophotometry. Compounds 1 and 2 form an insoluble ternary complex with CT-DNA in Tris-HCl buffer at pH 7.2, since the addition of CT–DNA to a solution of complexes 1 or 2 led to precipitation. By replacing the buffer with a mixture of dmso:Tris-HCl 1:2 (v:v), it was possible to solubilize the 1-DNA and 2-DNA complexes. Therefore, the interaction of 1 and 2 with CT-DNA was followed in the dmso:Tris-HCl 1:2 (v:v) solution. The control test showed that the free DNA band was not affected by using this solvent mixture.

DNA addition induced a hypochromic effect with a slight hypsochromic shift (of approximately 1 to 5 nm), indicating the interaction of the copper complex with calf thymus DNA. A titration of a 2 × 10⁻⁴ mol L⁻¹ solution of complex 1 with [DNA] varying from 0 to 5 × 10⁻⁴ mol L⁻¹ is shown in Figure 5. The binding constant, K, was calculated by the following equation:

[DNA]/(ε_a − ε_f) = [DNA]/(ε₀ − ε_f) + 1/K(ε₀ − ε_f)

where [DNA] is the concentration of DNA in base pairs, ε_a is the quotient of the absorbance/[Cu], ε_f is the extinction coefficient of the free Cu^II complex, and ε₀ is the extinction coefficient of the complex in the fully bound form. The ratio of slope to intercept in the plot of [DNA]/(ε_a − ε_f) versus [DNA] gives the value of K (inset Figure 5).

The spectrophotometric titration of complex 2 with DNA is shown in Figure S8. The calculated binding constants for complexes 1–3 are 1.90 × 10⁴, 6.38 × 10⁴, and 2.75 × 10⁴, respectively. These values are in the same range reported for copper(II) complexes with phenanthroline-type ligands (ranging from 10² to 10⁵ M⁻¹) [29,30,31,32].

Spectrofluorimetric titrations were performed using a fluorescent probe, ethidium bromide, EB. EB, a typical DNA intercalator, is almost non-fluorescent in the free form, but is intensively fluorescent when intercalated between base pairs. This property is often used to investigate if a molecule can intercalate between DNA base pairs because, in this case, EB is displaced from DNA sites and the fluorescence quenched. The addition of the studied copper(II) compounds caused the quenching of the fluorescence emission. The observed hypochroism and displacement of EB from DNA are coherent with DNA intercalation. The fluorescence intensity is directly proportional to the concentration of the copper complex. Fluorescence data were analyzed by the Stern–Volmer equation:

F₀/F = 1 + kqτ₀ [Q] = 1 + K_sv

where F₀ and F are the fluorescence intensities in the absence and presence of the complex, respectively, [Q] is the concentration of the complex, K_sv is the Stern–Volmer quenching constant, kq is the quenching rate constant, and τ₀ is the average lifetime of the fluorophore in the excited state, usually 10^–8 s for a biomacromolecule [33].

The binding constant, K_b, was calculated by the following equation:

log [(F₀ − F)/F] = log K_b + n log [Q]

where [Q] is the concentration of the quencher, n is the number of binding sites, F₀ and F are the fluorescence intensities in the absence and presence of the quencher, and n is the number of binding sites. A plot of log [(F₀ − F)/F] versus log [Q] provided a straight line, whose slope is the value of n, and the intercept the value of K_b. A representative assay made with complex 2 is depicted in Figure 6. The Stern–Volmer (K_sv) value and binding constant (K_b) obtained for the three complexes are listed in

Table 4. Stern–Volmer (K_sv), quenching rate (k_q), binding constants (K_b), and number of binding sites (n) for competitive binding of complexes 1, 2, and 3 and EB to CT-DNA.

Complex	Ksv/ L mol−1	kq/ L mol−1 s−1	Kb/ L mol⁻1	n	R2
1	8.110 × 103	8.110 × 1011	5.912 × 103	0.975	0.992
2	9.030 × 103	9.030 × 1011	4.506 × 103	0.933	0.992
3	2.155 × 104	2.155 × 1012	3.911 × 104	1.065	0.997

.

2.3.2. DNA Cleavage

The nuclease activity of the compounds was tested in plasmids, which are circular DNA molecules present in three conformations: the supercoiled form, or FI, the circular form, or FII, or the linear form, FIII. These three plasmid DNA forms can be separated when subjected to agarose gel electrophoresis. A relatively rapid migration is observed for FI; FIII migrates between FI and FII, which migrates at a slower rate.

Complex 3 is more active than 2; it nicked pUC19 to FII from 2.5 µM, and at 5 µM, the plasmid DNA was completely cleaved. From 7.5 µM, the linear form FIII started to appear. Furthermore, 2,9-dimethyl-1,10-phenanthroline was barely inactive; only a small percentage of FII was noticeable at 50 µM, as shown in Figure 7. The DNA cleavage activity of the complexes increases in the following order: 1 < 2 < 3. For comparison purposes, the nuclease activity of the complex [Cu(phen)₂]⁺ is also blocked by a substitution at ortho to the chelating nitrogens [34].

2.4. Cell Sensitivity to Compounds

The effect of the copper complexes on cell viability was evaluated in two cancer cell lines: the human squamous carcinoma (A431) and chronic myelogenous leukemia (K562). All complexes affected cell viability in a concentration-dependent manner. The IC₅₀ of complex 1 was equal to 160 nmol L⁻¹ and that of complex 2 was 9.8 µmol L⁻¹ in A431 cells. The complexes were also cytotoxic to K562 cells with IC₅₀ values of 99 nmol L⁻¹ (1), 12.7 µmol L⁻¹ (2), and 1.84 µmol L⁻¹ (3) (

Table 5. Growth inhibition values of K562 and A431 cells by compounds 1 and 2.

Compound	IC50 a (µmol L−1 ± s.d.)
	K562 Cell Line	A431 Cell Line
Complex 1	0.099 ± 0.001	0.16 ± 0.02
Complex 2	9.80 ± 0.10	12.7 ± 0.13
Complex 3 b	1.84 ± 0.07	-
[Cu(phen)2](ClO4)2	3.44 ± 0.30	-

^a IC₅₀ is the concentration needed to inhibit 50% of cell growth after 72 h of incubation. The values are the means of triplicate determinations. ^b Data from reference [10].

). The cytotoxic activity increased in the following order: 2 < 3 < 1.

3. Discussion

The antitumor drugs of the anthracycline family intercalate in DNA and inhibit topoisomerase II activity, by stabilizing the DNA–anthraquinone–topoisomerase II ternary complex [23]. In the absence of DNA topoisomerase, anthracyclines alone did not produce DNA double-strand breaks.

On the other hand, the nuclease activity of [Cu(phen)₂]²⁺, firstly reported by Sigman [35], involves its reduction to [Cu(phen)₂]⁺, which binds to the minor groove. Oxidation by hydrogen peroxide generates copper-oxo species, which are responsible for oxidative DNA cleavage [16]. In contrast, the bis(2,9-dimethyl-1,10-phenanthroline)copper(I) complex is not able to cleave DNA [34].

In the present work, two heteroleptic copper(II) complexes containing 1-hydroxyanthracene-9,10-dione and a phenanthroline-like ligand were studied. The spectroscopic results indicate that the heteroleptic environment is maintained in an aqueous solution. The complexes bind to DNA and inhibit the growth of cancer cells. The nature of the N,N-donor influences the cytotoxicity of the complexes, which decreases in the following order: 1 > 3 > 2. There is no direct correlation between the affinity of complexes to DNA and their effect on cell growth. Curiously, the presence of dmp in complex 1 rendered it the least effective nuclease and the most cytotoxic in two cancer cell lines. Inside the cells, the complexes are reduced to the tetrahedral cuprous state, which prevents full intercalation. The high cytotoxic activity of the complexes in two cancer cell lines presents them as potential anticancer drug candidates.

4. Materials and Methods

Copper(II) perchlorate hexahydrate, 1,10-phenanthroline, 2,2′-bipyridine, 2,9-dimethyl-1,10-phenanthroline hydrate, and 1-hydroxyanthracene-9,10-dione were purchased from Sigma-Aldrich, San Luis, MO, USA; 1,4-dihydroxyanthracene-9,10-dione was purchased from Fluka, San Luis, MO, USA, and used as received without any further purification. Calf thymus DNA salt (CT-DNA) and pUC 19 plasmid DNA were used as received from Sigma-Aldrich, San Luis, MO, USA. All other reagents and solvents were of analytical grade and purchased from commercial sources at the highest purity available and were employed without further purification.

4.1. Syntheses

Complexes [Cu(dmp)(L)]ClO₄ (1) and [Cu(bpy)(L)(H₂O)]ClO₄ (2), in which dmp is 2,9-dimethyl-1,10-phenanthroline, bpy is 2,2′-bipyridine, and HL is 1-hydroxyanthracene-9,10-dione, were synthesized using the following procedure. A solution of 0.6 mmol of Cu(ClO₄)₂·6H₂O (0.222 g) in 3 mL of ethanol was added to 6 mL of an ethanolic solution of dmp (0.4 mmol, 0.083 g) or bpy (0.4 mmol, 0.042 g). The mixture was stirred and refluxed for 4 h, followed by the addition of 0.4 mmol of HL (0.090 g) in 8 mL of chloroform, and the system was stirred and refluxed for 4 more hours. A purple powder was obtained, separated by filtration, washed with EtOH and CHCl₃, and dried under vacuum. The complexes were recrystallized in a mixture of dmso:acetone.

[Cu(dmp)(L)]ClO₄ (complex 1): yield: 190.2 mg, 80%. Anal. Calc. for [Cu(C₁₄H₁₂N₂)(C₁₄H₇O₃)]ClO₄ (594.46 g mol⁻¹): C, 56.57; H, 3.22; N, 4.71; Cu, 10.69. Found: C, 56.56; H, 3.04; N, 4.60; Cu, 10.10. Λ_M = 57.63 μS cm⁻¹ in DMF. FTIR (ν_max/cm⁻¹): 1605; 1586; 1512; 1396; 1301; ν (C–O) 1271; ν (Cl–O) 1088. ESI-MS(+) (CH₃OH) m/z 494.13 [Cu(dmp)(L)]⁺. Electronic spectrum (H₂O:dmso = 1:1) (λ_max) = 492 nm (ε = 3270 L mol⁻¹ cm⁻¹); 424 nm (ε = 2920 L mol⁻¹ cm⁻¹); 275 nm (ε = 35,700 L mol⁻¹ cm⁻¹); 259 nm (ε = 30,600 L mol⁻¹ cm⁻¹).

[Cu(bpy)(L)(H₂O)]ClO₄ (complex 2): yield: 186.1 mg, 83%. Anal. Calc. for [Cu(C₁₀H₈N₂)(C₁₄H₇O₃)(H₂O)]ClO₄ (560.40 g mol⁻¹): C, 51.44; H, 3.06; N, 5.00; Cu, 11.34. Found: 51.48; H, 2.40; N, 4.56; Cu, 11.28. Λ_M = 60.14 μS cm⁻¹ in DMF. FTIR (ν_max/cm⁻¹): 1608; 1586; 1515; 1449; 1397; 1312; ν (C–O) 1272; ν (Cl–O) 1096. ESI-MS(+) (CH₃OH) m/z 442.15 [Cu(bpy)(L)]⁺. Electronic spectrum (H₂O:dmso = 1:1) (λ_max) = 492 nm (ε = 5690 L mol⁻¹ cm⁻¹); 311 nm (ε = 19,200 L mol⁻¹ cm⁻¹); 302 nm (ε = 20,200 L mol⁻¹ cm⁻¹).

4.2. Spectroscopic Measurements

Infrared spectra were recorded over the region of 400–4000 cm⁻¹ with a Perkin–Elmer 283 B spectrometer, Waltham, MA 02451 USA. The samples were examined in KBr pellets.

A Cary100 Varian spectrometer was used for UV and visible absorption measurements. The complexes were dissolved in a mixture of dmso:H₂O (1:1). For the interactions with CT DNA, the complex concentration used was 2.0 × 10^–4 mol L^–1 and the DNA concentration varied from 0 to 4 × 10^–4 mol L^–1. The DNA concentration per nucleotide was determined by the ε = 6600 L mol^–1 cm^–1 at 260 nm. The ionic strength was maintained constant with 50 × 10^–3 mol L^–1 of KCl, and the pH was fixed at 7.2 with 50 mmol L^–1 of TRIS-HCl buffer. The absorbance of the DNA itself was subtracted by adding an equal quantity of DNA to both the complex and reference solutions. Fluorescence measurements were performed on a Varian Cary Eclipse fluorescence spectrophotometer.

The full-scan mass spectra of the complexes were obtained on a MicroTOF LC Bruker Daltonics spectrometer equipped with an electrospray source operating in a positive ion mode. The samples were dissolved in CH₃OH and injected in the apparatus by direct infusion.

EPR spectra were registered on a Bruker EMX instrument, operating at the X-band frequency (9.49 GHz), a 100 kHz modulation frequency, and 20 mW power, using standard Wilmad quartz tubes, at 77 K (liquid nitrogen) or at room temperature (298 K). DPPH (α,α’-diphenyl-β-picrylhydrazyl) was used as the frequency calibrant (g = 2.0036). The spectra of the complexes were registered both in the solid state and dissolved in dmso, using a 15 G modulation amplitude. The simulation and analyses of the spectra were performed by using the EasySpin 5.2.35. software package [35] in a MATLAB environment.

4.3. X-ray Crystallographic Study

The single-crystal XRD data were collected in an Oxford-Rigaku Sinergy diffractometer, using CuKα (λ = 1.54184 Å) radiation at 301(2) K. Data collection, reduction, and cell refinement were performed using CRYSALISPRO 1.1.4 software [36]. The structures were solved by direct methods through SHELXT and refined by SHELXL-2018/3 [37], with the OLEX2 system [38]. All non-hydrogen atoms were refined with anisotropic thermal parameters. H atoms connected to carbon were placed in idealized positions and treated using the rigid model, with Uiso(H) = 1.2 Ueq (C or N) for aromatic rings, CH groups and NH of the imine group, and Uiso(H) = 1.5 Ueq (C) for the methyl group. Figures were drawn using ORTEP-3 for Windows [39] and Mercury [40]. The crystallographic data for complexes 1 and 2 were deposited with the Cambridge Crystallographic Data Centre as Supplementary Publications CCDC, Nos. 2298434 and 2298433. Copies of the data can be obtained free of charge by applying to CCDC at the following website: https://www.ccdc.cam.ac.uk/structures/.

4.4. Conductivity Measurements

Conductivity measurements were performed with a Digimed DM 31 conductivity meter using a cell of constant 1.013 cm⁻¹. The solvent used was spectroscopic-grade dimethylformamide (ΛM = 1.42 μS cm⁻¹).

4.5. Elemental Analyses

Carbon, nitrogen, and hydrogen were determined on a Perkin-Elmer CHNS/O model 2400 series II, Waltham, MA 02451 USA. The copper content was determined by atomic absorption on a Hitachi spectrophotometer model 8200 (Hitachi, Ltd., Tokyo, Japan).

4.6. Plasmid DNA Cleavage

The DNA cleavage ability of complexes 1–3 was examined following the conversion of pUC 19 supercoiled DNA (FI) to open circular (FII) and linear DNA (FIII) using agarose gel electrophoresis to separate the cleavage products. In general, 400 ng of pUC19 plasmid DNA (20 ng/mL) were treated with different concentrations of copper complex for 30 min at 37 °C in Tris-HCl buffer (50 mM, pH = 7.2) containing 25 mM of NaCl and 1 mM of ascorbic acid. Thereafter, the reactions were quenched with a loading buffer solution (Tris-HCl 1 molL⁻¹, pH 6.8, 0.1% bromophenol blue, 10% glycerol) and the mixture was subjected to electrophoresis on 1.2% agarose gel containing 0.6 μg/mL of ethidium bromide in 1 × TAE buffer (40 mmol L⁻¹ of Tris-HCl, 40 mmol L⁻¹ of acetic acid, and 1 mmol L⁻¹ of EDTA; pH 8.0) at 70 V for 60 min.

4.7. Cells and Cultures

The human squamous carcinoma A431 cell line was obtained from ATCC (Manassas, VA, USA). A431 cells were grown at 37 °C with 90% humidity and a 5% CO₂ incubator in DMEM culture medium containing 1 mM of sodium pyruvate from Gibco (Grand Island, NY, USA), supplemented with 5% fetal bovine serum from LGC Biotechnology (Cotia, SP, Brazil). Trypsin solution was purchased from Gibco (NY, USA). T25 Tissue culture flasks and multi-well plates were obtained from Kasvi (Kasvi, Curitiba, PR, Brazil) and Sarstedt (SARSTEDT AG & Co. Nümbrecht, Germany). The K562 cell line was purchased from the Rio de Janeiro Cell Bank (number CR083 of the RJCB collection). This cell line was established from the pleural effusion of a 53-year-old female with chronic myelogenous leukemia in a terminal blast crisis. Cells were cultured in RPMI 1640 (Sigma Chemical Co., St Louis, MO, USA) medium supplemented with 10% fetal calf serum (CULTILAB, São Paulo, Brazil) at 37 °C in a humidified 5% CO₂ atmosphere. Cultures grow exponentially from 10⁵ cells mL⁻¹ to about 8 × 10⁵ cells mL⁻¹ in 3 days. The cell number was determined by Coulter counter analysis (Beckman Coulter, Fullerton, CA, USA).

4.8. Cell Sensitivity to Compounds

The cytotoxic activity of the complexes was assessed in two cancer cell lines: A431 from human squamous cell carcinoma and K562 from chronic myeloid leukemia. The sensitivity of A431 cells was evaluated using the 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay. This colorimetric assay is based on the reduction of yellow tetrazolium salt to purple-colored formazan crystals by metabolically active cells. Viable cells contain oxidoreductase enzymes that reduce MTT to formazan. Cells at a concentration of 1 × 10⁵ cells mL⁻¹ were incubated in culture plates in a humidified atmosphere with 5% CO₂ at 37 °C, for 24 h, to adhere. After this time, the cells were treated with different concentrations of complex 1, ranging from 0.05 to 0.3 µmol L⁻¹, or 2 ranging from 5 to 50.0 µmol L⁻¹, and re-incubated for 72 h. Then, MTT was added and the plates were incubated for an additional 4 h. The supernatant was then removed and the formazan was dissolved with dmso. Subsequently, the absorbance of formazan was measured spectrophotometrically at 570 nm. The concentration that inhibits 50% of cell viability (IC₅₀) was determined. Cytotoxicity in K562 cells was measured as described above and, after the incubation period, cell viability was checked by Trypan Blue exclusion, which stains only the dead cells.

5. Patent

Elene C. Pereira Maia and Ívina P. de Souza: copper (II) heteroleptic complexes, production process, pharmaceutical compositions and use, and patent application BR 10 2018 013398 5, filed in 2018, by Universidade Federal de Minas Gerais and Fundação de Amparo à Pesquisa do Estado de Minas Gerais.