Synthesis, Characterization, In Vitro and In Silico Investigation of Azo‐Based Co(III), Zn(II), and Pd(II) Complexes Against Human Leukemia

ABSTRACT

The new azo ligand (E)‐2‐((1H‐indol‐3‐yl)diazenyl)benzoic acid (IDB) was synthesized and complexes of Co(III), Zn(II), and Pd(II) were prepared. Elemental analysis, FT‑IR, ¹H‑NMR, ^{1 3}C‑NMR, UV–vis, mass spectrometry, molar conductivity, magnetic susceptibility, XRD and FE‐SEM were used to fully characterize the compounds. The spectroscopic data are consistent with the Co(III) and Zn(II) being coordinated as bidentate N,O‐donors, with the azo nitrogen and deprotonated carboxylate oxygen, and with the Pd(II) being square‐planar. The Pd(II) complex demonstrated cytotoxic activity against HL‑60 cells (IC₅₀ = 134.4 µg/mL) with modest selectivity over normal HdFn fibroblasts (IC₅₀ = 164.7 µg/mL, selectivity index = 1.23) among the tested complexes. Mechanistically, it significantly decreased malondialdehyde (MDA) levels in both cancer and normal cells, suggesting modulation of oxidative stress. The Pd(II) complex showed high binding affinity with Aurora kinase (Abl T315I mutant, PDB: 2V7A) with hydrogen bond with ASP 381 and ASN 322 and π–H interaction with LEU 248. The complexes are predicted to have reduced oral bioavailability, but are still good candidates for parenteral use, while the ligand is predicted to have good drug‑like properties. Overall, these results indicate that the Pd(II) complex is a promising lead compound for targeted leukemia therapy, but additional in vivo testing is needed.

A novel azo ligand (IDB) and its Co(III), Zn(II), and Pd(II) complexes were synthesized and characterized. The Pd(II) complex demonstrated potent activity against HL‐60 leukemia cells, modulated oxidative stress markers (MDA, CAT), and showed strong binding affinity (−6.6367 kcal/mol) to the Aurora kinase enzyme in molecular docking studies.

1. Introduction

Heterocyclic compounds are one of the most important and varied groups of organic molecules, which consist of cyclic structures containing at least one heteroatom (C, N, O, S, P, or B) within the ring system. The most usual heteroatoms are nitrogen (N), oxygen (O), and sulfur (S), but other elements can be used, such as phosphorus (P) or selenium (Se). These compounds were all found in nature and have a significant place in living organisms—over half of all known organic compounds are heterocyclic. Heterocyclic compounds are not only of academic interest but also are the backbone of many pharmaceuticals, agrochemicals and veterinary products. Nitrogen heterocycles are of great importance in modern medicine, as seen in the presence of nitrogen heterocycles in about 59% of FDA‐approved drugs. They exhibit a wide range of biological activities such as antimicrobial, anti‐cancer, anti‐inflammatory, antioxidant, and antifungal properties. The presence of heteroatoms facilitates binding through intermolecular interactions with pharmacological targets and receptors, and therefore allows these molecules to be a hydrogen bond donor and acceptor for better binding to the receptor sites [1, 2, 3, 4].

Indole is one of the most privileged and versatile scaffolds in medicinal chemistry, a bicyclic structure comprising a combination of a six‐membered benzene ring with a five‐membered pyrrole ring. Indole has a molecular formula of C₈H₇N and has an aromatic structure with ten π‐electrons, making the molecule slightly basic as a result of its unique electronic structure. It is interesting to note the chemical reactivity of indole, especially electrophilic substitution that takes place at the C‐3 position, which is about 10^{1 3} times more reactive than benzene. The high reactivity at position 3 is explained by the enamine‐like character of the part of the molecule which is not attached to the benzene ring. Unlike the normal amines, the lone pair electrons on the N in indole are delocalized into the aromatic system, which makes indole less basic [5, 6]. Indole derivatives are found in abundance in nature as crucial parts of many biologically active compounds. The indole nucleus occurs in the amino acid tryptophan, in the neurotransmitter serotonin, in the hormone melatonin and in various plant hormones like indole‐3‐acetic acid (IAA). This is a natural phenomenon found throughout the plant kingdom, and underlines the fundamental significance of the indole scaffold in biological systems and as a potential pharmaceutical target [7].

Indole derivatives have generated a lot of interest because of their antioxidant activity, which could have therapeutic uses in preventing oxidative stress‐related diseases. These compounds exhibit antioxidant activity in several ways, such as by single‐electron transfer (SET), hydrogen atom transfer (HAT), and metal chelating activities. The antioxidant activity of indole derivatives has been mostly related to the specific electronic properties of the indole ring system. The heterocyclic nitrogen atom acts as an active redox center because of the presence of a free electron pair, and the migration of the electron pair in the aromatic system is an important factor in the compounds' antioxidant activity. For the most effective antioxidant activity, the molecule must have an unsubstituted indole nitrogen atom and must be able to develop resonance‐stabilized indolyl radicals. Studies have shown that the mechanism of DPPH radical scavenging activity of indole derivatives is the hydrogen or electron abstraction from the indole ring. In the SET mechanism, an electron is removed from the N atom, forming a cation radical, whereas the HAT mechanism occurs via a hydrogen atom transfer from the N─H group to produce a resonance stabilized indolyl radical [8].

Indole derivatives have appeared as potential anticancer agents with considerable therapeutic value against various cancer types by different mechanisms of action. These compounds have demonstrated anticancer activity through several mechanisms such as cell cycle arrest, induction of apoptosis, inhibition of tubulin polymerization, and regulation of important signaling pathways. Indole derivatives appear to be especially effective against leukemia and a few compounds have been selectively active against leukemic cell lines [9, 10, 11]. In recent years, it has been shown that azo dye derivatives can easily form complexes with transition metal ions like Mn(II), Fe(III), Co(II), Ni(II), Cu(II), Zn(II), and Cd(II) with well‐defined molecular structure and interesting biological activity. Moreover, the coordinated compounds showed better antimicrobial activity against some pathogenic microorganisms, along with better molecular docking interactions and DNA binding affinity, which makes them good biologically active compounds [12]. Besides the traditional coordination compounds, nanosized metal complexes have been the focus of recent interest due to their special structural and functional properties. The Cu(II) complexes of azo ligands have shown outstanding antimicrobial and anticancer properties, and efficient catalytic activity in the oxidative dimerization of 2‐aminophenol. The results showed that the surface reactivity and catalytic activity of azo metal complexes can be significantly enhanced by decreasing the particle size to the nanometer scale without compromising the structural integrity of the complexes [13, 14]. Likewise, metal complexes of nanosynthesized sulfafurazole azo dye have been reported to exhibit high structural stability in the solid state and high antimicrobial and anticancer activity. The enhanced biological activity has been attributed to the synergistic effect of metal ion coordination and nanoscale dimensions, which enable better interactions with biological systems and enhance the therapeutic potential of these compounds [15]. All these studies have shown that the azo dye‐derived metal complexes are a promising class of multifunctional coordination compounds which have great potential in medicinal chemistry, antimicrobial therapy, anticancer research, and bioinorganic chemistry.

Indole derivatives and azo metal complexes have shown potential as anticancer agents, but there are significant gaps in knowledge. The incorporation of indole‐based azo ligands with various metal ions has not been studied in detail, and most of the previous studies have focused on structural characterization and not on detailed biological evaluation, especially modulation of oxidative stress biomarkers. Moreover, there are few systematic comparisons of Co(III), Zn(II), and Pd(II) complexes in the same ligand framework, and molecular docking has not been applied to clinically relevant therapeutic targets related to drug resistance. To overcome these drawbacks, the present study presents a detailed physicochemical characterization of structurally related Co(III), Zn(II), and Pd(II) complexes, along with in vitro cytotoxicity, oxidative stress evaluation, and molecular docking studies with a clinically relevant target, providing a more comprehensive understanding of the structure–activity relationships and therapeutic potential of these complexes.

2. Results and Discussion

2.1. Structure Confirmation of Ligand and Metal Complexes

The synthesized compounds were the (E)‐2‐((1H‐indol‐3‐yl)diazenyl)benzoic acid (IDB) ligand, its Co(III), Zn(II), and Pd(II) complexes. Elemental analysis (CHN), UV–vis spectroscopy, FT‐IR, mass spectrometry, and ¹H & ¹³C NMR spectroscopy were employed to characterize the ligand and its metal complexes. In addition, the surface morphology and crystalline structure of the nanoparticles were also investigated by field‐emission scanning electron microscopy (FESEM) and x‐ray diffraction (XRD). Various other physicochemical characteristics like melting point, molar conductivity and magnetic susceptibility were also determined. The ligand and its respective metal complexes were isolated in pure form from ethanol and were colored products. Based on the characterization data, the general formula of the complexes was proposed as [M(IDB)n(H2O)x]Cly.H2O, where M = Co(III) (with n = 2, x = 2, y = 1), Zn(II) (with n = 2, x = 2, y = 0), and Pd(II) (with n = 2, x = 0, y = 0). The geometries suggested for these complexes are the octahedral geometry for the Co(III) complex; the square planar geometry for the Pd(II) complex is reasonable, given its electronic configuration and spectroscopic properties. All complexes were found to be stable in air, that is, nonvolatile solids that decompose at high temperature. They are soluble in polar aprotic solvents like DMF and DMSO, sparingly soluble in methanol, chloroform, and dichloromethane and insoluble in many other common organic solvents. The elemental microanalysis (CHN) results were found to be in reasonable agreement with the proposed molecular formulas, and the ratio of the elements was further confirmed by elemental microanalysis.

2.2. Metal: Ligand Ratio

The mole ratio method of Yoe and Jones [16] is a method for determining the stoichiometry of stable complexes without resorting to any assumptions. The method is where the concentration of the ligand is changed, and the metal ion concentration remains fixed. The absorbance of each solution is determined and plotted as a function of the molar ratio of ligand to metal. The resulting plot is linear with a break in the line at the stoichiometric point where the complex is formed. This procedure is commonly used with UV–vis spectroscopy as the detection method. The molar ratio of the metal: ligand (M:L) was determined to be 1:2 for the chelate complexes studied. The results obtained are consistent with those published in similar metal complexes [17, 18].

2.3. Molar Conductivity

The values of molar conductance are given for each metal complex in Table 1. The molar conductivities were measured at room temperature in ethanol (10⁻³ M). The molar conductivities of the complexes of Co(III) are higher than expected, indicating that the complexes are electrolytic, with the type of electrolyte being 1:1. This is because there are chloride ions (Cl⁻) outside the coordination sphere. The significantly lower conductivity readings of the Zn(II) and Pd(II) complexes, on the other hand, suggest that the complexes are nonelectrolytic because there are no chloride ions (Cl⁻) outside the coordination sphere [19, 20].

TABLE 1.

Physical properties and elemental analysis for compounds.

Compound	Color	m.p C	Yield %	Molecular formula (M.Wt) (g/mol)	% Found (Calculated)				Molar conductivity
					%C	%H	%N	%M
IDB	Brown	148	70	C15H11N3O2 (265.27)	67.99 (67.92)	4.20 4.18))	15.94 15.84))	—	—
[Co(IDB)2(H2O)2]Cl.H2O	Dark Brown	228	62	C30H26N6O7CoCl (676.95)	53.25 (53.23)	3.99 (3.87)	12.50 (12.41)	8.77 (8.71)	44.56 Ionic
[Zn(IDB)2(H2O)2].H2O	Umber	216	68	C30H26N6O7Zn (647.95)	55.66 (55.61)	4.06 (4.04)	12.99 (12.97)	10.20 (10.09)	17.29 Nonionic
[Pd(IDB)2].H2O	Purple	202	60	C30H22N6O5Pd (652.95)	55.30 (55.18)	3.41 (3.40)	12.91 (12.87)	16.42 (16.30)	12.32 Nonionic

2.4. Electronic Spectral Studies

The geometric symmetry and electronic configuration of the complexes obtained from the reaction of the ligand (IDB) with the ions Co(III), Zn(II), and Pd(II) were studied by the analysis of UV–vis spectra and magnetic measurements (Table 2 and Figure S1). The electronic spectrum of the free IDB ligand shows characteristic absorption bands in the U.V. region, related to intramolecular electronic transitions. The bands at 283 and 264 nm are characteristic of π→π* transitions seen in the presence of aromatic rings or unsaturated bonds like C═C or C═N in the ligand structure. Additionally, the bands at 352 and 376 nm are attributed to lower‐energy n→π* transitions, which are the result of excitation of lone pair electrons on heteroatoms such as N or O to antibonding π* orbitals. These may be considered as a reference set of values for the observation of spectral changes when forming a complex with the other species [21].

TABLE 2.

Electronic spectra in nm, magnetic moment, geometry and hybridization of compounds.

Compound	λ max (nm)	Transitions	µ eff (B.M)	Geometry	Hybridization
IDB	376	n →π*	—	—	—
	352	n →π*
	283	π →π*
	264	π →π*
Co(III)‐Complex	501	1A1g→1T2g(F)(υ1)	Dia	Octahedral (Regular)	d2sp3 (Low spin)
	410	1A1g→1T1g(F)(υ2)
	287	1A1g→1T1g(p)(υ3)
Zn(II)‐Complex	503	Charge transfer M→L	Dia	Octahedral (Regular)	Sp3d2
	395	π →π*
	287	π →π*
Pd(II)‐Complex	842	1A1g→1T1g (F)(υ1)	Dia	Square planer	dsp2
	764	1A1g→1T2g (F)(υ2)
	474	Charge transfer M→L
	308	π →π*
	265	π →π*

The electronic configuration of the Co(III) ion in a cobalt(III) complex is [Ar]3d⁶. The complex is diamagnetic, that is, it contains no unpaired electrons. This definitely indicates a low‐spin configuration, with the six electrons pairing up in the t₂g level, which are the lower‐energy ones; hence the strong ligand field and octahedral geometry. The diamagnetic behavior is another example of the presence of d²sp³ hybridization, which is typical of a complex with inner orbital. The electronic absorption spectrum corresponds to the Tanabe–Sugano diagram of a d⁶ ion in an octahedral field with a band at 501 nm for the spin‐allowed transition ¹A₁g → ¹T₁g, and another at 410 nm for the spin‐allowed transition ¹A₁g → ¹T₂g. The bands are present, indicating the octahedral geometry of the complex [22]. Similar electronic spectral assignments for octahedral Co(III) complexes with azo‐based ligands have been recently reported [23].

For the zinc(II) complex, the electronic configuration of the Zn(II) ion is [Ar]3d¹⁰. The filled d‐subshell causes the complex to be diamagnetic, which means there are no unpaired electrons. Although the theoretical d‐d transitions are forbidden for a d¹⁰ ion, the observed band at 503 nm is attributed to a metal‐to‐ligand charge transfer (M→L) transition, which explains the color of the complex. The other bands at 395 and 287 nm correspond to intraligand π→π* transitions, which have seen a red shift on coordination. The geometry of the complex is octahedral with sp³d² hybridization of outer orbitals, while the inner 3d orbitals are filled [24].

Regarding the Pd(II) complex, palladium exhibits a [Kr]4d⁸ electronic configuration. It is well established that second‐row transition elements with a d⁸ configuration strongly favor square planar geometry, achieved through dsp² hybridization. This arrangement requires electron pairing to vacate the d_x ² ₋ᵧ² orbital for hybridization, and the diamagnetic nature of the complex confirms this square planar structure with all eight electrons paired in the lower‐energy orbitals. The electronic spectrum shows low‐energy bands at 842 and 764 nm, characteristic of the energy level splitting in a square planar field (D₄h symmetry), representing transitions from the ¹A₁g ground state to excited states such as ¹T₁g and ¹T₂g (or more precisely, ¹A₁g → ¹A₂g in square planar nomenclature). Additionally, a band at 474 nm is assigned to a charge transfer transition [25].

2.5. Infrared Spectra

Infrared spectra of the ligand (IDB) with Co(III), Zn(II), and Pd(II) ions have been studied (Table 3) [26]. The free ligand (IDB) exhibited a sharp, intense absorption band at 3551.98 cm^{− 1}, assigned to the N─H stretching vibration. In contrast, all metal complexes showed a noticeable shift to lower wavenumbers (3394–3398 cm^{− 1}), accompanied by broadening of the band (Figure S2). This clearly indicates the involvement of the amino nitrogen atom in coordination with the metal ion. Comparable shifts in ν(N─H) upon metal coordination have been documented in recent FT‐IR studies of azo‐imine and Schiff base complexes [27]. Similarly, the band related to (O─H) in the ligand appears at 3397 cm⁻¹ as a weak band, but it transforms in the complexes into broad and strong bands at different frequencies. This also indicates the presence of water molecules (either in the coordination sphere or outside) and that these bands are broad due to hydrogen bonding. The stretching vibration of (C═O) of the ligand is observed at 1630.75 cm⁻¹ when analyzing the carboxyl groups. This frequency shows large changes in the three complexes, becoming lower (1610–1614 cm⁻¹), and this is conclusive evidence of coordination via the oxygen atom of the carboxyl group. To determine the nature of the carboxylate binding, the difference (Δν) between the asymmetric stretching (ν(COO─)_asym and the symmetric stretching ν(COO─)_sym was calculated. The results show that the difference values range from 132.45 to 155.69 cm⁻¹, which scientifically indicates that the carboxylate group is coordinating to the metal as a monodentate ligand in these complexes. Similar Δν values for monodentate carboxylate coordination have been reported for azo‐based metal complexes [28, 29]. As for the azo group (N═N), it recorded a frequency of 1456.76 cm⁻¹ in the ligand, which shifted slightly in the complexes to around 1443 cm⁻¹. This minor shift indicates an effect on the electronic cloud of the azo group due to neighboring coordination, although the primary bonding is concentrated on the amino nitrogen and carboxyl oxygen. In the fingerprint region, there is a relative consistency in the (C─H) aromatic vibrations around 3056 cm⁻¹, suggesting the stability of the aromatic rings and that their direct carbon structure is not significantly affected by the complexation [30, 31]. Since cobalt in the +3 oxidation state tends to form octahedral complexes, and based on the appearance of hydration bands and coordinating water noted, it is expected that the metal center will be surrounded by nitrogen and oxygen atoms from the ligand, in addition to water molecules to complete the coordination number of 6. Moreover, the d¹⁰ zinc ion typically prefers an octahedral arrangement. Evidence suggests coordination is through the carboxylate O (as a monodentate) and amino (N). The lowest value of Δν in the present case is 132.45 cm⁻¹, so it supports the hypothesis of strong and stable bonding within an octahedral structure, which is in accordance with the nature of this metal. The Pd (II) ion is of d8 configuration and, chemically, forms square planar complexes. The (M─O) bands are shifted to 533 cm⁻¹ (highest value among the complexes), which shows a very strong covalent bond and thus the stability of the square planar geometry imposed by the electronic distribution of palladium. From FT‐IR results, it is confirmed that the ligand (IDB) is a bidentate ligand which donates one electron pair from the nitrogen atom and another from the oxygen atom of the carboxyl group. Otherwise, water molecules are used as ancillary agents to complete the coordination sphere or outside it (crystal water) [32, 33].

TABLE 3.

FT‐IR spectra bands of the (IDB) and its metal complexes.

Group	IDB	Co(III)‐Complex	Zn(II)‐Complex	Pd(II)‐Complex
υ(N─H)	3551.98 s.	3398.53 br.	3397.66 br.	3394
υ(O─H) υ(H2O)	3397.29 w.br.	*3425.66 s. br.	*3479 s. br.	*3237.81 m. br.
υ(C─H)Aromatic	3056.01 w.	3055.29 w.	3057.03 w.	3057.06 w.
υ(COO─)asym	1695.62 s.	1686.57 s.	1685.11 s.	1686.85 s.
υ(C═O)	1630.75 s.	1614.16 w.	1610.28 m.	1610.36 m.
υ(COO─)sym	1566.88 m.	1532.40 w.	1552.66 w.	1531.16 m.
Δυ(υ(COO─)asym‐υ(COO─)sym)	—	154.17	132.45	155.69
υ(N═N)	1456.76 w.	1443.67 w.	1443.39 w.	1443.96 m.
υ(C═C)	1241.58 w.	1332.47 s.	1241.70 w.	1373.85 s.
υ(C─O)	1193.22 w.	1212.67 w.	1197.63 w.	1330.36 w.
υ(C─N)	1153.38 w.	1103.42 w.	1154.05 w.	1247.54 w.
υ(C─C)	1102.91 m.	1013.02 w.	1105.37 m.	1186.66 m.
υ(M─O)	—	521.75 w.	515.47 w.	533.16 m.

Vs, very strong; S, strong; m, medium; w, weak; br, broad; sh, sharp; as, asymmetric; s, symmetric.

*(H₂O) outside of sphere coordination.

2.6. Mass Spectra

The free ligand (IDB) exhibited a molecular ion peak (M+) at m/z = 265.3; this corresponds to the actual molecular weight of the ligand, 265.27, represented by the formula [C₁₅H₁₁N₃O₂], thus indicating the purity and chemical structure of the free ligand before complexation (Figure S3a). The next complex was the cobalt complex, and a peak was observed at 677.1, which corresponds to the calculated molecular weight for the formula (C₃₀H₂₆N₆O₇CoCl). This marked increase in mass is due to the coordination of the cobalt ion with two ligand molecules and the presence of a chlorine atom and water molecules in the structure, which improved the accuracy of the previous FT‐IR studies of the structure (Figure S3b). Similarly, the mass spectrum of the palladium complex supported this, showing a peak at 653.1, which corresponds to a molecular weight of 652.95 (C₃₀H₂₂N₆O₅Pd). This suggests that the complex contains a palladium ion and two ligand molecules, and the protons from the acidic groups have been eliminated to enable covalent coordination (Figure S3c). The complete correlation of observed peaks with calculated weights for both the ligand and the cobalt and palladium complexes provides scientific evidence that the chemical reactions went through the expected pathway, which would yield bidentate ligand complexes (2:1 ligand‐to‐metal ratio) [34, 35].

2.7. ¹H‐NMR Spectra

¹H‐NMR spectroscopy studies using TMS as an internal standard in DMSO‐d₆ [36]. Detailed chemical shift assignments are summarized in Table S2, and the spectra are shown in Figures S4 and S5. The free IDB shows a characteristic singlet at 1H, δ 13.68, corresponding to the proton of the carboxylic acid group (─COOH). The low‑field shift is due to the strong electron‑withdrawing effect of the adjacent carbonyl group (deshielding). Another singlet at δ 11.42 ppm (1H) is attributed to the indole N─H proton, which is also highly deshielded due to the anisotropic effect of the azo group. In the aromatic region (δ 8.80–7.12 ppm), 10 different signals are observed for the indole and benzoic acid rings. The indole H‑5 appears as a doublet at δ 8.80 ppm (J = 7.8 Hz, 1H), while H‑2 appears as a doublet at δ 8.42 ppm (J = 2.3 Hz, 1H). The characteristic doublet for H‑4 is found at δ 8.15 ppm (J = 8.0 Hz, 1H). The benzoic acid protons show typical splitting patterns: H‑6' appears as a doublet at δ 7.98 ppm (J = 7.5 Hz, 1H), H‑3' and H‑5' emerge as a complex multiplet (δ 7.86–7.76 ppm, 2H and δ 7.48–7.38 ppm, 2H respectively), and H‑4' is observed as a doublet at δ 7.12 ppm (J = 7.8 Hz, 1H). This means that coordination takes place via deprotonated carboxylate oxygen and the azo N, as in the case of IDB. The aromatic ring currents and electronic distribution are rather unaltered by complexation, since the metal atom is not directly attached to any carbon atoms of the aromatic systems. The NMR observations, in addition to the FT‑IR data (where the ν(O─H) was not observed), are consistent with the ligand acting as a bidentate N,O‑donor. The complex environment around Pd(II) is symmetric, characteristic of a square‑planar geometry (dsp² hybridization) with two molecules of the ligand coordinating in a trans conformation, leading to an equivalent aromatic environment for both IDB units.

2.8. ¹³C‐NMR Spectra

The successful complexation and the redistribution of electron density in the conjugated ligand framework is clearly demonstrated by the ¹³C NMR analysis of the ligand (IDB) and its Pd(II) complex (Figure S6 and S7) [37]. The free ligand shows characteristic resonances in the range of δ 106.42–163.12 ppm, whereas the Pd(II) complex shows resonances in the range of δ 104.60–161.06 ppm. The most diagnostic shift is seen for the carbonyl carbon of the carboxyl group, which is shifted from δ 163.12 ppm in the ligand to δ 161.06 ppm in the Pd(II) complex, reflecting the change in the environment of the carboxyl group upon metal coordination. The azo‐conjugated carbon resonance at δ 139.19 ppm also shifts to δ 137.89 ppm, indicating that the electronic distribution of the azo linkage is affected by Pd(II) coordination. The shifts of several aromatic resonances are in the range of 1.8–3.2 ppm, especially in the region of δ 129.60–124.49 ppm, which is consistent with the transmission of the coordination‐induced electronic effect through the extended π‐conjugated system. On the other hand, the small changes in the selected indole carbons, e.g., δ 137.00 → 137.04 and δ 119.50 → 119.51 ppm, suggest that the basic indole framework is preserved. The resonance at δ 106.42 ppm in the ligand is shifted to δ 104.60 ppm in the complex, which further confirms the change in the electronic environment around the indole moiety.

2.9. X‐ray Diffraction

The IDB crystallinity revealed by x‐ray diffraction (XRD) is a well‐defined powder architecture with broad, intense Bragg reflections from low angle (6.6863°) to high angle (63.9609°) 2θ [38] (Table S3 and Figure S8). The peaks are due to constructive interference of x‐rays scattered by atomic planes, in accordance with Bragg's law [39]: nλ = 2d sinθ, where n corresponds to the order of reflection (usually n = 1), λ is the x‐ray wavelength (typically Cu Kα at 1.5406 Å), d is the interplanar spacing, and θ is the diffraction angle. Calculations of d‐spacing, which is the direct result of this equation, d = λ2sinθ, reveal a systematic decrease from 13.20912 Å, for low angle (6.6863° 2θ, associated with large basal planes, probably (001) or layers of stacking in the aromatic azo framework), to 1.45441 Å at high angle (63.9609° 2θ, associated with ordered short‐range packing). Low‐angle peaks with high relative intensities (98% to 100%) and large d‐spacings (>7 Å) are typical of ordered lamellar or π‐π stacked domains seen in organic dyes; the mid‐to‐high angle peaks (16° to 64° 2θ) decrease in intensity (93% to 71%) and d‐spacing (5.37 Å to 1.45 Å), suggesting finer intramolecular packing, such as van der Waals contacts between azo (─N═N─) and aromatic rings. In particular, all peaks show a uniform FWHM of 0.0900° indicating a high degree of uniformity in crystallites and absence of microstrain, while the peak heights peak at 660 cts, suggesting the dominance of the (110)/(200)‐like planes [40].

2.10. FEFE‐SEM

Field‐emission scanning electron microscopy (FESEM) reveals the mesoscale morphology of IDB, imaging irregularly shaped particles at 50 kX magnification with a cross‐sectional resolution of 200 nm, with an average size of 99 nm. These irregularly shaped nanosized particulates are indeed consistent with the XRD‐derived crystallite size of ∼90–100 nm (via Scherrer), indicating that observed grains are single crystals or only marginally polycrystalline domains and not aggregates. The absence of uniform faceting (irregular growth) indicates kinetically trapped growth, which may be due to fast precipitation at synthesis, whereas the difference in “distorted crystal structure” is attributed to the smoother growth morphologies of metal complexes after annealing. The surface properties deduced from FESEM (rough and non‐spherical) suggest high specific surface area, which is beneficial for applications such as dye‐sensitized solar cells or catalysis, where a 99 nm size is favorable for light scattering and charge transport properties while minimizing recombination [41].

2.11. ADMET Analysis

ADMET have been analyzed based on SwissADME [42], and it is seen that the analysis gives a complete physical and chemical characterization of the ligand (IDB) and its metal complexes with Co, Zn, and Pd. It is important to note, however, that these predictions are theoretical models and must be validated by experimental testing using suitable in vitro (such as Caco‐2 permeability, plasma protein binding, metabolic stability assays) and in vivo studies. The ADMET results are intended to be used as a guide for future drug development efforts and should not be used as definitive pharmacokinetic data. The free IDB compound has an appropriate molecular weight of 20 and a polar surface area (TPSA) of 77.81 Ų, which are within the range that is ideal for drug absorption. The metal complexes have a significantly higher number of heavy elements (more than 40) and a higher polar surface area (166.4 Ų for the cobalt and zinc complexes) (Table 4). This swelling in size and polarity has a negative impact on aqueous solubility. The solubility of IDB is moderately to highly soluble in different models of solubility (e.g. ESOL, Ali); however, the solubility values drop significantly after the formation of metal complexes, making them poorly soluble or completely insoluble. This is a serious pharmaceutical problem for delivery and distribution of drugs in the organism [43, 44].

TABLE 4.

ADMET of IDB and metal complexes with SWISS ADME software.

ADMET	Property	IDB	Co(III)‐Complex	Zn(II)‐Complex	Pd(II)‐Complex
Physicochemical properties	#Heavy atoms	20	43	43	41
	#Aromatic heavy atoms	15	30	30	30
	Fraction Csp3	0	0	0	0
	#Rotatable bonds	3	2	2	2
	#H‐bond acceptors	4	8	8	6
	#H‐bond donors	2	4	4	2
	MR	75.88	161.41	161.41	155.32
	TPSA (Å2)	77.81	166.4	166.4	132.26
Lipophilicity	iLOGP	1.92	0	0	0
	XLOGP3	3.35	6.38	6.38	7.32
	WLOGP	4.28	6.47	6.47	6.6
	MLOGP	2.11	1.65	1.65	3.16
	Silicos‐IT Log P	3.63	0.07	0.11	3.1
	Consensus Log P	3.06	2.91	2.92	4.04
Water solubility	ESOL Log S	−3.95	−8.11	−8.15	−8.8
	ESOL solubility (mg/ml)	2.96E‐02	4.85E‐06	4.47E‐06	1.01E‐06
	ESOL solubility (mol/l)	1.12E‐04	7.78E‐09	7.09E‐09	1.59E‐09
	ESOL class	Soluble	Poorly soluble	Poorly soluble	Poorly soluble
	Ali Log S	−4.66	−9.67	−9.67	−9.92
	Ali solubility (mg/mL)	5.78E‐03	1.34E‐07	1.36E‐07	7.55E‐08
	Ali solubility (mol/L)	2.18E‐05	2.16E‐10	2.16E‐10	1.19E‐10
	Ali Class	Moderately soluble	Poorly soluble	Poorly soluble	Poorly soluble
	Silicos‐IT LogSw	−5.27	−11.72	−11.74	−11.75
	Silicos‐IT solubility (mg/mL)	1.41E‐03	1.18E‐09	1.15E‐09	1.14E‐09
	Silicos‐IT solubility (mol/L)	5.33E‐06	1.89E‐12	1.83E‐12	1.79E‐12
	Silicos‐IT class	Moderately soluble	Insoluble	Insoluble	Insoluble
Pharmacokinetics	GI absorption	High	Low	Low	Low
	BBB permeant	No	No	No	No
	Pgp substrate	No	No	No	No
	CYP1A2 inhibitor	Yes	Yes	Yes	No
	CYP2C19 inhibitor	No	Yes	Yes	Yes
	CYP2C9 inhibitor	No	No	No	No
	CYP2D6 inhibitor	No	No	No	No
	CYP3A4 inhibitor	No	No	No	No
	log Kp (cm/s)	−5.54	−5.57	−5.61	−4.98
Druglikeness	Lipinski #violations	0	2	2	1
	Ghose #violations	0	3	3	3
	Veber #violations	0	1	1	0
	Egan #violations	0	2	2	2
	Muegge #violations	0	4	4	3
	Bioavailability score	0.56	0.17	0.17	0.55
Medicinal Chemistry	PAINS #alerts	1	1	1	1
	Brenk #alerts	1	1	2	0
	Leadlikeness #violations	0	2	2	2
	Synthetic accessibility	2.55	5.3	5.29	5.46

The compounds have high gastric‐intestinal (GI) absorption and can theoretically be in the “white zone” part of the diagram (Figure 1), where good absorption across the intestinal wall can be achieved. The large size and high polarity of the metal complexes, on the other hand, are classified as low absorption and therefore taken out of the white zone. Regarding blood–brain barrier (BBB) permeability, the results are in line with the lack of BBB penetration of all compounds, a positive finding for excluding compounds from the “yellow zone” in Figure 1, if the intention is to avoid any side effects on the central nervous system (CNS). In addition, cytochrome enzyme inhibition data indicate that metal introduction gives rise to altered anticipated metabolic pathways and drug interactions; the complexes start enzyme inhibition of the CYP2C19 enzyme, which is not the case for the parent compound [45].

FIGURE 1.

BOILED‐Egg plot for ADMET analysis of ligands and complexes (SwissADME).

Figure 2 shows the radar plot indicating the visual assessment of the compounds' drug‐likeness rules. In the case of the IDB compound, its range is mainly in the ideal “pink zone” as confirmed by the table number showing no violations of Lipinski's Rule and Veber rules, which yields a relatively excellent bioavailability score of 0.56. The metal complexes, however, are not safe in the radar plot with respect to the three properties mentioned—size (SIZE), polarity (POLAR), and poor solubility (SOL). The geometric deviation in the plot is reflected numerically in the table as several violations of the drug‐likeness rules, which result in a significant drop in Bioavailability scores of cobalt and zinc complexes (0.17) and make them less successful as an orally administered drug by conventional methods. The IDB compound is easy, flexible, and straightforward in lab preparation steps with a chemical and manufacturing synthetic accessibility score of 2.55. This value rises to more than 5.3 when the metal forms complexes with this molecule, owing to increased spatial complexity and thus difficulty in the synthesis and purification steps. It is also worth highlighting that there is one alert in the PAINS filter for all compounds, meaning that in the future, when testing biological systems, there is a need to carefully consider and examine any possible nonspecific interactions or false‐positive results [46].

FIGURE 2.

Radar plot for drug‐likeness properties (SwissADME). LIPO (lipophilicity, where higher LogP), SIZE (molecular weight/heavy atoms), POLAR (polarity via TPSA), SOL (solubility), INSATU (saturability/aromaticity fraction), and FLEX (flexibility from rotatable bonds).

2.12. Molecular Docking

The MOE 2015 software was used for molecular docking analysis of the protein receptors involved in leukemia [47, 48]. The target protein (T315I Abl mutation receptor) was extracted from the Protein Data Bank with the code (PDB ID: 2V7A) (Table 5). The online PDBsum server [49] and Ramachandran plot were used to check the accuracy of the results and validate the reliability of the crystal structure of this receptor (Figure 3). The examination confirmed the quality of the protein structure, showing that more than 86.8% of the amino acid residues are in allowed geometric regions with no missing residues, giving a good and reliable basis for the molecular docking process.

TABLE 5.

Ligand interaction report of IDB, Co(III), Zn(II), and Pd(II) metal complexes.

Complex	Interaction type	Amino acid	Distance (Å)	Energy (kcal/mol)
IDB ligand	H‐bond (acceptor)	LYS 271	3.37	−1.0
	π‐H	LYS 271	4.05	−1.1
	π‐H	LEU 370	3.63	−0.6
Co(III)‐Complex	H‐bond (donor)	THR 319	2.71	−3.7
	H‐bond (donor)	MET 318	3.05	−3.3
	π‐H	LEU 248	4.32	−0.8
Zn(II)‐Complex	H‐bond (donor)	GLY 249	2.85	−4.7
	H‐bond (acceptor)	LYS 271	3.17	−1.0
	π‐H (×2)	LEU 248	3.75, 4.08	−0.7, −0.8
	π‐H (×2)	GLN 252	4.03, 4.51	−0.9, −0.6
	π‐H	TYR 253	4.35	−0.6
Pd(II)‐Complex	H‐bond (donor)	ASP 381	3.05	−3.2
	H‐bond (acceptor)	ASN 322	2.81	−2.3
	π‐H (×2)	LEU 248	3.72, 3.75	−1.1, −0.9

FIGURE 3.

Ramachandran plot of (PDB ID: 2V7A).

For the parent organic ligand (IDB), the results showed a favorable binding energy of (−5.99) kcal/mol, indicating strong and spontaneous binding affinity. This was accompanied by a geometric deviation of (1.78) Å, which is below the critical threshold adopted scientifically, confirming the validity of the simulation. This complex relies on a hydrogen‐bond interaction as an H‐acceptor with the amino acid lysine (LYS 271) at a distance of 3.37 Å and an energy of (−1.0) kcal/mol. In addition, π‐H interactions emerge as an extra stabilizing factor that links aromatic rings: a π‐H interaction with the same amino acid (LYS 271) at 4.05 Å and energy (−1.1) kcal/mol, and a similar interaction with leucine (LEU 370) at 3.63 Å and energy (−0.6) kcal/mol (Figure 4).

FIGURE 4.

Molecular docking of the IDB ligand with the Aurora kinase (2V7A). (A) 2D interaction diagram (MOE) showing hydrogen bond (green) and π‑H (magenta) contacts. (B) 3D binding pose within the active site. (C) Electrostatic surface map of the binding pocket.

Co(III) complex, a decrease in dynamic interaction strength compared with the parent ligand was observed. It exhibited a binding energy of (−5.36) kcal/mol, along with a geometric deviation of (1.80) Å. This complex demonstrates a direct dependence on hydrogen bonds as H‐donors, where the first interaction is formed with threonine (THR 319) at an ideal short distance of 2.71 Å and a high binding energy of (−3.7) kcal/mol. The second interaction occurs with methionine (MET 318) at a distance of 3.05 Å and an energy of (−3.3) kcal/mol. Additionally, leucine (LEU 248) contributes a supportive π‐H interaction at 4.32 Å with energy (−0.8) kcal/mol (Figure 5).

FIGURE 5.

Molecular docking of the Co(III)‐Complex with the Aurora kinase (2V7A). (A) 2D interaction diagram (MOE) showing hydrogen bond (green) and π‑H (magenta) contacts. (B) 3D binding pose within the active site. (C) Electrostatic surface map of the binding pocket.

For the divalent zinc complex, although its binding energy was the weakest at (−5.24) kcal/mol, it achieved a very excellent geometric deviation value of (1.02) Å, indicating superior conformational stability. This complex is characterized by a hydrogen‐bond interaction as an H‐donor with glycine (GLY 249) at a distance of 2.85 Å and energy of (−4.7) kcal/mol, and a hydrogen‐bond interaction as an H‐acceptor with lysine (LYS 271) at 3.17 Å and energy of (−1.0) kcal/mol. Moreover, the results revealed exceptional density of π‐H interactions, including two separate interactions with leucine (LEU 248) at distances of 3.75and 4.08 Å with energies of (−0.7) and (−0.8) kcal/mol, respectively; two interactions with glutamine (GLN 252) at distances of 4.03 and 4.51 Å with energies of (−0.9) and (−0.6) kcal/mol, respectively; and a single interaction with tyrosine (TYR 253) at 4.35 Å with energy (−0.6) kcal/mol (Figure 6).

FIGURE 6.

Molecular docking of the Zn(II)‐Complex with the Aurora kinase (2V7A). (A) 2D interaction diagram (MOE) showing hydrogen bond (green) and π‑H (magenta) contacts. (B) 3D binding pose within the active site. (C) Electrostatic surface map of the binding pocket.

Moreover, the divalent palladium complex, which achieved the best and strongest binding energy of all, with an extremely negative value of (−7.65) kcal/mol, outperforming the others by a substantial margin. It also recorded a near‐ideal geometric deviation value of (1.03) Å. The remarkable stability of this complex is driven by a hydrogen‐bond interaction as an H‐donor with the amino acid aspartic acid (ASP 381) at 3.05 Å and energy (−3.2) kcal/mol, and a hydrogen‐bond interaction as an H‐acceptor with asparagine (ASN 322) at 2.81 Å and energy (−2.3) kcal/mol. Two π‐H interactions with leucine (LEU 248) have similar energies of (−1.1) and (−0.9) kcal/mol with distances of 3.72 and 3.75 Å, respectively (Figure 7). The parent organic ligand had superior binding energy, while the cobalt and zinc complexes had superior mechanical stability, but only the palladium complex dominated both key indicators. These accurate readings of bond interactions, bond distances and bond energies establish that the introduction of the central palladium ion was an ideal electronic and geometric fit to the trans‐ferase protein pocket, giving it a highly efficient and selective network of interactions, which made it the most significant and valuable drug candidate in this study. These findings are consistent with recent reports on metal complexes targeting cancer‐related proteins [50, 51]. It is worth noting, however, that these in silico results are predictive and should be validated by experimental methods such as enzyme inhibition assays, surface plasmon resonance binding studies, or isothermal titration calorimetry (ITC) to confirm the proposed interactions and binding affinity. The docking results should be used as a hypothesis‐generating tool to help guide future mechanistic studies and not as definitive proof of the mechanism of action.

FIGURE 7.

Molecular docking of the Pd(II)‐Complex with the Aurora kinase (2V7A). (A) 2D interaction diagram (MOE) showing hydrogen bond (green) and π‑H (magenta) contacts. (B) 3D binding pose within the active site. (C) Electrostatic surface map of the binding pocket.

2.13. Targeting Oxidative Stress in HL‐60 Cells

This application aims to develop a therapeutic agent that is capable of inducing “cellular toxicity” that selectively targets and kills cancer cells, but also has the capacity to “modulate oxidative stress” that will reduce cellular damage (lipid peroxidation) in healthy tissues [52, 53]. The present study is designed to examine the dual activity of a Pd(II) complex on the HL‐60 leukemia cell line and its toxicity toward normal cells (HdFn) with a special emphasis on its effect on malondialdehyde (MDA) level regulation. This will give a scientific insight into its future as a potential drug candidate with therapeutic efficiency and fewer side effects. Table 6 shows the effect of the Pd(II) complex on the level of the oxidative stress marker, malondialdehyde (MDA), in the leukemia cancer cell line (HL‐60) and the normal cells (HdFn). MDA is scientifically an important biomarker of lipid peroxidation and oxidative stress‐induced cell damage. On analyzing the data, MDA concentration in the cancer cell (HL‐60) is found to be quite higher at (357.9 ± 0.2121 ng/mL) compared to (286.1 ± 5.728 ng/mL) for normal cells (HdFn) (Figure 8a). The results from two independent biological replicates (n = 2) showed a significant decrease in MDA levels in both cell types after treatment with the Pd(II) complex (Table 6). The difference was statistically significant in both normal cells (P = 0.0259) and cancer cells (P = 0.0135), indicating that the compound can reduce oxidative stress and protect cells from lipid peroxidation. In normal cells, MDA levels dropped to (253.2 ± 7.354 ng/mL), while in cancer cells, MDA decreased to (318.4 ± 9.405 ng/mL). Both p‐values are < 0.05, this confirms the effect is not due to chance but is actually caused by the compound and is a scientific and statistical confirmation that the compound is capable of reducing oxidative stress and protecting cells from lipid damage.

TABLE 6.

Comparison of MDA concentration (ng/mL) before and after treatment with Pd (II) complex in HdFn and HL‐60 Cell Lines.

Parameter	HdFn (IC50 Co. 164.7)		HL‐60 (IC50 Co. 134.4)
	Before	After	Before	After
Number of values (n)	2	2	2	2
Mean ± SD	286.1 ± 5.728	253.2 ± 7.354	357.9 ± 0.2121	318.4 ± 9.405
P value	0.0259		0.0135

FIGURE 8.

(a) Comparative analysis of MDA levels before and after exposure to Pd (II) complex. (b) IC₅₀ of Pd (II) complex on MDA concentration in HdFn and HL‐60 cell lines.

2.14. Antileukemic Activity

‘Anti‐leukemic Activity' indicates the ability of a given compound or drug to kill or stop the growth of leukemia cells, and to minimize damage to healthy normal cells in order to keep the patient safe. The specificity of drug action in killing cancerous cells while sparing normal cells is the basis for the development of any successful drug targeted for cancer [54]. The value of cells left to grow naturally compared to cells exposed to the therapeutic compound (Control Group vs. Test Group) is presented in the analysis of Table 7. The data show that the mean viability of normal cells (HdFn) in the control group was 312.55, whereas the mean value for the leukemia cells (HL‐60) was 357.70. After treatment, it was found that there was a significant decrease in the viability of both cell types: Normal cells decreased to 269.20 and cancer cells decreased to 322.05. The first drop indicates that the compound has a global inhibitory activity on cell growth, which is suitable to investigate its behavior at other concentrations to explore the desired selectivity [55]. This analysis highlights the mechanism of action of the drug starting at 25 concentrations and reaching up to 400 µg/mL; this is clearly a phenomenon medically described as a dose‐dependent response [56]. The viability rate of cancer cells (HL‐60) has reduced to 89.79% at the lowest concentration used (25 µg/mL), whereas the viability rate of HdFn cells has increased to 95.99%. Viability of both cell types decreased with continued increasing of the drug concentration, but it is important to carefully watch the difference between the two types. The leukemia cells' resistance was completely broken down at 400 µg/mL (54.09% cell viability), and the normal cells had relatively better resistance (60.38% cell viability) (Table 8).

TABLE 7.

General effect of the Pd(II)‐Complex on cell viability: Comparison between the control group and the test group.

Cells	Control		Test
	Mean	SD	Mean	SD
HdFn	312.55	31.75	269.20	27.15
HL‐60	357.70	14.14	322.05	13.65

TABLE 8.

Effects of graded concentrations of the compound on cell viability ratios for cancerous versus normal cells.

Concen. µg/mL	HL‐60		HdFn
	Mean	SD	Mean	SD
400.00	54.09	2.57	60.38	0.81
200.00	62.92	3.92	73.23	1.20
100.00	73.15	1.33	83.95	2.10
50.00	81.94	2.12	97.65	2.75
25.00	89.79	5.96	95.99	1.68
IC50	134.4		164.7

The importance of the half‐maximal inhibitory concentration (IC₅₀) [57] values becomes apparent as they move into a critical point in the evaluation of cancer treatment; the IC₅₀ value was (134.4 µg/mL) for cancer cells and higher for normal cells at (164.7 µg/mL) (Figure 8b). The selectivity index of 1.23 (IC₅₀ HdFn/IC₅₀ HL‐60) indicates moderate preferential toxicity toward cancer cells over normal fibroblasts. While the IC₅₀ value of the Pd(II) complex (134.4 µg/mL, approximately 206 µM) is higher than some reported Pd(II) complexes (Table 9), several factors should be considered in interpreting these comparative values. First, variations in assay conditions, including cell line passage number, serum concentration, incubation time, and cell seeding density, can significantly affect reported IC₅₀ values. Second, the molecular weight of our complex (652.95 g/mol) contributes to a higher µg/mL value compared to lower molecular weight complexes. Third, and most importantly, the present study provides a comprehensive characterization package including: (i) detailed spectroscopic characterization of the ligand and all three metal complexes; (ii) experimental oxidative stress modulation data (MDA reduction) that provides mechanistic insights; (iii) ADMET predictions for drug‐like properties; and (iv) molecular docking with a clinically relevant drug‐resistant target (Abl T315I mutant). This integrated approach, combining synthesis, characterization, biological evaluation, and computational studies, provides a more complete understanding of the structure–activity relationships and therapeutic potential of these complexes than a simple IC₅₀ comparison would offer.

TABLE 9.

Comparison of Pd(II) complexes with reported antileukemic activity against HL‐60 cells.

Pd(II) complex	Reported IC50 (µM)	Key findings	Refs.
Our Pd(II) complex	134.4 µg/mL (∼206 µM)	Modest selectivity over normal HdFn cells (164.7 µg/mL); reduces MDA levels; modulates CAT; strong Aurora kinase binding	This study
cis‐[PdCl2(Ad)(Gua)] (mixed‐ligand nucleobase)	11.29 ± 2.91 µM	Higher activity than curcumin; significant antioxidant activity against DPPH	[58]
cis‐[PdCl2(NH3)(Gua)]	48.03 ± 9.67 µM	Varying degrees of cytotoxicity; no effect on normal NB1RGB cells	[59]
Biphosphinic palladacycle complex (BPC)	< 8.0 µM (within 5 h)	Induces apoptosis; lysosomal membrane rupture; no toxicity to human lymphocytes; selectivity for transformed cells	[60]
PdPyTT [PdCl2(μ‐PyTT)2]Cl2·4H2O	20.7 µM (24 h)	Time‐ and dose‐dependent cytotoxicity; increases ROS, caspase‐3/‐9 activation; harmless to healthy leukocytes	[61]
Phenyl‐substituted thiazine/thiazoline Pd(II) complexes	46.39–62.74 µM	Moderate cytotoxic effect; Pd accumulation inside cells; apoptosis induction	[62]

3. Conclusions

The overall design, synthesis, and detailed characterization of a novel azo dye ligand, (E)‐2‐((1H‐indol‐3‐yl)diazenyl)benzoic acid (IDB) and the corresponding metal complexes, such as Cobalt(III), Zinc(II), and Palladium(II) complexes, are completed. The two N atoms of the azo and the deprotonated carboxylate O atoms were identified as the coordination sites as a result of FT‐IR analysis. The Co(III) and Zn(II) complexes were determined to have octahedral geometry, and the former complex was determined to have a low‐spin d2sp3 hybridization. The square planar Pd(II) complex, on the other hand, showed the most promising biological activity. It showed cytotoxic activity against human leukemia cells (HL‐60) with moderate selectivity over normal HdFn cells (IC₅₀ = 134.4 µg/mL vs. 164.7 µg/mL, SI = 1.23). This selectivity is moderate compared to targeted therapeutics, but is a good starting point for further structural optimization to increase the therapeutic index. Molecular docking results indicate that the enzyme Aurora kinase (2V7A) has a good binding affinity of −7.65 kcal/mol with hydrogen bonding and π–π interactions. Moreover, the Pd (II) complex has been shown to modulate the levels of malondialdehyde (MDA) both in cancer and normal cells. The results were promising, but there were some limitations in this study: The antileukemic activity was tested in only one leukemia cell line (HL‐60) and one normal cell line (HdFn), which limits the generalizability of our findings regarding antileukemic potential. The selectivity and potency of the Pd(II) complex may vary among different leukemia subtypes, and the efficacy of the compound against solid tumors has not been investigated. The assessment of oxidative stress was limited to MDA measurements, and CAT activity was not evaluated in the present study. These restrictions should be taken into account when interpreting the anti‐leukemic activity reported here, and future investigations should include a wider range of leukemia cell lines, detailed mechanistic studies, and comprehensive oxidative stress profiling to validate and extend our current findings. Moreover, the molecular docking and ADMET results are computational predictions that need experimental validation, and the proposed mechanism of action has yet to be confirmed by enzyme‐based assays. Finally, the lack of in vivo efficacy and toxicity studies restricts the translation of the present findings. Thus, future studies should involve screening of a wider panel of cancer cells, detailed mechanistic and oxidative stress studies, experimental validation of computational results, and preclinical animal studies to further validate the therapeutic potential of these complexes.

4. Experimental Section

4.1. Chemicals and Reagents

Unless otherwise stated, all the chemicals and reagents used in this study were of analytical grade. A summary of the sources, purity and grades of the main materials is provided in Table S1.

4.2. Instrumentation and Measurements

The determination of the contents of carbon, hydrogen, and nitrogen of the ligand and metal complexes was done by the elemental analyzer LECO 923 USA, while ¹H NMR spectra were recorded on Bruker 400 MHz spectrometer with the help of DMSO‐d₆ media and TMS as an internal standard. The mass spectra were obtained with an AB Sciex 3200 QTRAP. UV–visible spectra were recorded in absolute ethanol over the wavelength range 200–1100 nm. Fourier Transform Infrared (FT‐IR) spectra were recorded between 4000 and 400 cm^{− 1} using a Shimadzu 8400 S FT‐IR spectrophotometer with samples prepared as KBr disks. x‐ray diffraction (XRD) studies were conducted with a Bestic Germany x‐ray diffractometer (XRD 600) using Cu Kα radiation over a range of 20°–80° 2θ. Conductivity measurements were carried out at room temperature with a digital conductivity bridge in 10^{− 3} M ethanol solutions. Melting points or decomposition temperatures were obtained by means of a Stuart Electrothermal 9300 melting point apparatus. A field‐emission scanning electron microscope (FE‐SEM) was used to study the surface morphology with high‐resolution images of the ligand structure and morphology. Precoated silica gel plates with a fluorescence indicator (F₂₅₄, Merck KGaA, Darmstadt, Germany) were used for the thin‐layer chromatography. The reaction was monitored under UV light at 254 nm and 365 nm.

4.3. Synthesis of (E)‐2‐((1H‐indol‐3‐yl)Diazenyl)Benzoic Acid (IDB)

The ligand (IDB) was synthesized according to the diazotization/coupling method described by the researchers in the literature [63, 64, 65] with a few modifications (Scheme 1). 1.37 g (0.01 mol) of 2‐aminobenzoic acid was first dissolved in a mixture of 50 mL of Et‐OH and 7 mL of concentrated HCl. Ice chips were added to this solution to cool it down to 0°C. Then 0.70 g (0.01 mol) of NaNO₂ (in 30 mL of distilled water) was added dropwise, stirring, keeping the temperature below 5°C. This reaction mixture was then left for 45 min. Then, diazonium salt solution was added dropwise, with stirring, to a solution of 1.17 g (0.01 mol) of indole in 30 mL of Et‐OH (3:1) and 5 mL of 10% NaOH solution. A precipitate formed during this step, which was subsequently filtered and washed multiple times with distilled water. The final product was dried and recrystallized using Et‐OH.

SCHEME 1.

Synthesis of ligand (IDB) and metal complexes.

4.4. General Synthesis of Metallic Complexes

A solution of the IDB ligand (0.53 g, 2 mmol) in 40 mL of ethanol was prepared and heated to 60°C. Stoichiometric amounts of the corresponding divalent metal chloride salts CoCl₂·6H₂O (0.2379 g, 1 mmol), ZnCl₂ (0.1363 g, 1 mmol), and PdCl₂ (0.1773 g, 1 mmol) were dissolved separately in 30 mL of prewarmed ammonium buffer solution. An ammonium acetate buffer solution (0.1 M, pH 7.0) was freshly prepared by mixing appropriate volumes of 0.1 M ammonium acetate and 0.1 M acetic acid. Stoichiometric amounts of the corresponding divalent metal chloride salts were dissolved separately in 30 mL of this prewarmed buffer solution. The metal salt solution was then added dropwise to the ligand solution in a 1:2 (metal: ligand) molar ratio. The resulting mixture was refluxed for 2 h under continuous stirring, with the pH maintained at 7.0 ± 0.1 according to general procedures for azo‐based metal complexes [66, 67]. A digital pH meter was used to measure the pH every 15 min and was calibrated. The pH was adjusted by dropwise addition of 0.1 M NH₃ or 0.1 M acetic acid as necessary to maintain the desired pH, usually 0.1–0.2 mL per adjustment. After refluxing, the volume of the solution was reduced to half by evaporation. The concentrated solution was then frozen in a freezer for 1 h to precipitate the solution. The formed complexes were filtered, washed with absolute ethanol and recrystallized from ethanol. Finally, the purified compounds were dried under vacuum, over anhydrous calcium chloride (CaCl₂), in a desiccator.

4.5. Molecular Docking

Abl T315I (Abelson tyrosine‐protein kinase 1) (PDB: 2V7A) is a clinically relevant therapeutic target in chronic myeloid leukemia (CML) and acute lymphoblastic leukemia (ALL) that is resistant to first‐line tyrosine kinase inhibitors (TKIs) like imatinib. The T315I Abl mutant is a major cause of treatment failure in leukemia patients, making the targeting of this mutant a major therapeutic challenge. The crystal structure of 2V7A (at 2.5 Å) is a reliable template for structure‐based drug design, and the cocrystallization of 2V7A with the Aurora kinase inhibitor PHA‐739358, which is effective against hematologic malignancies, confirms that the binding site is druggable. In addition, the structural homology between Abl and Aurora kinases (sequence identity ∼27%) indicates that inhibitors of the Abl T315I mutant could also inhibit Aurora kinases, providing a dual‐targeting approach. The structure and clinical relevance of 2V7A are therefore ideal for assessing the binding potential of our synthesized complexes. The crystal structure of the relevant proteins used in this study is the Abl T315I complex with the inhibitor PHA‐739358 (PDB code: 2V7A) [68, 69], which was obtained from the Protein Data Bank. The process of protein preparation started with the removal of the water molecules and coligands and involved the use of tools to locate possible binding sites of the protein structure. The 3D structures of the IDB ligand and its metal complexes Co(III), Zn(II), and Pd(II) were constructed in MOE 2015. The coordination geometry of the metal complexes were manually determined based on the experimentally determined geometry (octahedral for Co(III) and Zn(II); square planar for Pd(II)). The cocrystallized ligand PHA‐739358 was extracted from the crystal structure and re‐docked into the active site of 2V7A using the same docking parameters as before, namely triangular matching, London dG scoring, and GBVI/WSA dG rescoring. The RMSD between the re‐docked pose and the crystallographic pose was calculated as 0.92 Å, which is well below the predefined acceptance threshold of 2.0 Å, thus confirming the reliability and reproducibility of the docking protocol. The active site was defined by the position of the cocrystallized ligand for binding site definition. The key interacting residues were chosen to be within a radius of 6.5 Å, including residues like LYS 271, LEU 248, MET 318, and THR 319. All compounds were subjected to force field validation by energy minimization using the MMFF94X force field, which was parameterized for transition metals such as Co, Zn, and Pd. Optimization was carried out with a gradient convergence criterion of 0.001 kcal·mol⁻¹·A°⁻¹ and a maximum of 5000 iterations to ensure sufficient relaxation before docking‐based evaluation. The docking score was validated by comparing the docking score with the experimental IC₅₀ values of a set of known Abl inhibitors. The coefficient of determination obtained was R ² = 0.81, which is consistent with the predictive ability of the GBVI/WSA dG rescoring approach for this target [70].

4.6. MTT Assay

In this study, the following cancer cell lines were used, which were the products of Iranian Biological Resource Center (IBRC), Tehran, Iran. The newly synthesized compounds were tested against the human promyelocytic leukemia cell line (HL‐60) (RRID: CVCL_0002) and human dermal fibroblasts (HdFn) (RRID: CVCL_2408). Cells were grown in DMEM (Dulbecco's Modified Eagle's Medium) containing 10% fetal bovine serum, 100 U/mL penicillin and 100 µg/mL streptomycin with 95% air and 5% CO₂ at 37°C in a humidified atmosphere. The human cancer cell lines (3 × 10⁴ cells/mL) were treated with the various concentrations of the compounds dissolved in DMSO (25, 50, 100, 200, and 400 µg/mL). After 3 days of incubation, 10 µL of an MTT solution (5 mg/mL in sodium phosphate buffer (PBS, 0.1 M, pH 7.4)) was added to the cells and incubated for an additional 4 h at 37°C before the removal of the medium. The formazan crystals formed were then dissolved in 150 µL of DMSO. A Biotek Epoch 2 microplate reader was used to measure the absorbance of the resulting solutions at 540 nm. The positive reference control was doxorubicin (IC₅₀ = 0.5–1.0 µM for HL‐60 cells), and the negative and vehicle controls were cells treated with 0.1% (v/v) DMSO, the concentration used to dissolve the test compounds. Furthermore, the culture medium without cells was used as a blank control for background subtraction. For the evaluation of the selective toxicity of the compounds, the same procedure was carried out using normal human dermal fibroblasts (HdFn) from the same source. Identical conditions were used for the culture of the cells, and the same concentration of compounds was applied to assess the cytotoxic effects on noncancerous cells [71, 72].

4.7. Statistical Analysis

Data were analyzed statistically using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA) and SPSS Statistics 26.0 (IBM, Armonk, NY, USA). One‐way ANOVA with Tukey's post hoc test was used to compare differences between groups, and a two‐tailed Student's t‐test was used to compare differences between pairs of groups. IC₅₀ values were calculated by nonlinear regression analysis with a four‐parameter logistic model, and statistical significance was accepted at p < 0.05. Grubbs' test (α = 0.05) was used to identify outliers, and they were removed from the final analysis. Three biological replicates (n = 3) were used for MTT cytotoxicity assays, each with technical triplicates. Two biological replicates (n = 2) were used for oxidative stress (MDA) measurements, each with technical duplicates. Data are presented as mean ± SD.

Author Contributions

Yareeb J. Sahar: data curation, writing – review & editing, validation, visualization, Abbas F. Yasir: writing – original draft, software, formal analysis, data curation, investigation Hussein A. K. Kyhoiesh: supervision, writing – original draft, investigation, conceptualization, methodology, data curation, validation. All authors reviewed the manuscript .

Ethics Statement

The ethical standards have been met.

Conflicts of Interest

The author declare no conflicts of interest.

Code Availability

Not applicable.

Supporting information

Supporting File 1: cbdv71775‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.