Copper(II) complexes with a hybrid phosphinyl-phenanthroline ligand: synthesis, crystal structures and effect of phosphinyl functionalization on the cytotoxicity

Abstract

Three new mononuclear copper(II) coordination compounds with 2-(diphenylphosphinyl)-1,10-phenanthroline (L) have been synthesized: [CuLCl₂]·1.2H₂O (1), [Cu(H₂O)LBr₂]·0.14EtOH (2) and [Cu(bipy)(H₂O)L](NO₃)₂·2.8H₂O (3). The complexes have been characterized in the solid state by CHN and thermogravimetric analyses, IR spectroscopy, and powder and single-crystal X-ray diffraction, while conductometry and UV-vis spectroscopy were applied for characterization in solution. In the crystal structures, the ligand L adopts a tridentate coordination mode; the metal centers display distorted square-pyramidal geometries in the case of halide complexes, while an octahedral geometry is observed in the nitrate complex. The cytotoxicity of L and complexes 1–3 was investigated against four tumor cell lines (A549, Hep2, HepG2, and MCF7) and non-tumor lung fibroblasts MRC5. All complexes possess cytotoxic activity in the micromolar concentration range. The modification of 1,10-phenanthroline with a diphenylphosphinyl fragment reduces the cytotoxicity of the complexes by an order of magnitude while providing no selectivity advantage over the reference compound [Cu(phen)Cl₂] in the 2D cell model. Meanwhile, the cytotoxicity of complexes 1 and 3 increases at least twofold in the 3D cell model (Hep2 spheroids). Potential mechanisms of action, including apoptosis induction and ROS generation, have been investigated.

---

Introduction

Transition metal complexes with polydentate ligands represent a substantial area of coordination chemistry and have been intensively studied over the past few decades. Among the great diversity of bifunctional ligands, significant attention has been given to those combining highly electronegative O or N donor atoms with P or S atoms.^1–7 The corresponding complexes of a wide range of metals (e.g. Co, Ni, Cu, Pd, and Au) with ligands containing P,N-(phosphine–amine/oxazoline/nitrile) and P,O-donor atoms (phosphine–ether/phosphine oxide) have been extensively investigated from the point of view of the catalytic activity,^1,8,9 as the synergistic interplay between hard and soft donor atoms modulates electronic and steric properties at the metal center influencing reactivity, stability and selectivity in catalytic transformations. Bipyridyl- and phenanthroline-based phosphine oxides appear to be promising ligands for rare earth metal complexes due to their remarkable photophysical properties and ability to discriminate ions of f-elements by their size.^10–13 Nevertheless, relatively few studies have addressed both the synthesis and biological activity of transition metal complexes with the aforementioned type of ligand.

Previously, we synthesized and characterized cytotoxic copper(II) complexes incorporating 2,2′-bipyridine or 1,10-phenanthroline derivatives paired with either diphenylphosphinic/di(o-tolyl)phosphinic acid,^14,15 but faced the problems of diphenylphosphinate complex dissociation and di(o-tolyl)phosphinate complex insolubility in aqueous solution. Building on these findings, it was hypothesized that a hybrid ligand combining structural features of both diphenylphosphine oxide and 1,10-phenanthroline could overcome these challenges. Taking into account the stability of 1,10-phenanthroline-based copper(II) complexes (log K₁ = 9.1¹⁶), the incorporation of an additional donor atom to create a tridentate coordination sphere would additionally enhance the chelate effect through the formation of two metallocyclic rings, thereby improving the complex stability against dissociation. It is worth mentioning that the nature of the substituents in 1,10-phenanthroline's aromatic rings affects not only the physicochemical characteristics of the ligand but also biological ones, and alters the cytotoxic profiles of the resulting complexes.^17,18 For instance, alkyl substituents at various ring positions usually increase the cytotoxicity of complexes,^17,19 with similar effects observed for keto functionalization at the 5- and 6-positions.^20,21

In the present study, two halide copper(II) complexes with 2-(diphenylphosphinyl)-1,10-phenanthroline and one complex with 2,2′-bipyridine and 2-(diphenylphosphinyl)-1,10-phenanthroline were synthesized. New crystal structures of the copper(II) complexes were established by single-crystal X-ray diffraction analysis. The obtained compounds were examined for their cytotoxic activity against larynx carcinoma Hep2, breast carcinoma MCF-7, lung carcinoma A549, hepatocellular carcinoma HepG2 and non-tumor lung fibroblast MRC5 human cell lines. The effect of diphenylphosphinyl functionalization of 1,10-phenanthroline on the cytotoxicity of the complexes was investigated.

Results and discussion

Synthesis and characterization

2-(Diphenylphosphinyl)-1,10-phenanthroline (L) was synthesized from 2-chloro-1,10-phenanthroline by adding Ph₂PLi, followed by oxidation of the resulting product with hydrogen peroxide. The complexes [CuLCl₂]·1.2H₂O (1) and [Cu(H₂O)LBr₂]·0.14EtOH (2) were obtained from an ethanol solution of 2-(diphenylphosphinyl)-1,10-phenanthroline upon addition of the corresponding copper(II) halides. The complex [Cu(bipy)(H₂O)L](NO₃)₂·2.8H₂O (3) was prepared by reacting Cu(NO₃)₂·3H₂O with 2,2′-bipyridine and 2-(diphenylphosphinyl)-1,10-phenanthroline in ethanol (Scheme 1). The yields of the target compounds ranged from 57% to 64%. Complexes 1–3 are soluble in ethanol, dimethyl sulfoxide, acetonitrile and water.

Scheme 1 Synthesis of complexes 1–3.

The composition of complexes 1–3 was confirmed by CHN and thermogravimetric analyses and IR spectroscopy (Fig. S1 and S2). The TGA curve of 1 displays the first mass loss of 3.9% between 45 and 140 °C, which is assigned to the release of non-coordinated water molecules. Complex 2 exhibits solvent loss from room temperature to 70 °C, corresponding to the elimination of outer-sphere ethanol molecules. Subsequently, the TGA curve reveals a further mass loss of 3.7% within the range of 175–235 °C, indicating the removal of coordinated water molecules, which are absent in crystal 2a. In the case of 3, between 40 and 120 °C, a mass loss of 8.4% aligns with the release of all water molecules (theoretical: 8.6%), while the next stages (above 205 °C) are attributed to the decomposition of the complex. Complexes 1–3 were characterized by powder X-ray diffraction. The presence of both coordinated and outer-sphere water and ethanol molecules in the complexes likely contributes to the observed discrepancy between the experimental and simulated X-ray diffraction patterns (Fig. S3). Complexes 1–3 were recrystallized from an EtOH : MeCN (1 : 1) mixture. The obtained XRD patterns agreed well with the theoretical X-ray powder patterns of complexes 1a–3a, providing strong support for the originally proposed coordination sphere of the copper atoms.

The recorded IR spectra of complexes 1–3 contain characteristic frequencies of the P=O bond stretch vibrations (≈1101–1180 cm⁻¹; strong), stretch vibrations of the aromatic rings (≈1435–1639 cm⁻¹; variable intensity) and C–H bonds (≈2924–3074 cm⁻¹; weak). The intense absorption bands at 1587 cm⁻¹ (1, 2) and at 1566 cm⁻¹ (3) belong to 1,10-phenanthroline²² and 2,2′-bipyridine^14,15 oscillations, respectively. The bands at 829, 864 and 1329 cm⁻¹ are assigned to stretch vibrations of the nitrate anion in 3. All complexes contain crystallization solvent molecules as confirmed by the appearance of broad bands of O–H bond stretch vibrations at 3400 (1), 3413 (2) and 3389 (3) cm⁻¹. Meanwhile, the lower frequencies (3287, 3296 (2), and 3250 (3) cm⁻¹) indicate the presence of coordinated water molecules.

Crystal structure description

The structure of [CuLCl₂]·EtOH·0.5H₂O (1a). The copper(II) chloride complex is mononuclear, and the ligand exhibits tridentate coordination via an oxygen atom and two nitrogen atoms of the phenanthroline moiety (Fig. 1). The coordination number up to five is completed by two chlorine atoms. As a result, the coordination polyhedron can be described as a square pyramid since the τ₅-parameter equals 0.071.²³ The bond lengths of the central atom with nitrogen or oxygen atoms are in the range of 1.99–2.07 Å (Table 1), whereas the Cu–Cl bonds are longer, 2.2129(9) and 2.4611(9) Å. A non-coordinated water molecule forms hydrogen bonds with an ethanol molecule (2.888 Å) as well as with chlorine atoms (3.244 Å) of neighboring complexes (Fig. S4a). The presence of phenanthroline in the ligand composition results in intermolecular π–π interactions with a distance of 3.880 Å (Fig. S4b). These interactions contribute to the packing of molecules into a chain (Fig. S4c).

Fig. 1 The structure of [CuLCl₂]·EtOH·0.5H₂O. Solvent molecules are not shown.

Table 1 Bond lengths (Å) as well as distances in intermolecular hydrogen bonds (Å) and π-stacking (Å) observed in the complexes

Bond	1a	2a	3a
Cu–N(L)	1.993(3)	1.987(3)	2.074(2)
	2.071(3)	2.044(3)	2.256(3)
Cu–O(L)	2.072(2)	2.039(2)	2.406(2)
Cu–Hal	2.2129(9)	2.3575(6)	–
	2.4611(9)	2.7245(7)
Cu–O(H2O)	–	–	1.988(2)
Cu–N(bipy)	–	–	2.005(3)
			2.026(3)
O⋯O	2.888	–	2.698
			2.707
O⋯Hal	3.244	–	–
phen⋯phen	3.880	–	3.845
			3.882

---

The structure of [CuLBr₂] (2a). The bromide complex has a similar structure to that of 1a, except for the presence of solvent molecules (Fig. 2). The ligand is also tridentate, and the coordination compound is neutral due to two terminal bromide ions. The coordination polyhedron is close to a square pyramid, with τ₅ = 0.194. As shown in Table 1, the bond lengths of the ligand donor atoms to the copper(II) ion are from 1.99 to 2.04 Å, while the Cu–Br bonds are longer than those in the chloride complex (2.36 and 2.72 Å). In contrast to the previously described complex, π-stacking is not observed here.

Fig. 2 The structure of [CuLBr₂].

The structure of [CuL(bipy)(H₂O)](NO₃)₂ (3a). This coordination compound consists of a cationic part, [CuL(bipy)(H₂O)]²⁺, and two nitrate anions as counterions. The central atom is surrounded by neutral ligands: L (a tridentate ligand), a terminal water molecule and bipy (a chelating ligand) (Fig. 3). Thus, the coordination number of the copper(II) ion is six, and the bond lengths range from 1.99 to 2.41 Å, with the Cu–O(L) bond being longer than those in the two previous complexes (Table 1).

Fig. 3 The structure of [CuL(bipy)(H₂O)](NO₃)₂. Hydrogen atoms are not shown except for those involved in hydrogen bonds (blue dotted lines).

Two non-coordinated nitrate anions are involved in the formation of hydrogen bonds with the water molecule (Fig. 3). An intramolecular π–π interaction occurs between the phenyl group of L and bipy, with a distance of 3.845 Å (Fig. S5). Additionally, π-stacking is also observed between the phen moieties of neighboring molecules, similar to the chloride complex (Fig. S5).

Stability study in solution

The molar conductivities of the copper(II) complexes were measured in ethanol and water at room temperature to evaluate their electrolytic behaviour. The observed values for the halide complexes (31 and 38 Ω⁻¹ cm² mol⁻¹ for 1 and 2, respectively) lie near or within the range typical for 1 : 1 electrolytes in ethanol (35–45 Ω⁻¹ cm² mol⁻¹),²⁴ suggesting partial dissociation of one halide ion (e.g., [CuLCl]⁺ + Cl⁻ or [Cu(H₂O)LBr]⁺ + Br⁻). The Λ_m value (61 Ω⁻¹ cm² mol⁻¹) for [Cu(bipy)(H₂O)L](NO₃)₂·2.8H₂O (3) exceeds the 1 : 1 electrolyte range and approaches expectations for a 1 : 2 electrolyte (70–90 Ω⁻¹ cm² mol⁻¹), where the complex fully dissociates into a dicationic species [Cu(bipy)(H₂O)L]²⁺ and two nitrate counterions. In water, all three complexes act as 1 : 2 electrolytes (Λ_m = 229–265 Ω⁻¹ cm² mol⁻¹).

Complexes 1–3 were characterized in solution (ethanol, PBS/ethanol mixture, and water/ethanol mixture) by UV-Vis spectroscopy. Spectra were recorded immediately after dissolution (t = 0 h) and monitored over 48 h (t = 24, 48 h) to evaluate the stability of the compounds. Broad absorptions were observed in ethanol solutions in the visible range at 810–820 nm for 1 and 2, and at ∼720 nm for 3, attributed to the ²E_g → ²T_2g transitions in Cu(II). The absence of spectral shifts, hyperchromic or hypochromic effects during the observation period confirms the stability of the cationic forms of the complexes formed in non-aqueous solution (Fig. 4).

Fig. 4 Absorption spectra of 1 and 3 in the visible range in ethanol at t = 0, 24, and 48 hours.

Initial ethanol solutions of the complexes were diluted with water or PBS to achieve the desired concentrations and the spectra were recorded in the UV range at t = 0, 24, and 48 h (Fig. S6–S8). Over the 48-hour period, the spectra showed no band shifts and minimal intensity variation, indicating aqueous stability.

Cytotoxic activity

High-content analysis (Hoechst 33342/propidium iodide staining) was used to evaluate the cytotoxicity of L and complexes 1–3 on human tumor cells derived from lung (A549), liver (HepG2), breast (MCF7) and larynx (Hep2) tissues. To assess the selectivity of action, the study was also conducted on non-tumor lung fibroblasts (MRC5). The cytotoxicity study results expressed as LC₅₀ (50% lethal concentration) values, along with cell viability graphs, are summarized in Table 2 and Fig. S9–S13. L demonstrated cytostatic activity in all tested cell lines, and was cytotoxic to non-tumor lung fibroblasts with an LC₅₀ value of 33 ± 7 µM. Exposure to the ligand resulted in weaker cytotoxicity in other cell lines (the LC₅₀ value was not reached), accompanied by an elevated apoptosis level as observed in fibroblasts (Fig. S9–S13). 1,10-Phenanthroline and its derivatives are known to exhibit both cytotoxic and cytostatic properties on tumor cells within the micromolar concentration range.¹⁹

Table 2 Cytotoxic activity (expressed by LC₅₀ and SI) of the compounds against Hep2, MCF7, HepG2, A549 and MRC5 cell lines after 48 hours of exposure

Compound	LC50, µM (SI)
	Hep2 cells (2D model)	Hep2 cells (3D model)	MCF7 cells	HepG2 cells	A549 cells	MRC5 cells
L	>100 (<0.3)	—	>100 (<0.3)	—	>100 (<0.3)	33 ± 7
1	47 ± 8 (0.6)	<25	46 ± 7 (0.6)	57 ± 4 (0.5)	>50 (<0.6)	29 ± 3
2	16.8 ± 1.4 (1.5)	21 ± 1	30 ± 3 (0.8)	58 ± 3 (0.4)	>50 (<0.5)	25 ± 4
3	27.6 ± 2.1 (0.2)	10 ± 1	23.7 ± 2.0 (0.3)	25.9 ± 0.4 (0.2)	>50 (<0.1)	6.5 ± 0.2
[Cu(bipy)Cl2]	34.7 ± 0.4 (1)	>50	72.5 ± 0.6 (0.5)	52 ± 2 (0.7)	>100 (<0.4)	37.1 ± 0.8
[Cu(phen)Cl2]	3.0 ± 0.4 (1)	16 ± 7	3.6 ± 0.3 (0.8)	5.46 ± 0.14 (0.5)	5.0 ± 0.1 (0.6)	2.87 ± 0.14
Cisplatin	9.2 ± 0.5 (>5)	>25	33.7 ± 1.8 (>1.5)	33 ± 5 (>2)	>50 (−)	>50

---

Compared to the ligand, copper(II) complexes 1–3 showed increased cytotoxicity (LC₅₀ = 6.5–58 µM) in all cell lines except A549, which turned out to be the least sensitive to the effects of coordination compounds (LC₅₀ > 50 µM, cytostatic effect). While the halide complexes 1 and 2 possessed generally similar toxicological profiles (except in Hep2 cells, where the bromide analogue demonstrated enhanced activity), the bipyridine-containing complex 3 displayed greater toxicity, especially toward MRC5 cells. Notably, the complexes showed no tumor selectivity (SI ≤ 1) in our evaluation. Under the influence of the complexes, the level of apoptosis generally didn’t increase significantly, except in MRC5 and A549 cells upon incubation with complex 1, where the apoptosis rate reached up to 40% (Fig. S12 and S13). A comparison of the cytotoxic properties of complexes 1–3 with the reference compound [Cu(phen)Cl₂]²⁵ allows us to conclude that the modification of phenanthroline with a diphenylphosphinyl fragment reduces cytotoxicity of the complexes by an order of magnitude, while providing no selectivity advantage over reference compounds in vitro in a 2D cell model.

Next, the cytotoxicity of 1–3 was evaluated in a 3D cell model – Hep2 spheroids – which exhibits nutrient gradient effects (hypoxic core, proliferating outer layer), restricts drug penetration into deeper layers, and thus more closely mimics the in vivo tissue architecture (Table 2). However, for these reasons, spheroids are usually less sensitive to the cytotoxic action of compounds compared to 2D cell culture models.^26,27 This exact behavior with an increase of the LC₅₀ values was noted for reference compounds – [Cu(bipy)Cl₂], [Cu(phen)Cl₂] and cisplatin. According to the obtained data (Fig. 5, Fig. S14 and Table 2), the cytotoxicity of complexes 1 and 3 increased at least twofold, whereas that of 2 showed a slight decrease. For example, the LC₅₀ value of 3 decreased from 27.6 ± 2.1 µM in the 2D model to 10 ± 1 µM in spheroids. One possible explanation for the increased activity of 1 and 3 in the 3D cell model may be that background levels of various enzymes in 3D cultures may differ significantly from those in 2D monolayer culture. For example, the study²⁸ showed that HepG2 spheroids (hepatocellular carcinoma) were more sensitive than a 2D model for detecting DNA damage caused by some genotoxic compounds. A compound that shows higher activity in the 3D model may be more effective in vivo and work better against actual tumors as it can overcome barriers like limited drug penetration and hypoxia. These results suggest that complexes 1 and 3 are promising candidates for further investigation.

Fig. 5 Hep2 spheroids (one of Z-planes) treated with 5 and 12.5 µM concentrations of complex 3 and untreated Hep2 spheroids (control) after dual staining with Hoechst 33342 (on the left) and propidium iodide (on the right).

ROS generation study

Since many cytotoxic complexes exhibit their effects through ROS-mediated pathways,^29–31 we decided to study ROS generation in Hep2 cells following 1 h of incubation with complexes 1–3 (1–30 µM concentration range) using 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA) staining. According to the obtained results (Fig. 6), the complexes displayed comparable behavior – the ROS levels remained close to those of the negative control (untreated cells), indicating no additional ROS generation. Therefore, the observed cytotoxicity likely arises from alternative mechanisms of action.

Fig. 6 Total intracellular reactive oxygen species in Hep2 cells determined by fluorescent H₂DCFDA signal intensity. Green bars indicate intensity in cells incubated with complexes 1–3, and red bars indicate H₂DCFDA intensity in the positive control (cells incubated with 100 µM H₂O₂).

Experimental

Reagents and materials

2,2′-Bipyridine (Scheme 2) was acquired from ABCR (Germany). Commercially available copper(II) chloride dihydrate, copper(II) bromide and copper(II) nitrate trihydrate were of analytical grade. Solvents and reagents were used as purchased without any further purification.

Scheme 2 Structural formulas of 2-(diphenylphosphinyl)-1,10-phenanthroline (L) and 2,2′-bipyridine (bipy).

Methods and instrumentation

Elemental analysis (C, H and N) was performed using a CHNS-analyzer vario MICRO cube (Elementar Analysensysteme GmbH, Hanau, Germany). Infrared (IR) absorption spectra of neat samples were recorded on a Bruker SCIMITAR FTS 2000 spectrophotometer (Bruker, Billerica, MA, USA) at 4000–400 cm⁻¹. Powder X-ray diffraction (PXRD) analysis of the complexes was performed on a Shimadzu XRD-7000 (Shimadzu, Kyoto, Japan) diffractometer (CuKα radiation, Ni filter, 5–40° 2θ range, 0.03° 2θ step, room temperature). Thermogravimetric analysis (TGA) of the coordination compounds was carried out under He flow (30 cm³ min⁻¹) at a 10 K min⁻¹ heating rate on a NETZSCH thermobalance TG 209 F1 Iris (Erich NETZSCH GmbH & Co. Holding KG, Selb, Germany) in Al crucibles (loads of 5–10 mg).

The single-crystal XRD data for complexes 1a and 3a were collected using a Bruker D8 Venture diffractometer with graphite-monochromated CuKα radiation (λ = 1.54178 Å). Absorption corrections were performed by the SADABS program.³² In the case of compound 2a, an automated Agilent Xcalibur diffractometer with an area AtlasS2 detector was applied for the collection of data (graphite monochromator, λ(MoKα) = 0.71073 Å). The CrysAlisPro program package was used for integration, absorption correction, and determination of the unit cell parameters.³³ The SHELXTL^34,35 programs and OLEX2 GUI³⁶ were applied to solve structures by the direct method and refine by the full-matrix least-squares method. The atomic thermal displacement parameters for non-hydrogen atoms were refined anisotropically. The positions of the hydrogen atoms were calculated geometrically and refined using the riding model. Table S1 summarizes the crystallographic data and details of the structure refinements.

The conductance was determined for ethanol solutions of the complexes (10⁻³ M) using a COM-100 conductometer at T = 23 °C. Time-dependent UV-vis absorption spectra of complexes 1–3 in ethanol (C₁ = 4.0 mM, C₂ = 2.5 mM, C₃ = 4.2 mM), a water–ethanol mixture (C₁ = C₂ = C₃ = 20 µM), and a phosphate buffered saline (PBS)/ethanol mixture (0.01 M phosphate-buffered saline, pH 7.3–7.5, C(NaCl) = 0.137 M, C(KCl) = 0.0027 M, C₁ = C₂ = C₃ = 20 µM) were recorded at ambient temperature by means of a SF-102 spectrophotometer at t = 0, 24, and 48 hours using cuvettes of 1 cm path length. The complexes were dissolved in ethanol, and the resulting solutions were diluted with water or phosphate buffered saline to the required concentration.

Synthesis of 2-(diphenylphosphinyl)-1,10-phenanthroline (L). The ligand was synthesized by the treatment of 2-chloro-1,10-phenanthroline with Ph₂PLi³⁷ followed by oxidation of the formed tertiary phosphine with hydrogen peroxide (H₂O₂/acetone, r.t., 1 h). The ¹H NMR spectrum of the synthesized sample of L is similar to the described one.³⁸

Synthesis of [CuLCl₂]·1.2H₂O (1). A sample of L (0.049 g, 0.13 mmol) was dissolved in 4.0 ml of ethanol. The copper(II) chloride dihydrate sample (0.023 g, 0.13 mmol) was then added and dissolved with stirring, forming a bright-green solution. Stirring for 1.5 h resulted in the formation of a green precipitate, which was filtered out, washed with ice-cold ethanol and dried in air. Single crystals of [CuLCl₂]·EtOH·0.5H₂O (1a), suitable for single-crystal XRD, were obtained after repeating the synthesis with the addition of 1.5 ml of acetonitrile. Yield: 61% (0.043 g). Elemental analysis (%): Calc. for C₂₄H_19.4Cl₂CuN₂O_2.2P: C 53.8; H 3.6; N 5.2. Found: C 53.2; H 3.7; N 5.0. IR-spectra (ν, cm⁻¹): 1126, 1136, 1167 ν(P=O); 1437, 1454, 1483, 1502, 1572, 1587, 1601, 1632 R_rings; 2924, 2992, 3019, 3034, 3061 ν(C–H); 3444, 3543 ν(O–H). TGA: 3.9% weight loss at 140 °C (calculated for 1.2H₂O – 4.0%). Λ_m = 31 (ethanol) and 265 (water) Ω⁻¹ cm² mol⁻¹. UV-vis spectra, λ_max (ethanol) = 813 nm.

Synthesis of [Cu(H₂O)LBr₂]·0.14EtOH (2). A sample of L (0.076 g, 0.20 mmol) was dissolved in 4.0 mL of ethanol. The copper(II) bromide sample (0.044 g, 0.20 mmol) and 1 mL of distilled water were then added. After 20 minutes, a dark-yellow precipitate formed, which was filtered out, washed with ice-cold ethanol and distilled water, and dried in air. Single crystals of [CuLBr₂] (2a), suitable for single-crystal XRD, were obtained from the mother liquor after 10 days. Yield: 57% (0.072 g). Elemental analysis (%): Calc. for C_24.28H_19.84Br₂CuN₂O_2.14P: C 46.4; H 3.2; N 4.5. Found: C 46.1; H 3.4; N 4.4. IR-spectra (ν, cm⁻¹): 1123, 1138, 1169 ν(P=O); 1435, 1483, 1501, 1535, 1570, 1587, 1603, 1624 R_rings; 2964, 3007, 3038, 3074 ν(C–H); 3287, 3296, 3413 ν(O–H). TGA: 0.9% weight loss at 70 °C (calculated for 0.14C₂H₅OH – 1.0%), 3.7% weight loss at 235 °C (calculated for H₂O – 3.0%). Λ_m = 38 (ethanol) and 229 (water) Ω⁻¹ cm² mol⁻¹. UV-vis spectra, λ_max (ethanol) = 818 nm.

Synthesis of [Cu(bipy)(H₂O)L](NO₃)₂·2.8H₂O (3). A sample of 2,2′-bipyridine (0.032 g, 0.20 mmol) was dissolved in 3.5 mL of ethanol. A copper(II) nitrate trihydrate sample (0.049 g, 0.20 mmol) was then added and dissolved with stirring, forming a light blue solution. The sample of L (0.076 g, 0.20 mmol) was dissolved in 3 mL of ethanol with gentle heating (55 °C) and subsequently added to the mixture. One-third of the solution was evaporated under heating up to 70 °C. After one month, ethanol completely evaporated. The green precipitate was collected with an additional portion of ethanol (200 µL), washed with distilled water and dried in air. Single crystals of [Cu(bipy)(H₂O)L](NO₃)₂ (3a), suitable for single-crystal XRD, were obtained from the mother liquor after treatment with 2 ml of acetonitrile. Yield: 64% (0.10 g). Elemental analysis (%): Calc. for C₃₄H_32.6CuN₆O_10.8P: C 51.5; H 4.2; N 10.6. Found: C 51.1; H 4.3; N 10.2. IR-spectra (ν, cm⁻¹): 829, 864 ν(NO₃⁻); 1101, 1121, 1155, 1180 ν(P=O); 1329 ν(NO₃⁻); 1439, 1447, 1476, 1485, 1497, 1508, 1566, 1601, 1610, 1622, 1639 R_rings; 3065 ν(C–H); 3250, 3389 ν(O–H). TGA: 8.4% weight loss at 120 °C (calculated for 3.8H₂O – 8.6%). Λ_m = 61 (ethanol) and 235 (water) Ω⁻¹ cm² mol⁻¹. UV-vis spectra, λ_max (ethanol) = 719 nm.

Cell cultures

Lung adenocarcinoma A549 (BioloT Ltd, Russia), larynx carcinoma Hep2, hepatocellular carcinoma HepG2, breast carcinoma MCF-7 and non-tumour lung fibroblasts MRC5 (State Research Center of Virology and Biotechnology VECTOR, Novosibirsk, Russia) were seeded on 96-well plates and cultured in DMEM/F12 (A549; PanEco Ltd, Russia) or DMEM (Hep2, HepG2, MCF7, MRC5; PanEco Ltd, Russia) supplemented with 10% fetal bovine serum (BSA, HyClone) under standard conditions (humidified atmosphere, 5% CO₂ and 95% air, 37 °C).

For experiments on 2D models, cells were seeded into 96-well plates at a density of 5 × 10³ cells per well. Spheroids were used as a 3D cell culture model. For their preparation, Hep2 cells were seeded into 96-well U-shaped plates (83.3925.400, Sarstedt AG&CO. KG, Germany) at a density of 1.5 × 10³ cells per well.

Treatment of the cells with the compounds

The cells were treated with complexes 1–3 or ligand 24 hours after seeding on the plates. The solutions of the ligand and complexes were prepared as follows: (a) the compounds were dissolved in DMSO; (b) the obtained solutions were diluted with cell culture medium (final DMSO content was less than 1% (v/v)). For cytotoxicity studies, the incubation time was 48 hours, and for the assessment of reactive oxygen species generation it was 1 hour. Cisplatin dissolved in water and [Cu(bipy)Cl₂] and [Cu(phen)Cl₂] dissolved in DMSO were used for comparison purposes.

Cytotoxic activity

The cytotoxic effects of the compounds were tested by the Hoechst/propidium iodide (PI) double staining protocol.³⁹ For identification of live, apoptotic and dead cells, treated cells and control cells were stained with a mixture of fluorescent dyes Hoechst 33342 (Sigma-Aldrich, Switzerland) and propidium iodide (PI, Invitrogen, USA) for 30 min at 37 °C (2D cell culture). The cell viability and proliferation in a 3D cell model were evaluated as described earlier⁴⁰ with several changes. The spheroids were stained with a mixture of fluorescent dyes Hoechst 33342 and PI for 3 hours. The concentration range was selected based on the LC₅₀ values obtained from the 2D cell model.

Assessment of reactive oxygen species generation

Cellular ROS were determined by cell staining with 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFDA; Lumiprobe, Russia). After 1-hour incubation with compounds, the Hep2 cells were washed once with warm HEPES buffer solution (PanEco Ltd, Russia) and incubated with 10 µM H₂DCFDA and Hoechst 33342/PI for 30 min at 37 °C. Cells incubated with 100 µM H₂O₂ were used as a positive control in this experiment, and non-treated cells served as a negative control.

Image acquisition and analysis

An IN Cell Analyzer 2200 (GE Healthcare, UK) was used to perform automatic imaging in brightfield and fluorescence channels. To study cytotoxic activity and ROS generation in 2D cell cultures, images of four fields per well were acquired at 200× magnification. For cytotoxic activity in 3D cell cultures, z-stacks of spheroids were captured at 100× magnification (7–11 images separated by 15 µm, starting from the bottom).⁴⁰ The IN Cell Investigator (GE Healthcare, UK) software was used for the analysis of the images. Separate z-planes were segmented and analyzed as 2D images to count alive/dead cell nuclei, and objects displaced relative to one another in each plane were summed (maximum nuclei offset: 5–10 µm). Thus, nuclei were segmented and quantified within the multilayered three-dimensional volume, avoiding both missed objects and double counting of the same nuclei.

Data analysis

All data shown are the mean values of three wells. The LC₅₀ values were calculated from curves constructed by plotting cell survival (%) against drug concentration (µM) after the nonlinear approximation. The quantitative data were expressed as mean ± standard deviation (SD). All statistical analyses were performed using Microsoft Excel 2016 and OriginPro 8.0.

Conclusions

Hybrid phosphinyl-phenanthroline ligand L was used for the synthesis of three mononuclear copper(II) complexes: [CuLCl₂]·1.2H₂O (1), [Cu(H₂O)LBr₂]·0.14EtOH (2) and [Cu(bipy)(H₂O)L](NO₃)₂·2.8H₂O (3). A crystallographic study revealed that in the crystal structures the ligand L adopts a tridentate coordination mode; the metal centers display distorted square-pyramidal geometries in the case of the halide complexes, while an octahedral geometry is observed in the nitrate complex. According to the conductometry data, partial dissociation of one halide ion in ethanol solution occurs in the case of 1 and 2, while 3 fully dissociates into a dicationic species, [Cu(bipy)(H₂O)L]²⁺, and two nitrate counterions. In water, all three complexes act as 1 : 2 electrolytes. UV-Vis spectroscopy confirmed the stability of the cationic forms of the complexes formed in aqueous solution over a 48-hour period. The cytotoxicity of L and complexes 1–3 was studied in vitro in 2D (A549, Hep2, HepG2, MCF7, MRC5) and 3D (Hep2 spheroids) cell models. L demonstrated cytostatic activity in all tested cell lines in the 2D cell model, and was cytotoxic to non-tumor lung fibroblasts with an LC₅₀ value of 33 ± 7 µM. Compared to the ligand, copper(II) complexes 1–3 showed increased cytotoxicity (LC₅₀ = 6.5–58 µM) in all cell lines except A549 and no tumor selectivity (SI ≤ 1). The modification of 1,10-phenanthroline with a diphenylphosphinyl fragment reduces the cytotoxicity of the complexes by an order of magnitude, while providing no selectivity advantage over reference compound [Cu(phen)Cl₂] in the 2D cell model. However, the cytotoxicity of complexes 1 and 3 increased at least twofold in the 3D cell model, which more closely mimics the in vivo tissue architecture, making these complexes promising candidates for further investigation. Studies on potential mechanisms of action revealed no additional ROS generation upon incubation of the cells with the complexes, and the level of apoptosis generally didn’t increase significantly in tumor cells. Therefore, the observed cytotoxicity was likely due to alternative mechanisms of action.

Author contributions

P. E. Savinykh: investigation, formal analysis, writing – original draft, and visualization. Yu. A. Golubeva: investigation, formal analysis, validation, methodology, and writing – original draft. K. S. Smirnova: investigation, formal analysis, and visualization. L. S. Klyushova: investigation, formal analysis, and visualization. E. H. Sadykov: investigation and formal analysis. A. V. Artem’ev: investigation, writing – review and editing. E. V. Lider: conceptualization, supervision, and writing – review and editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: IR, UV-Vis spectra, TGA and X-ray diffraction data, and selected cytotoxicity data. See DOI: https://doi.org/10.1039/d5nj03235k.

CCDC 2477101–2477103 contain the supplementary crystallographic data for this paper.^41a–c

Acknowledgements

This work was supported by the Russian Science Foundation (Project No. 20-73-10207). The authors thank Anna P. Zubareva and Natalia N. Komardina for carrying out the elemental analysis, Aleksandra A. Shapovalova for IR spectroscopy data, M.O. Shin for the powder X-ray analysis data, and Elisaveta A. Sanzhenakova, Taisiya S. Sukhikh and Danil V. Kochelakov for providing single-crystal XRD data collected in NIIC SB RAS. The research (Nikolaev Institute of Inorganic Chemistry SB RAS) was supported by the Ministry of Science and Higher Education of the Russian Federation, project No. 125020401317-8. The investigation of biological activity was performed on the equipment of the Center for Collective Use “Proteomic Analysis” (FRC FTM) and supported by the Ministry of Science and Higher Education of the Russian Federation, No. 125031203556-7.

References

1. T. M. Dau, B. D. Asamoah, A. Belyaev, G. Chakkaradhari, P. Hirva, J. Jänis, E. V. Grachova, S. P. Tunik and I. O. Koshevoy, Dalton Trans., 2016, 45, 14160–14173.
2. D. C. Babbini and V. M. Iluc, Organometallics, 2015, 34, 3141–3151.
3. A. W. G. Platt, Coord. Chem. Rev., 2017, 340, 62–78.
4. N. E. Borisova, A. V. Kharcheva, S. V. Patsaeva, L. A. Korotkov, S. Bakaev, M. D. Reshetova, K. A. Lyssenko, E. V. Belova and B. F. Myasoedov, Dalton Trans., 2017, 46, 2238–2248.
5. S. Menzel, S. P. Höfert, S. Öztürk, A. Schmitz and C. Janiak, Z. Anorg. Allg. Chem., 2021, 647, 803–808.
6. G. A. Senchyk, A. B. Lysenko, H. Krautscheid, E. B. Rusanov, M. Karbowiak and K. V. Domasevitch, CrystEngComm, 2022, 24, 2241–2250.
7. P. Braunstein and A. A. Danopoulos, Chem. Rev., 2021, 121, 7346–7397.
8. A. S. V. Anand, S. Perinbanathan, I. Singh, R. R. Panicker, T. Boominathan, A. S. Gokul and A. Sivaramakrishna, ChemCatChem, 2024, 16, e202400844.
9. M.-Y. Heng, H.-L. Shao, J.-T. Sun, Q. Huang, S.-L. Shen, G.-Z. Yang, Y.-H. Xue and S.-N. Xiao, Rare Met., 2025, 44, 1108–1121.
10. Y. Liu, C. Z. Wang, Q. Y. Wu, J. H. Lan, Z. F. Chai, Q. Liu and W. Q. Shi, Inorg. Chem., 2020, 59, 11469–11480.
11. C. A. P. Goodwin, A. W. Schlimgen, T. E. Albrecht-Schönzart, E. R. Batista, A. J. Gaunt, M. T. Janicke, S. A. Kozimor, B. L. Scott, L. M. Stevens, F. D. White and P. Yang, Angew. Chem., Int. Ed., 2021, 60, 9459–9466.
12. X. P. Lei, Q. Y. Wu, C. Z. Wang, J. H. Lan, Z. F. Chai, C. M. Nie and W. Q. Shi, Inorg. Chem., 2023, 62, 2705–2714.
13. L. Xu, X. Yang, A. Zhang, C. Xu and C. Xiao, Coord. Chem. Rev., 2023, 496, 215404.
14. P. E. Savinykh, Y. A. Golubeva, K. S. Smirnova, L. S. Klyushova, A. S. Berezin and E. V. Lider, Polyhedron, 2024, 261, 117141.
15. P. E. Savinykh, Y. A. Golubeva, K. S. Smirnova, L. S. Klyushova, M. P. Davydova, A. V. Artemiev and E. V. Lider, J. Struct. Chem., 2024, 65, 1465–1476.
16. W. A. E. McBryde, A Critical Review of Equilibrium Data for Proton- and Metal Complexes of 1,10-Phenanthroline, 2,2′-Bipyridyl and Related Compounds: Critical Evaluation of Equilibrium Constants in Solution: Stability Constants of Metal Complexes, 2013.
17. A. C. Hachey, D. Havrylyuk and E. C. Glazer, Curr. Opin. Chem. Biol., 2021, 61, 191–202.
18. M. E. Bravo-Gómez, J. C. García-Ramos, I. Gracia-Mora and L. Ruiz-Azuara, J. Inorg. Biochem., 2009, 103, 299–309.
19. J. A. Eremina, E. A. Ermakova, K. S. Smirnova, L. S. Klyushova, A. S. Berezin, T. S. Sukhikh, A. A. Zubenko, L. N. Fetisov, K. N. Kononenko and E. V. Lider, Polyhedron, 2021, 206, 115352.
20. J. A. Eremina, E. V. Lider, T. S. Sukhikh, L. S. Klyushova, M. L. Perepechaeva, D. G. Sheven, A. S. Berezin, A. Y. Grishanova and V. I. Potkin, Inorg. Chim. Acta, 2020, 510, 119778.
21. Y. A. Golubeva, K. S. Smirnova, L. S. Klyushova, V. I. Potkin and E. V. Lider, Appl. Biochem. Biotechnol., 2022, 11, 531–539.
22. M. Reiher, G. Brehm and S. Schneider, J. Phys. Chem. A, 2004, 108, 734–742.
23. A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
24. W. J. Geary, Coord. Chem. Rev., 1971, 7, 81–122.
25. Y. A. Golubeva, E. A. Ermakova, K. S. Smirnova, L. S. Klyushova, N. P. Zaytsev, N. A. Osik, A. S. Berezin, A. S. Potapov and E. V. Lider, Inorg. Chim. Acta, 2025, 587, 122803.
26. J. Lee, G. D. Lilly, R. C. Doty, P. Podsiadlo and N. A. Kotov, Small, 2009, 5, 1213–1221.
27. V. Zingales, M. R. Esposito, M. Quagliata, E. Cimetta and M.-J. Ruiz, Foods, 2024, 13, 564.
28. M. Štampar, J. Tomc, M. Filipič and B. Žegura, Gen. Carc., 2019, 93, 3321–3333.
29. H. Jenkins, L. MacLean, S. McClean, G. Cooke, M. Devereux, O. Howe, M. D. Pereira, N. V. May, É. A. Enyedy and B. S. Creaven, J. Inorg. Biochem., 2023, 249, 112383.
30. S. Abdolmaleki, S. Khaksar, A. Aliabadi, A. Panjehpour, E. Motieiyan, D. Marabello, M. H. Faraji and M. Beihaghi, Toxicology, 2023, 492, 153516.
31. B. Rogalewicz and A. Czylkowska, Eur. J. Med. Chem., 2025, 292, 117702.
32. Bruker Apex3 software suite: Apex3, SADABS-2016/2 and SAINT, version 2018.7-2; Bruker AXS Inc.: Madison, WI, 2017.
33. CrysAlisPro 1.171.38.46. Rigaku Oxford Diffraction: The Woodlands, TX, USA, 2015.
34. G. M. Sheldrick, Acta Crystallogr A, 2015, 71, 3–8.
35. G. M. Sheldrick, Acta Cryst., 2015, C71, 3–8.
36. O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
37. A. V. Artem’ev, M. P. Davydova, L. S. Klyushova, E. H. Sadykov, M. I. Rakhmanova and T. S. Sukhikh, Dalton Trans., 2024, 53, 18027–18036.
38. G. Zakirova, D. Mladentsev and N. Borisova, Synthesis, 2019, 2379–2386.
39. J. A. Eremina, E. V. Lider, N. V. Kuratieva, D. G. Samsonenko, L. S. Klyushova, D. G. Sheven, R. E. Trifonov and V. A. Ostrovskii, Inorg. Chim Acta, 2021, 516, 120169.
40. O. Sirenko, T. Mitlo, J. Hesley, S. Luke, W. Owens and E. F. Cromwell, Assay Drug Dev. Technol., 2015, 13, 402–414.
41. (a) CCDC 2477101: Experimental Crystal Structure Determination, 2025 DOI:10.5517/ccdc.csd.cc2p4mhz; (b) CCDC 2477102: Experimental Crystal Structure Determination, 2025 DOI:10.5517/ccdc.csd.cc2p4mj0; (c) CCDC 2477103: Experimental Crystal Structure Determination, 2025 DOI:10.5517/ccdc.csd.cc2p4mk1.

---