Crystal and molecular structure of [PdCl(TOTP)(MTUC)] (1): ORTEP diagram of 1 along with selected bond distances and angles are shown in Figure 1.

The geometry around the Pd(II) is square planar. The MTUC chelates the metal center through its N,S-donor atoms forming a high-energy, four-membered ring. The distribution of bond lengths in 1 indicates the bond types depicted in Scheme 2. Specifically, the measured distance of the C23-O1 bond of 1.241 Å is notably closer to the reported length of 1.23 Å for a C=O double bond, as opposed to the C-O single bond length of 1.43 Å. The measured bond distances of C23-N2 at 1.402 Å, and C23-C24 at 1.414 Å indicate single rather than double bonds (typical lengths for single C-N and C-C bonds are 1.37 Å and 1.53 Å, respectively) (Scheme 2). The bond lengths C22-N2 at 1.350 Å, and N1-C25 at 1.369 Å tend towards the single C-N bond, while the C22-N1 at 1.316 Å is close to double C=N bond (typical lengths for double C=N is 1.27 Å). The C25-C24 bond length at 1.351 Å implies a double C=C bond (as depicted in Scheme 2). The measured C22-S1 bond length of 1.725 Å falls in the range of the C-S single bond (the typical length for a single C-S bond is 1.76 Å). Based on the bond length distribution, it is deduced that mono-deprotonation occurs at the N-C-S thioamide group of the MTUC upon its coordination with the Pd(II) metal center, as depicted in Scheme 2.

One P atom from the TOPT ligand, one Cl atom, and the N,S ones from MTUC form a square planar arrangement around the Pd(II) ion. The bond angles P-Pd-X (X= Cl and S) are close to the ideal value of 90° (Cl1-Pd1-P1= 90.22(6) and S1-Pd1-P1= 97.54(6))°, while the S1-Pd1-N1 varied significantly from the ideal 90° of a square planar geometry (S1-Pd1-N1= 69.69(15)°). The bond angles Cl1-Pd1-S1 (168.72(7)°) and P1-Pd1-N1 (167.11(15)°) lie away from the ideal value of 180°.

The Pd1-P1 bond length (2.2416(14) Å) is close to the corresponding distance found in previously reported compound with TPP of formula [PdCl₂(TPP)(MTUC)] (2.2363(8) Å) (Banti et al., 2015) and [PdCl₂(TPP)(TU)] (TU= thiourea) 2.2440(1) Å (Moro et al., 2006). The Pd1-S1 bond in 1, is 2.3188(17) Å and is similar with the corresponding found in [PdCl₂(TPP)(MTUC)] (2.3197(9) Å)) and [PdCl₂(TPP)(TU)] (2.319(1) Å) (Moro et al., 2006). The length of the Pd-Cl bond (2.3170(16) Å) is in accordance with the corresponding ones, observed in [PdCl₂(TPP)(TU)] (Pd–Cl(l) 2.311(l), Pd–Cl(2) 2.401(l) Å) (Moro et al., 2006). Moreover, the Pd-Cl bond distance of 1 is close to the corresponding one in [PdCl₂(TPP)(MTUC)] ((Pd1–Cl1 = 2.3892(9), Pd1-Cl2 = 2.2892(9) Å) (Banti et al., 2015).

A robust intermolecular hydrogen bond O1···H[N2] at a distance of 2.752(7) Å connects two monomeric units of [PdCl(TOTP)(MTUC)], resulting in the formation of a dimer (depicted in Figure 1B). These robust hydrogen bonds could potentially facilitate a strong interaction between DNA and compound 1.

Vibrational spectroscopy: The vibrational band at 1,620 cm⁻¹ in the ATR-FTIR spectrum of MTUC is assigned to the v(C=O) bond vibration which is shifted at 1,651 cm⁻¹ in the spectrum of 1. This validates the increase in the C=O bond order as shown by the bond distribution outlined in Scheme 2. The bands at 1,552 and 1,240 cm⁻¹ in the ATR-FTIR spectrum of MTUC are assigned to the thioamide I (δ(N–H) + ν(C-N)) and II (ν(C–N) + δ(N–H) + ν(C=S)) bands (Aulakh et al., 2019), which mainly consisted as ν(ring), bond vibrations (Grosmaire and Delarbre, 2012). The bands at 990 and 654 cm⁻¹ in the IR spectrum of MTUC are attributed to the ν(ring), ν(CS) vibrations (thioamide III (ν(C–N) + ν(C–S)) and IV (ν(C–S) bands) (Supplementary Figure S1) (Grosmaire and Delarbre, 2012; Charalampou et al., 2014; Banti et al., 2015). These vibration bands are shifted at 1,517, 1,238, 1,000, and 662 cm⁻¹, respectively. The significant shift of the thioamide I vibrational band from 1,552 to 1,517 cm⁻¹ is likely due to the elongation of the C=N upon coordination of MTUC to Pd(II) (Supplementary Figure S1). Both C=N and N-H contribute to the thioamide I band mentioned earlier. This stands as the most compelling evidence of MTUC coordination to the Pd(II) apparent in the FTIR spectrum, as the other shifts observed are comparatively minor. The upward shift of the thioamide band III to 1,000 cm⁻¹ from 990 cm⁻¹ is likely due to the elongation of the C-S bond resulting from the coordination of MTUC toward Pd(II). The bands at 555-514 cm⁻¹ in the spectrum of free TOTP are attributed to the v(C−P) vibrations which are shifted at 563-536 cm⁻¹ (Supplementary Figure S1).

¹H Nuclear Magnetic Resonance (¹H NMR) Spectroscopy: The broad resonance signal in the ¹H-NMR spectrum of 1 at 12.85 ppm is attributed to the H(N¹) and H(N³) protons of the MTUC. The corresponding signal is observed at 12.23 ppm in the free ligand (Scheme 1; Supplementary Figure S2). The single resonance signal at 5.65 ppm is attributed to the H(C⁵) of MTUC and it is shifted at 5.72 ppm in 1. The singlet signal at 2.03 ppm is assigned to the H(CH₃-) of the methyl-substituent of MTUC. This signal is shifted at 1.64 ppm in 1 (Charalampou et al., 2014; Banti et al., 2015). The resonance signals at 7.54–6.92 and 2.22 ppm in the spectra of 1 are attributed to Haromatic and H(CH₃) of TOPT, respectively. The corresponding signals of the free TOPT are observed at 7.28–6.55 and 2.27 ppm, respectively (Banti et al., 2016) (Supplementary Figure S2).

Stability studies: The stability of 1 was verified by UV and ¹H-NMR spectroscopies for a period of 48 h (Supplementary Figures S3, S4), which corresponds to the period of incubation of the cells with the compound. No changes were observed between the initial spectra and the corresponding ones after 48 h, confirming the retention of its structure in solution.

In vitro antiproliferative activity: The in vitro antiproliferative activity of 1 was evaluated against two human adenocarcinoma breast cell lines, MCF-7 (hormone depended (HD)) and MDA-MB-231 (hormone independent (HI)) by sulforhodamine B (SRB) assay after its incubation for 48 h. The IC₅₀ values of 1 against MCF-7 and MDA-MB 231 cells are 39.1 ± 1.3 and 57.1 ± 3.4 μΜ, respectively (Table 1). The corresponding IC₅₀ value exhibited by cisplatin against MCF-7 and MDA-MB 231 cells is 5.5 ± 0.4 and 26.7 ± 1.1 μΜ, respectively, suggesting lower antiproliferative activity of 1. The low activity of 1 follows the corresponding one of cis-Cl-[PdCl₂(TPP)(MTUC)]·3(H₂O), which shows no antiproliferative activity up to the concentration of 30 μΜ tested (Banti et al., 2015).

The heteroleptic palladium(II) complexes of 2-thiouracil and TPP, [PdCl(TPP)(μ-Ν,S-TUC)PdCl(TPP)₂] and [PdCl(TPP)(Ν,S-TUC)] (Shaheen et al., 2007) were assessed across a selection of seven human tumor cell lines: MCF-7 and EVSA-T (two breast cancers), WiDr (colon cancer), IGROV (ovarian cancer), M19 MEL (melanoma), A248 (renal cancer), and H226 (non-small cell lung cancer). Their IC₅₀ values spanned a range from 0.4 to 25.1 μΜ. Particularly, the IC₅₀ values of [PdCl(TPP)(μ-Ν,S-TUC)PdCl(TPP)₂] and [PdCl(TPP)(Ν,S-TUC)] against MCF-7 cells, are 21.0 and 4.6 μΜ (Shaheen et al., 2007) (Table 1). Moreover, the IC₅₀ value of the silver(I) complex of 2-thiouracil (TUC) [Ag(TUC)(TPP)₂](NO₃)·H₂O) against human osteosarcoma cells (MG63) is 16.6 μM (Aulakh et al., 2019) (Table 1), while the corresponding one of copper(I) halide complexes with 5-carbethoxy-2-thiouracil (eitotH₂), [CuX(eitotH₂)₂]₂ and [CuX(TPP)₂(eitotH₂)] (X= Cl, Br, I) against A549 (human pulmonary carcinoma cell line), HeLa (human epithelial carcinoma cell line) and MRC5 cells range between 3.4 and 118 μΜ (Papazoglou et al., 2014) (Table 1). The copper compounds of TPP show higher antiproliferative activity than those without TPP. Generally, the palladium(II) compounds with MTUC or TUC exhibit antiproliferative activity in higher concentrations than those of silver or copper complexes.

The in vitro toxicity of 1 was also estimated against normal human fetal lung fibroblast cells (MRC-5). Its IC₅₀ value is 38.1 ± 2.9 μΜ (Table 1). This led to a Therapeutic Potency Index (TPI= (IC₅₀ agent against non-cancerous cells)/(IC₅₀ against cancerous cells)) of 0.98, against MCF-7 (HD) and 0.50 against MDA-MB-231(HI), respectively. The TPI values of cisplatin against MCF-7 (HD) and MDA-MB-231(HI) are 0.20 and 0.04. Therefore, 1 exhibits higher selectivity for cancer cells than for normal ones, while it is less toxic, towards MRC-5 cells, than cisplatin by 35-fold.

In vitro Mechanism of Action. The in vitro mechanism of action for 1 is elucidated through (i) Pd(II) localization (ii) observation of MCF-7 cell morphology, (iii) staining with acridine orange/ethidium bromide (AO/EB), and (iv) conducting a mitochondrial membrane permeabilization test.

[i] Pd(II) localization: In an attempt to ascertain if the palladium metallodrug penetrates the cytoplasm, we conducted X-ray fluorescence spectroscopy (XRF) measurements. The XRF spectrum of 1 confirms the presence of Pd in the cytoplasm (Supplementary Figure S5). The outcomes indicated that, following incubation of MCF-7 cells with the metallodrug at a concentration of 4.13 ppm, the cytoplasm contained approximately 3.2 ± 0.4 ppm of palladium. Before quantifying the palladium, the cells underwent a triple wash with PBS to remove the medium, followed by centrifugation each time. At present we lack the tools, such as luminescence tracers or confocal microscopy, to precisely track the Pd compound’s localization to demonstrate its selective targeting of DNA. Furthermore, the Pd(II) metallodrug, when excited at λ_exc= 367 nm, emits no radiation between 380 and 700 nm hindering its identification within DNA via its own fluorescence (Supplementary Figure S6).

Upon treatment of MCF-7 cells by 1, a significant increase in apoptosis was observed. The percentage of apoptotic cells is (58.0 ± 7.9) % in contrast to the untreated cells (17.1 ± 1.0) %. No necrotic cells were observed (Afsar et al., 2016). Therefore, it is concluded that the antitumor activity of 1 was triggered due to apoptosis instead of the necrosis pathway.

[iv] Loss of the mitochondrial membrane permeabilization (MPP): To investigate whether the metallodrug 1 interferes with mitochondrion, the loss of mitochondrial membrane permeabilization (MMP), which is an indicator of mitochondrial dysfunction, was measured. The fluorescence of the cationic mitochondriotropic dye, which bioaccumulates in the membrane of the normal mitochondrion, and quenchs when MMP collapses with a consequent release of cytochrome c in the cytosol activating the apoptotic pathway.

The % quenching of the fluorescence emitted when MCF-7 cells are treated with 1 is 18.7%. Upon treatment of the cells with [PdCl₂(TPP)(MTUC)] the fluorescence emitted quench by only 8.2%. The fluorescence quenching caused by cisplatin, rises up to 54.9% (Chrysouli et al., 2018). These observations provide additional validation for the previously inferred apoptosis detected through cell morphology and the acridine/ethidium assay. Certainly, the quenching in fluorescence emitted from the dye, which is bonded to the mitochondrial membrane, upon cell exposure to the Pd(II) metallodrug is not evidence of a direct interaction of the drug with the mitochondrial membrane. The loss in mitochondrial membrane permeabilization might arise from alternative pathways. However, it is crucial to demonstrate that the Pd(II) metallodrug either triggers these pathways or directly engages with them, ultimately leading to the loss of mitochondrial membrane permeabilization.

DNA Binding Studies: DNA has been extensively studied as a potential pharmacological target of metalotherapeutics. The DNA binding affinity towards 1 was investigated (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023). The DNA-1 complex formation is studied here by (i) viscosity measurements, (ii) absorption titration, (iii) fluorescence spectroscopy and (iv) by the DNA Thermal Denaturation (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023).

Viscosity Studies: The preliminary understanding of the DNA interaction mode with 1 was initially derived from viscosity measurements. The viscosity of a DNA solution is highly responsive to alterations in DNA length, serving as a pivotal indicator of the DNA binding mode in solution, notably influenced by the formation of the DNA-1 complex (Kapetana et al., 2022; Banti et al., 2023). In the case of intercalation interaction, the DNA-agent solution experiences a notable rise in relative viscosity, attributable to the unwinding and elongation of the double helix (Kapetana et al., 2022; Banti et al., 2023). On the other hand, when groove binding or electrostatic interactions occur, there is an insignificant impact on DNA length, resulting in no significant change in viscosity (Kapetana et al., 2022; Banti et al., 2023). Moreover, when an agent induces DNA strand cleavage, the DNA length decreases, consequently leading to a significant decrease in viscosity (Kapetana et al., 2022; Banti et al., 2023). Lastly, if a covalent bond forms between an agent and DNA, such as in the case of cisplatin-DNA, the solution viscosity decreases due to kinking in the DNA backbone, shortening the axis length of the helix (Kapetana et al., 2022; Banti et al., 2023).

The values of relative specific viscosity (n/nο)^1/3 were plotted vs. the values r= [complex]/[DNA] (Figure 4). The increasing amounts (up to r= 0.37) of 1 in the DNA solution increase its relative viscosity slightly. This indicates groove binding interaction. Intercalation mode of DNA interaction with a binder, on the other hand, would significantly increase the viscosity of the solution (Figure 4), e.g., in the case of the DNA-ethidium bromide (EB) interaction (Figure 4). The minor groove binding agent Hoechst 33258 leads to a moderate decrease of the DNA solution viscosity similar to the one observed in the case of DNA-1 complex (Figure 4).

UV-Vis Spectroscopic study: Electronic absorption spectroscopy is used in the study of the binding properties (covalent, non-covalent, and electrostatic or groove binding mode) of metallodrugs with DNA.

The absorbance of CT-DNA (λ_max= 258 nm) was recorded upon the presence and absence of 1 at various r values (r = [agent]/[DNA]) at a constant [DNA] (Supplementary Figure S7). A slight hyperchromic effect was detected (3.4%), indicative of either slight groove binding or electrostatic mode (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023). The intrinsic binding constant (K _b ) of 1 was calculated (1.1 ± 3.3)×10⁴ M^-1 (Supplementary Figure S8), while the corresponding metallodrug of TPP (cis-Cl-[PdCl₂(TPP)(MTUC)]·3(H₂O)) shows similar K _b value (4.5 ± 1.5)×10⁴ M^-1, with slight also hyperchromicity effect at 258 nm (Banti et al., 2015). However, since the hydrolysis rate for palladium is faster than that of platinum complexes (10⁵ times faster according to instrumental methods, and 10⁶ times estimated by ab initio studies (Largy et al., 2011)), this leads to instability of Pd-DNA complexes relative to their Pt counterparts. This results in the limited use of Pd derivatives for DNA targeting and for biomedical applications (Largy et al., 2011). Generally, the absorption spectra indicated that the complex interacts with DNA weakly, and therefore a groove binding is concluded in accordance with the viscosity measurements (see above).

Fluorescence Spectroscopic Studies: The binding properties of 1 towards CT-DNA were further evaluated by fluorescence spectroscopic studies. Intercalative or minor groove binding mode can be detected by the quenching of fluorescence emitted from the CT-DNA/EB complex when the EB is substituted by a DNA binder (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023). The recorded emission spectra of the CT-DNA-EB complex upon excitation at λ_exc = 527 nm in the absence and presence of 1 are shown in Figure 5. The intensity of the radiation emitted at 588 nm forms the solution of the CT-DNA-EB complex undergoes quenching by 82%, by increasing the concentrations of 1, as compared to the initial fluorescence intensity of the solutions EB-DNA. The Pd(II) metallodrug does not emit at λ_em= 588 nm when excited at λ_exc= 527 nm, having no impact on the corresponding emission of EB-DNA (Supplementary Figure S9). The value of the apparent binding constant (K _app ) was calculated and the concentration of the drug at 50% reduction of the fluorescence is derived from the diagram of (I _x /I ₀) (I₀ and I_x are the emission intensities of the CT-DNA-EB in the absence and presence of 1, respectively) vs. the concentration of 1 (Figure 5) (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023). The K _app value is (29.8 ± 6.0)×10⁴ M^-1) suggesting minor groove binding. The value of K _app lies in the range 10⁴-10⁵ M^-1 implies groove binding mode (Sheng et al., 2007; Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023). For comparison the corresponding K _app of EB is 10⁷ M^-1, indicating that 1 binds through minor groove mode (Sheng et al., 2007; Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023).

DNA Thermal Denaturation studies: Thermal denaturation studies of double-helical DNA to single-strands can be detected by a “hyperchromic effect” at 258 nm in the UV spectrum of the DNA with or without binder agents. A “sigmoid” melting transition curve is produced, simultaneously (Banti et al., 2024). As the temperature of DNA rises, the stacking interactions among bases and the hydrogen bonding between the bases in the double-stranded DNA gradually become disrupted (Banti et al., 2024). The DNA melting temperature (T _m ) denotes the temperature at which half of the double-stranded DNA separates into two single strands, as the molar extinction coefficient of single-stranded DNA significantly exceeds that of double-stranded DNA (Banti et al., 2024).

The T _m of unbound DNA was found to be 59.5 ± 1.0°C (Figure 6). Under the same experimental conditions, the observed T _m of the DNA-1 complex is 60.2 ± 0.5°C. The effect of 1 on the melting profile of CT-DNA was smaller than that produced by the well-known intercalator, ethidium bromide (EB), causing a considerable increase of T _m of DNA (ΔT _m of 13°C) (Cory et al., 1985; Kelly et al., 1985; Banti et al., 2024)). In the case of a groove binding or electrostatic interaction mode, the ΔT _m value is <2°C (Cory et al., 1985; Kelly et al., 1985; Banti et al., 2024). Since the ΔT _m value of the DNA-1 complex is lower than 2°C, then a groove or electrostatic interaction mode can be concluded.

Materials and Instruments: Reagent-grade solvents, such as 6-methyl-2-thiouacil (Aldrich) and tris(o-tolyl)phosphine (Sigma–Aldrich, Merck), were employed without additional purification. These included dimethyl sulfoxide, methanol, and acetonitrile from Riedel-de Haën, toluene from Sigma Aldrich-Merck, and dichloromethane from Panreac. Calf thymus (CT)-DNA and ethidium bromide (EB) were procured from Sigma-Aldrich. The MCF-7, MDA-MB-231, and MRC-5 cell lines used in this study were obtained from ATCC (American Type Culture Collection).

Melting point measurements were performed in open tubes using a Stuart Scientific apparatus and remain uncorrected. Elemental analyses for carbon and hydrogen were conducted using the Carlo Erba EA MODEL 1108 elemental analyzer. The Cary 630 ATR-FTIR spectrometer by Agilent Technologies was employed to record infrared spectra ranging from 4,000 to 400 cm⁻¹. The spectrometer’s resolution, ranging from 2–16 kaiser, was set at 4 cm⁻¹ for the experiments. This specific instrument configuration features a KBr beam splitter, incorporates a diamond crystal, and utilizes a DTGS (Deuterated Triglycine Sulfate) detector. The ¹H NMR spectra were acquired using a Bruker AC 250 MHz FT-NMR instrument in DMSO-d ₆ solution, with chemical shifts (δ) reported in parts per million (ppm) using (CH₃)₄Si as an internal reference. Electronic absorption spectra were recorded at 450 to 190 nm in DMSO and Buffer solution using a VWR UV-1600 PC series spectrophotometer. Fluorescence spectra were captured at 700 to 380 nm in DMSO and Buffer solution irradiating with λ_ex= 367 or 527 nm utilizing a Jasco FP-8200 Fluorescence Spectrometer. XRF measurements were carried out employing a Rigaku NEX QC EDXRF analyzer situated in Austin, TX, USA using PdCl₂ for the standard curve.

Synthesis and crystallization of 1 : For the preparation of complex 1, 0.50 mmol of PdCl₂ (0.09 g), 0.50 mmol of tris(o-tolyl)phosphine (0.152 g) and 0.50 mmol of 6-methyl-2-thiouacil (MTUC) (0.07 g) were suspended to 20 cm³ methanol/acetonitrile solution (1:1) under continues stirring and reflux for 3 h. The solution was then filtered off and the filtrate was kept in ambient condition. Crystals suitable for X-ray analysis were grown from the mother solution after a few days.

1: Orange crystals; Elemental analysis found: C: 53.75; H: 4.33; N: 4.57, P: 5.49; Pd: 18.23; S:5.18%; calculated for C₂₆H₂₆ClN₂OPPdS: C: 53.71; H: 4.46; N: 4.77, P: 5.27; Pd: 18.11; S:5.46%; IR (cm^-1): 1649vs, 1518m, 1449m, 1399m, 1363s, 1280m, 1239s, 1192s, 1131m, 1044m, 1001m, 956m, 833s, 803s, 754vs, 712m, 678m, 662m, 596m, 565s, 537s, 463vs; ¹H-NMR (ppm) in DMSO-d ₆: 12.85 (s, H(N¹ and N³) of MTUC), 7.54–6.92 (m, H^aromatic and H(CH₃) of TOPT), 5.72 (s, H(C³) of MTUC), 1.64 (w, H(CH₃-) of MTUC); UV-vis (DMSO): λ=296 nm (logε= 4.23) and λ=367 nm (logε= 3.55).

X-ray structure determination: Intensity data for the crystals of 1 were collected using an Oxford Diffraction CCD instrument, utilizing graphite-monochromated CuKa radiation (λ= 1.54184 Å). Cell parameters were determined via least-squares refinement of diffraction data extracted from 25 reflections. All data underwent correction for Lorentz-polarization effects and absorption (Oxford Diffraction Ltd, 2006). The structural elucidation involved utilizing direct methods with SHELXS97 (Sheldrick, 1990) and subsequent refinement through full-matrix least-squares procedures on F2 with SHELXL97 (Sheldrick, 1997). Anisotropic refinement was applied to all non-hydrogen atoms, while hydrogen atoms were positioned at calculated sites and refined using the “riding model,” with isotropic thermal parameters set at 1.2 (1.3 for CH₃ groups) times the Ueq value of the relevant carrier atom.

1: C₂₆H₂₆ClN₂OPPdS, MW= 587.37 g/mol, monoclinic, space group P21/n, a= 10.3919(3) Å, b= 22.4640(7) Å, c= 11.5973(3) Å, β= 94.880(2), V= 2,697.50(13) ų, Z= 4, T= 100 K, Radiation CuKα, ρ(calc)= 1.446 g cm⁻³, μ= 7.903 mm⁻¹, F(000)= 1,192. 10776 reflections measured, 5132 unique (Rint= 0.082), 4,219 with I > 2s(I). The final R1= 0.0539 (for 4,219 reflections with I> 2s(I)) and wR2(F2)= 0.1887 (all data), S= 1.19.

Crystallographic data (excluding structure factors) for the structure reported in this paper has been deposited with the Cambridge Crystallographic Data Centre as supplementary publication no. CCDC-2303444 (1). Copies of the data can be obtained free of charge by application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: (+44) 1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).

All biological tests were conducted following established protocols detailed in prior literature (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023; Banti et al., 2024). The preparation of solutions adhered to the procedures outlined in the respective references (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023; Banti et al., 2024). The SRB assay, cell morphology observations, and DNA binding studies using UV-vis and fluorescence spectroscopies were carried out according to previously established methodologies (Banti et al., 2015; Banti et al., 2016; Chrysouli et al., 2018; Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023; Banti et al., 2024). AO/EB Staining and the Permeabilization of the mitochondrial membrane test were performed following documented procedures (Stathopoulou et al., 2021; Kapetana et al., 2022; Banti et al., 2023; Banti et al., 2024). Viscosity and DNA Thermal Denaturation studies were conducted in line with previously published methods (Cory et al., 1985; Kelly et al., 1985; Banti et al., 2024).

Each biological experiment underwent at least three independent replications.