Biological activity of curcumin coordination complexes with second and third-row transition metals

1 Introduction

Curcumin is a yellow-orange pigment obtained from the plant Curcuma longa Linn, a rhizomatous monocotyledonous perennial herbaceous plant belonging to the ginger family (Zingiberaceae), known as haldi in Hindi, and recognised by different names worldwide: Ukon, Woolgum, Kunyit, Oendre, Rame, Temu kuning, Temu kunyit, Goeratji, Kakoenji, Koenjet, Kondin, Tius, Kunir, Gianghuang, etc (Goel et al., 2008). Curcuma longa Linn is cultivated in India and tropical countries, predominantly in China and Southeast Asia. The powdered rhizome of this plant, called turmeric, has been used in Asian medicine since the second millennium BC for various disorders and ailments, including abdominal pain, sprains and swelling, anorexia, rheumatism, diabetic wounds, liver disorders, cough, arthritis, wounds, tumours, inflammation, colds, asthma, bronchitis, and jaundice. It is also applied externally for ulcers and skin infections. It has also been used to treat intestinal parasites and as a remedy for poisoning, snakebites, and other ailments. It is also used to preserve, add colour (e.g., yellow 3), and enhance the flavour of foods (Sharma et al., 2005; Fuloria et al., 2022).

Curcumin and turmeric products meet safety standards established by health authorities, such as the Food and Drug Administration (FDA), the Food and Agriculture Organisation of the United Nations (FAO), the World Health Organisation (WHO), and other relevant agencies that oversee food safety. Additionally, they are well tolerated, even at doses of 8 g/day for 3 months and 12 g/day for 3–4 months (Salehi et al., 2019; Esatbeyoglu et al., 2012; Hsu and Cheng, 2007). Turmeric contains essential oils, such as turmerones, ar-turmerone, zingiberene, zingiberol, curcumol, β-curcumene, and xanthorizol, as well as colourants known as curcuminoids, which are responsible for most of its medicinal properties. Besides, it contains sugars, proteins, fibre, and resins (Fuloria et al., 2022; Esatbeyoglu et al., 2012).

Curcumin is a bis-α,β-unsaturated β-diketone, named (1E,6E)-1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione, a hydrophobic polyphenol also called diferuloylmethane (Figure 1). This polyphenol represents the most essential curcuminoid and is isolated from the rhizome of the plant. It also contains demethoxycurcumin, bisdemethoxycurcumin (Jiang et al., 2021; Horosanskaia et al., 2020; Priyadarsini, 2014) and cyclocurcumin (Gasbarri and Angelini, 2024; Kiuchi et al., 1993). Curcumin was first isolated in an impure form in 1815 by Vogel and Pelletier (Vogel and Pelletier, 1815) and obtained in crystalline form in 1870 (Daube, 1870). Subsequently, Lampe et al. confirmed the chemical structure and synthesis in 1910 and 1913, respectively (Miłobȩ;dzka et al., 1910; Lampe and Miłobȩ;dzka, 1913; Gupta et al., 2012). Since then, different investigation groups have reconfirmed its crystalline form (Parimita et al., 2007; Tønnesen et al., 1982). Curcumin shows a keto-enol tautomerism, and the balance between the two forms depends on the polarity and pH of the solvent. The keto-form is major in acid or neutral solutions, while the enol-form is major in alkali solutions (Bernabé-Pineda et al., 2004), resulting in a brilliant yellow hue at pH 2.5 to 7 and a red hue at pH > 7 (Goel et al., 2008). Meanwhile, crystalline curcumin forms intramolecular hydrogen bonds, resulting in the enol form as the dominant configuration (Parimita et al., 2007; Tønnesen et al., 1982), as confirmed by computational studies (Shen and Ji, 2007). The strategic structural development of curcumin, exploring its natural scaffold for diverse therapeutic targets, remains a vibrant area of research (Ahmed et al., 2025). Analytical, spectroscopic and spectrometric data for this compound are given in Table 1.

TABLE 1. Analytical, spectroscopic, and spectrometric data of curcumin (C₂₁H₂₀O₆).

2 Biological studies and bioavailability

The use of curcumin in human diseases was first reported in 1937 in “The Lancet”, one of the most prestigious clinical medical journals (Oppenheimer, 1937). Later, its antibacterial effect and ability to decrease blood sugar levels were described in 1949 and 1972, respectively (Aggarwal and Sung, 2009). In subsequent studies, curcumin was revealed to be the most active secondary metabolite of the four curcuminoids (curcumin, demethoxycurcumin, bisdemethoxycurcumin, cyclocurcumin) (Shah et al., 2022). Moreover, hundreds of investigations have been carried out demonstrating the versatility of curcumin in medicinal chemistry (Sohn et al., 2021; Akbar et al., 2018; Hewlings and Kalman, 2017; Pulido-Moran et al., 2016; Noorafshan and Ashkani-Esfahani, 2013), as antioxidant, anti-inflammatory (Memarzia et al., 2021; Menon and Sudheer, 2007), antiviral (Ardebili, et al., 2021; Jennings and Parks, 2020; Rahban et al., 2020; Mathew and Hsu, 2018), anti-antiproliferative and anti-cancer (Chainoglou and Hadjipavlou-Litina, 2019; Heger et al., 2014; Sa and Das, 2008; Kunnumakkara et al., 2008; Duvoix et al., 2005), antitumoral (Kuttan et al., 2007), cognitive-enhancing and anti-Alzheimer (Abrahams et al., 2019; Meo et al., 2019; Bhat et al., 2019; Lee et al., 2013; Hamaguchi et al., 2010), anti-depressant (Lopresti, 2022; Ramaholimihaso et al., 2020), autoimmune diseases (Mollazadeh et al., 2019; Bright, 2007), antimicrobial (Hussain et al., 2022; Moghadamtousi et al., 2014), insecticidal (Kausar et al., 2024), immunomodulatory activity (Chamani et al., 2022; Abdollahi et al., 2018), diabetes (Quispe et al., 2022; Marton et al., 2021; Meng et al., 2013), cardiovascular diseases (Yang et al., 2024; Miriyala et al., 2007), hepatoprotective (Khan et al., 2019; Farzaei et al., 2018), renoprotective (Trujillo et al., 2013) (Figure 2). The vast diversity of biological activities of curcumin is due to its ability to interact with several molecular targets, including receptors, cytokines, hormones, growth factors, transcription factors (Islam et al., 2024; Patel et al., 2020; Saberi-Karimian et al., 2019; Shishodia, 2013; Zhou et al., 2011), enzymes (Murakami and Ohigashi, 2007; Surh et al., 2001), kinases (Dev et al., 2021; Golonko et al., 2019), and the inhibition of free radicals (Gupta et al., 2020; Calabrese et al., 2008).

The efficacy of curcumin is limited by its poor water solubility, chemical instability, and rapid metabolism, which may be attributed to its hydrophobic nature. Additionally, poor intestinal and hepatic absorption and biotransformation result in very low bioavailability after oral administration (da Silva Lopes et al., 2023; Jankun et al., 2016; Ahmed and Shafqat, 2025). Numerous investigations have been conducted to improve its effectiveness and bioavailability. For example, the use of nanoformulations (Pardeshi et al., 2024; Karthikeyan et al., 2020; Rafiee et al., 2019), curcumin nanoparticles (Chopra et al., 2021; Rai et al., 2020; Kanwal et al., 2023), and the combined administration of curcumin with other molecules (Ciuca et al., 2023; Tabanelli et al., 2021; Liu et al., 2016). adjuvants such as piperine (Heidari et al., 2023), the use of liposomal curcumin (Feng et al., 2017; Li et al., 2005), the use of curcumin phospholipid complex (Mirzaei et al., 2017) and the use of structural analogues of curcumin (Rodrigues et al., 2021; Noureddin and El-Shishtawy, 2019). The benefits reported to these formulations are (a) more extended release, (b) improved cell permeability, and (c) improved resistance to metabolic processes (Hegde et al., 2023; Toden and Goel, 2017). Besides, in the search to improve the bioavailability of curcumin, it has been reported to form complexes with most known metal ions, because its α,β-unsaturated β-diketo group acts as a chelating agent, resulting in stable structures with interesting biological activity (Wanninger et al., 2015; Prasad S. et al., 2021; Pröhl et al., 2016; Mittal et al., 2024; Mondal et al., 2024).

3 Second and third row complexes

The second and third-row transition metals are also known as 4d and 5d transition metals, respectively. These metals exhibit different tendencies and properties than first-row transition metals, or 3d transition metals. Despite being in different periods, they share similar chemical properties because of their similar electron configurations in their outer shells. Second-row atoms are larger and exhibit greater stability in higher oxidation states than first-row atoms due to their greater number of electron shells. The presence of lanthanides causes third-row atoms to be smaller than expected and to have chemical properties similar to those of the second row. Third-row atoms exhibit more complex properties due to the participation of the 5d, 4f, and 6s electrons; however, they tend to form very stable coordination complexes with various ligands. These elements have significant catalytic, industrial, technological, and biological applications (Suárez-Moreno et al., 2022; Gu et al., 2015; Wang et al., 2011).

The selection of second- and third-row transition metals for this review is based on their chemical, structural, and pharmacological advantages for the development of metallodrugs combined with curcumin. These metals form more stable and well-defined complexes than first-row metals, thanks to their larger size and extended d orbitals, which facilitate coordination with ligands such as curcumin and allow for modulation of metal or ligand release in biological media. Furthermore, many have known therapeutic relevance—such as platinum and other heavy metals like ruthenium, iridium, rhodium, and palladium—and electronic or photochemical properties that enable additional applications, such as photodynamic therapy, molecular imaging, and targeted drug delivery. In recent decades, the growing interest in these systems is reflected in the number of reported compounds and publications describing their biological activities, analytical, spectroscopic, and spectrometric characterisation, as well as the three-dimensional structures obtained using SCXRD or DFT calculations (Figure 3). To organise the discussion and figures, the second- and third-row metals are presented according to the number of publications and compounds with potential medical applications, regardless of their position in the periodic table (Figure 4). Therefore, in this work, we analysed the possible biomedical applications of curcumin complexes with second and third-row transition metals. To do so, we used the leading scientific databases SciFinder® and CCDC® (Chemical Abstracts Service, 2026).

3.1 Zirconium and niobium compounds

Khorshidi and Sadeghi (2017) obtained a hexacoordinate zirconium complex (Zr ^**IV** -01) with two curcumin ligands in the trans position. They analysed its antibacterial activity using the well diffusion method, measuring the diameter of the inhibition zone. Nevertheless, this derivative did not show a significant inhibitory effect against Escherichia coli, Staphylococcus aureus, Bacillus subtilis, and Salmonella Enteritidis compared to reference drugs (tetracycline, chloramphenicol, and cefotaxime). The authors note that Zr ^**IV** -01 can be used as a nontoxic, stable orange dye for colouring cotton fibres, as it lacks antibacterial activity. Besides, the antioxidant, antibacterial, and anticancer activities of the niobium(V) binuclear complex were evaluated (Khorshidi and Sadeghi, 2017).

Nb ^**V** -01 showed significant antibacterial activity, especially against E. coli and Pseudomonas aeruginosa (well-diffusion method), comparable to the traditional drugs tetracycline and chloramphenicol. Antioxidant activity assays (DPPH and ABTS methods) revealed radical scavenging activity across various concentrations, equivalent to that of ascorbic acid. Finally, the in vitro cytotoxicity (MTT assay, 48 h) against MCF-7 cancer cells was significantly higher than that against normal fibroblasts (Nowzari and Khorshidi, 2018) (Figure 5).

Research on curcumin complexes with zirconium and niobium provides valuable insights into the coordination chemistry of this natural ligand, particularly revealing the promising biological potential of the niobium(V) derivative. Nevertheless, several methodological aspects could be strengthened to more robustly support the conclusions presented. In particular, the inclusion of quantitative antimicrobial activity assays, mechanistic studies, and broader biological evaluations would enable a more comprehensive assessment of the pharmacological potential of these complexes.

3.2 Technetium and rhenium compounds

Based on the pharmacological activity of curcumin, Sagnou et al. (2011) synthesised three technetium(I)-99m complexes (Tc ^**I** -01 to Tc ^**I** -03), the most widely used radionuclide in nuclear medicine, in addition to their three rhenium(I) analogues (Re ^**I** -01 to Re ^**I** -03). In this series of facial isomerism complexes, the [M(CO)₃]⁺ nuclei (M = rhenium(I) or technetium(I)-99m) were used along with imidazole or isocyanocyclohexane coligands. The rhenium(I) complexes were successfully tested for the selective staining of β-amyloid plaques in Alzheimer’s disease. This work proposes that technetium(I) complexes could be used in radiodiagnostic imaging using single-photon emission computed tomography. The affinity of rhenium(I) complexes for β-amyloid plaques would facilitate the biological evaluation of their technetium(I) analogues as potential radiodiagnostic agents in Alzheimer’s disease (Sagnou et al., 2011).

In 2013, this working group published the neutral complexes of mixed ligands of rhenium(I) (Re ^**I** -04, Re ^**I** -05) and technetium(I) (Tc ^**I** -04, Tc ^**I** -05) using triphenylphosphine as a monodentate ligand. The monophosphine with facial isomerism (Re ^**I** -04, Tc ^**I** -04) and bisphosphine with cis-trans isomerism (Re ^**I** -05, Tc ^**I** -05) complexes showed selective binding to the β-amyloid plaques of Alzheimer’s disease. These complexes introduce a new combination of versatile ligands for developing rhenium(I) and technetium(I)-99m metallodrugs. Moreover, curcumin confirms its pharmacological potential in radiodiagnostic or radiotherapy systems. The development of curcumin complexes enables the development of technetium(I)-99m-based tools with potential diagnostic or therapeutic applications in treating Alzheimer’s disease (Triantis et al., 2013).

In 2024, this research group reported four neutral tricarbonyl complexes using two monodentate co-ligands with anticancer properties: 1-(4-(benzo[d]thiazol-2-yl)phenyl)-3-(methylidyne-azaneyl)propan-1-one (Re ^**I** -06, Tc ^**I** -06) and 1-(4-(benzo[d]thiazol-2-yl)-2-methylphenyl)-3-(methylidyne-azaneyl)propan-1-one (Re ^**I** -07, Tc ^**I** -07). The complexes combine these two ligands with curcumin into a single molecule, making them potential pharmacophores. The technetium(I)-99m complexes exhibited good stability toward cysteine and histidine and high bioavailability, as indicated by their lipophilicity values. These complexes were evaluated for their anticancer properties due to the possible synergy between the two pharmacophores. Additionally, these compounds exhibited suitable photophysical properties across different wavelengths, making them potentially useful for studying cancer cells by confocal microscopy (Stouraitis et al., 2024) (Figure 6).

The development of rhenium(I) and technetium(I)-99m tricarbonyl complexes incorporating curcumin has opened promising avenues in the design of diagnostic imaging probes and multifunctional systems with possible therapeutic relevance. Despite this progress, most investigations remain focused on synthesis and chemical characterisation, while biological evaluation remains at a preliminary stage. The incorporation of quantitative affinity studies, animal models, and mechanistic analyses would be necessary to solidify the potential of these complexes as viable candidates for diagnostic or therapeutic applications.

3.3 Ruthenium compounds

Cisplatin and other platinum(II)-based antitumor drugs are widely recognised for the treatment of various types of cancer, primarily ovarian, testicular, and lung cancer. Nonetheless, other tumours are resistant to platinum(II) complexes. In this regard, Caruso et al. (2012) designed a molecule that incorporates curcumin and a ruthenium(II) p-cymene complex in the form of a semi-sandwich (Ru ^**II** -01). This complex has been extensively investigated due to its potential antitumor activity, as it represents a promising strategy for developing more effective and safer therapies that address both tumour heterogeneity and resistance to conventional treatments. Ru ^**II** -01 demonstrated moderate antiproliferative activity in three tumour cell lines (MTT assay and 72 h of treatment), with moderate IC₅₀ values (HCT116, 13.98 ± 1.503 μM; MCF-7, 19.58 ± 2.367 μM; A2780, 23.38 ± 3.334 μM; CP8, 27.00 ± 2.332 μM). Besides, Ru ^**II** -01 was less effective in two tumour cell lines, U-87 and A549, and one healthy cell line, MCF-10A, with IC₅₀ values of 62.33 ± 8.934, 29.36 ± 1.842, and 102.1 ± 6.5 μM, respectively. In this study, the IC₅₀ values reported for cisplatin (HCT116, 5.217 ± 0.348 μM; MCF-7, 1.835 ± 0.237 μM; A2780, 1.325 ± 0.196 μM; CP8, 9.918 ± 1.155 μM; MCF-10A, 7.27 ± 0.56) were not improved. Additional studies (ESI-MS, SCXRD, DFT) showed that Ru ^**II** -01 binds to a guanine-rich B-DNA decamer, with the metal centre interacting with the N7 of guanine, a feature that may be necessary for its antitumor activity (Caruso et al., 2012).

Based on these findings, a series of complexes isostructural to Ru ^**II** -01 but with various curcuminoids (Ru ^**II** -01a to Ru ^**II** -01f) was reported 5 years later. The antiproliferative activity (MTT assay and 72 h of treatment) of the seven compounds was further evaluated against the cancer cell lines HeLa and HepG2, as well as against the noncancer cell line HEK-293T. As in the previous study, IC₅₀ values for Ru ^**II** -01 were moderate (HeLa, 52.0 ± 1.2 μM; HepG2, 47.0 ± 0.5 μM and HEK-293T, 49.3 ± 0.6 μM) The lowest IC₅₀ values were presented by Ru ^**II** -01f (HeLa, 32.6 ± 0.4 μM; HepG2, 30.7 ± 0.7 μM and HEK-293T, 42.0 ± 1.5 μM), but it failed to improve the values achieved by the reference drug (cisplatin: HeLa, 10.0 ± 0.1 μM; HepG2, 9.8 ± 0.1 μM and HEK-293T, 10.8 ± 0.5 μM). 11.6 ± 0.3 μM) (Li et al., 2017).

Despite the results reported in 2012 and 2017, Garufi et al. (2020) proposed evaluating the transcription factor NRF2 and the pathways altered by the deregulation of the oncosuppressor p53 in several cancer cell lines using Ru ^**II** -01. This study was based on the fact that free curcumin activates NRF2 and p53, which interact to activate oncogenic pathways, increasing resistance to anticancer therapies and tumour progression. Hence, understanding this interaction is essential to improving cancer cells’ response to current therapies. It was observed that Ru ^**II** -01 induced cell death independently of p53. Ru ^**II** -01 also activated NRF2, degraded mutant p53, and activated wild-type p53. NRF2 inhibition increased Ru ^**II** -01-induced cell death in cancer cells harbouring wild-type p53 and mutant p53, but did not increase cell death in p53-null cells. This suggests a crucial role for Ru ^**II** -01 in modulating NRF2, depending on p53 status, for successful cancer cell responses to therapy (Garufi et al., 2020).

Ru ^**II** -01 is hydrophobic in the cellular environment, which makes its efficient delivery to target sites difficultAccordingly, Ru ^**II** -01 was incorporated into hydrophobic compartments of various nanoassemblies, such as liposomes, micelles, and coacervate nanodroplets. Nanoassemblies are relevant in the biomedical field due to their application in delivering hydrophobic drugs. In this project, the authors studied the controlled-release profiles of different nanoassemblies as a function of pH and temperature. Among them, the physicochemical properties, structural adaptability of coacervate nanodroplets, high loading, and minimal drug leakage under physiological conditions make them ideal nanocarriers. The coacervate nanodroplets were fabricated from carbon dots or ATP in the presence of poly(diallyldimethylammonium chloride). The results revealed a controlled, sustained release mechanism via swelling and diffusion in the pH range 7.4–6.0, as well as diffusion and erosion at pH 5. Temperature also influenced the release profiles of Ru ^**II** -01 (Saini et al., 2021) (Figure 7).

Hybrid metallodrugs based on ruthenium and curcumin have attracted attention as potential anticancer agents, owing to the favourable pharmacological profile of ruthenium complexes combined with curcumin’s bioactivity. Although the molecular designs reported are attractive and some mechanistic insights have been explored, the observed cytotoxicity remains moderate, and selectivity is limited. In addition, the need for improved delivery strategies indicates that these compounds still require significant optimisation before being considered competitive therapeutic candidates relative to established drugs such as cisplatin. Future studies integrating structural optimisation, deeper biological evaluation, and in vivo models will be essential to clarify the true pharmacological potential of this class of complexes.

As previously mentioned, free curcumin is nontoxic but has low bioavailability, limiting its potential clinical use. Based on this evidence, multiple attempts have been made to design new curcumin-based compounds to address these challenges. Thusly, Bonfili et al. (2012) developed three ruthenium(II) complexes with different arene groups (p-cymene, Ru ^**II** -01; benzene, Ru ^**II** -02; hexamethylbenzene, Ru ^**II** -03). Viability in HCT116 cells was moderately inhibited at 4 and 24 h of treatment with these ruthenium(II) complexes, which differs from the report by Caruso et al. (2012), who reported comparable values at similar concentrations of Ru ^**II** -01 at 72 h. In addition, the ability of these compounds to regulate proteasomes, which are widely studied pharmacological targets to elucidate the mechanisms of action of anticancer compounds, was explored. Ru ^**II** -02 exhibited greater antioxidant capacity and inhibited proteasomal activity more potently than Ru ^**II** -01, Ru ^**II** -03, or free curcumin, triggering apoptosis. Apoptosis was also confirmed by the accumulation of the proapoptotic proteins p53 and p27, and by cleavage of full-length PARP (poly (ADP-ribose) polymerase), primarily in Ru ^**II** -02 at 24 h of treatment. The lipophilicity of Ru ^**II** -02 explains its effectiveness in remaining in the membrane longer than free curcumin. Although Ru ^**II** -01 showed greater affinity for DNA, in line with previous evidence, caspase-3 activation is consistent with DNA fragmentation primarily for Ru ^**II** -02 (Bonfili et al., 2012; Caruso et al., 2012).

Free curcumin has also shown a marked antihyperglycemic effect, improving insulin sensitivity. This has been attributed to its anti-inflammatory properties and potential interaction with the DPPIV (dipeptidyl peptidase IV) enzyme, which inactivates GLP-1 (glucagon-like peptide 1), a peptide responsible for glucose tolerance. In this way, Antonyan et al. (2014) evaluated the inhibitory activity of Ru ^**II** -01 to Ru ^**II** -03 on the DPPIV enzyme. Ru ^**II** -02 suppressed DPPIV activity more potently (K _*i* = 20.2 ± 0.8 μM) than Ru ^**II** -01, Ru ^**II** -03, or free curcumin. All three complexes exhibited antioxidant activity (DPPH and ABTS methods) comparable to that of free curcumin, effectively preventing glucose-induced oxidative stress. Molecular docking studies and spectrophotometric experiments showed a possible interaction of Ru ^**II** -02 with the S1 and S2 subdomains of DPPIV, with S1 mainly responsible for the enzyme’s substrate specificity (Antonyan et al., 2014).

Pettinari et al. (2018) synthesised eight ruthenium(II) complexes based on the ligands curcumin (Ru ^**II** -04 to Ru ^**II** -07) and bisdemethoxycurcumin (Ru ^**II** -04a to Ru ^**II** -07a), and functionalized η⁶-arene ligands (N,N-dimethyl-1-phenylmethanamine and 1-phenylethan-1-ol). Their RAPTA-type ionic derivatives containing the ligand 1,3,5-triaza-7-phosphaadamantane (PTA) were also obtained. The cytotoxic activity of the complexes, excluding Ru ^**II** -06 and Ru ^**II** -06a due to their instability, was evaluated against the A2780, A2780cisR (both cancerous), and HEK-293 (non-cancerous) cell lines using the MTT assay after 72 h of treatment. Ru ^**II** -02 and Ru ^**II** -04 showed moderate activity, with IC₅₀ values of 29 ± 4, 33 ± 6, and 21 ± 4 μM, and 29 ± 2, 37 ± 6, and 18 ± 3 μM, respectively. Importantly, these results did not improve on the values obtained for the reference drug cisplatin (1.0 ± 0.2, 25 ± 1, 7.3 ± 0.6 μM) (Pettinari et al., 2018b) (Figure 8).

Coordination of curcumin to ruthenium(II) centres can substantially alter its biological behaviour, enabling the exploration of a wide range of pharmacological activities, including anticancer effects, proteasome inhibition, and potential antihyperglycemic activity. Despite the structural diversity investigated, many of the reported complexes exhibit only moderate biological activity and limited cellular selectivity, and in several cases their efficacy does not surpass that of reference drugs such as cisplatin. Moreover, some of the proposed mechanisms of action still lack thorough experimental validation. Consequently, although these systems represent promising platforms for the design of curcumin-based metallodrugs, further structural optimisation and more comprehensive biological studies will be required to establish their true therapeutic potential.

As previously seen, ruthenium(II) complexes containing arene ligands exhibit significant cytotoxic effects in various cancer cell lines. Moreover, RAPTA-C complexes containing PTA ligands have been reported to exhibit superior solubility, cytotoxicity, and selectivity. Therefore, a series of novel ruthenium(II) complexes with p-cymene (Ru ^**II** -08/Ru ^**II** -08a and Ru ^**II** -09/Ru ^**II** -09a) or hexamethylbenzene (Ru ^**II** -10/Ru ^**II** -10a), containing curcumin (Ru ^**II** -08 to Ru ^**II** -10) or bisdemethoxycurcumin (Ru ^**II** -08a to Ru ^**II** -10a), and the PTA ligand were obtained. The antitumor activity of the six compounds was evaluated in vitro using the MTT assay against cancer cells (A2780 and A2780cisR) and healthy cells (HEK-293) after 72 h of treatment. These new compounds were highly cytotoxic, with IC₅₀ values ≤1 μM, surpassing that of cisplatin, the reference drug. Furthermore, some are less cytotoxic to HEK-293 cells, thus maintaining a significant selectivity index. The use of curcumin/bisdemethoxycurcumin or p-cymene/hexamethylbenzene did not significantly affect biological activity; however, the presence of PTA substantially enhances the pharmacological properties of these cationic complexes (Chatterjee et al., 2008; Pettinari et al., 2014).

Subsequently, the interactions of Ru ^**II** -08 and Ru ^**II** -09 with HSA (human serum albumin) were investigated using electrochemical and spectroscopic techniques. Both complexes, but primarily Ru ^**II** -09, quenched HSA’s intrinsic fluorescence in a concentration-dependent manner. Both complexes, primarily Ru ^**II** -08, spontaneously bound to the Sudlow I site of HSA via hydrogen bonding and van der Waals interactions in a 1:1 M ratio. Both complexes, particularly Ru ^**II** -08, altered HSA conformation, disrupted its secondary structure, and impaired its biological activity. These results provide new insights into the design and application of Ru ^**II** -08 and Ru ^**II** -09 in the biomedical field (Huang et al., 2017).

Additionally, dicationic ruthenium(II) complexes containing p-cymene ligands and PTA-Me, as well as curcumin (Ru ^**II** -11 and Ru ^**II** -12) or bisdemethoxycurcumin (Ru ^**II** -11a and Ru ^**II** -12a), have been designed. The antiproliferative activities of the metal complexes were determined against A2780 and A2780cisR (cancerous) cells, as well as HEK-293 (non-cancerous) cells, using the MTT assay after 72 h of incubation. Unlike previously described monocationic PTA complexes (IC₅₀ ≤ 1 μM), which were highly cytotoxic and selective, the PTA-Me complexes exhibited weak cytotoxicity (IC₅₀ ≥ 90 ± 3 μM) and no selectivity (IC₅₀ ≥ 90 ± 25 μM). This low anticancer activity is probably due to its dicationic character, which reduces cellular uptake (Pettinari et al., 2017).

Huang et al. (2019) conducted an in vitro study of the intercalative interactions between three ruthenium(II) complexes (Ru ^**II** -01, Ru ^**II** -01c, Ru ^**II** -09a) and CT-DNA (calf thymus DNA) using spectroscopic techniques, electrochemical methods, and viscosity measurements. The three complexes intercalate into the G-C nitrogen base pairs of CT-DNA through van der Waals interactions and hydrogen bonding. The three complexes did not significantly affect the helicity of the right-handed B form of CT-DNA or the internal structure of deoxyribose. Ru ^**II** -01, with lower steric hindrance and higher polarity, exhibited greater intercalative binding capacity with CT-DNA. These preliminary results suggest molecular mechanisms underlying the interactions between these compounds and DNA, which are helpful for future biological applications (Huang et al., 2019).

Composites were designed to combine the antibacterial activity of curcumin complexes with the porosity of carbon materials. To this end, the Ru ^**II** -01 (neutral) and Ru ^**II** -08 (cationic) complexes were selected based on their toxicity and solubility. Both commercially available activated carbon and activated carbon synthesised from carbon aerogel and/or almond shell were also used to evaluate their physicochemical properties. The homogeneous distribution of the poorly soluble Ru ^**II** -01 and Ru ^**II** -08 on the surface of the highly porous carbon materials prevented their leaching into the solution, resulting in a composite with high adsorption capacity and efficient antibacterial activity. Ru ^**II** -08 (cationic), which is substantially more soluble in water than Ru ^**II** -01 (neutral), was the most toxic and was deposited on the different activated carbon materials by impregnation. The surface of the composites remained free of Gram-negative bacteria (Vibrio fischeri and Escherichia coli), even after several days of exposure to high bacterial concentrations in solution. These results indicate that these hybrid materials could be used in air filters or aqueous treatments (Pettinari et al., 2018a) (Figure 9).

Despite the promising cytotoxic activity reported for several ruthenium(II)–curcumin complexes, the conclusions remain constrained by methodological limitations. The biological evaluation relies primarily on the MTT assay and involves a relatively small number of cancer cell lines, which restricts the robustness of the pharmacological interpretation. In addition, the proposed structure–activity relationships are based on a limited set of compounds, making it difficult to establish definitive correlations regarding the influence of arene ligands, curcuminoid backbones, or PTA coordination on biological behaviour. Moreover, mechanistic interpretations derived from spectroscopic studies of interactions with biomolecules such as DNA or HSA remain largely indirect and require further biochemical validation.

Adding triphenylphosphine derivatives to curcumin complexes with ruthenium(II) can further enhance their therapeutic properties, especially in cancer treatment. Triphenylphosphine derivatives can enhance the stability and solubility of compounds, thereby improving their bioavailability and penetration into cancer cells. Additionally, biotin is a vector that enables the introduction of cytotoxins into cancer cells, providing an alternative to chemotherapy. Based on this, Biancalana et al. (2020) functionalized 4-diphenylphosphinobenzoic acid with biotin to obtain 2-(biotinyloxy)ethyl 4-(diphenylphosphanyl)benzoate, which was subsequently used in the synthesis of Ru ^**II** -14. Moreover, Ru ^**II** -13, lacking biotin, was prepared and used as a reference compound. Ru ^**II** -14 showed suitable stability in a DMSO/water solution after 72 h and slow degradation in cell culture medium. Accordingly, the antiproliferative activity of both compounds was evaluated using the MTT assay after 72 h of treatment against the cancer cell lines A2780 and A2780cisR. The IC₅₀ values in both cell lines were >200 μM, indicating that neither compound exceeded the values observed for cisplatin, Ru ^**II** -01, biotin, and curcumin. The antiproliferative potential of Ru ^**II** -13 and Ru ^**II** -14 was also evaluated against the cancer cell lines COLO-205, HCT116, and SW-620, both in biotin-free DMEM (Dulbecco’s Modified Eagle Medium) and in biotin-rich RPMI 1640 (Roswell Park Memorial Institute 1640) medium. The cytotoxicity of both compounds increased significantly in biotin-free DMEM, with IC₅₀ values lower than those determined in biotin-rich RPMI 1640. Besides, the antiproliferative activity of Ru ^**II** -14 did not correlate with the expression of the biotin transporter SLC5A6. Nevertheless, this compound interacts with a biotin receptor, avidin, so its antiproliferative activity is greater when biotin is absent from the cell culture medium (Biancalana et al., 2020) (Figure 10).

Tröger bases are used in supramolecular scaffolds as chiral heterocyclic fluorophores. Tröger bases exhibit rapid cellular uptake, strong DNA-binding affinity, and the ability to induce apoptosis in cancer cells, thus potentially serving as cellular imaging probes (Murphy et al., 2021; Bright et al., 2024). In this regard, Shanmugaraju et al. (2018) developed a luminescent ruthenium(II) complex (Ru ^**II** -15) containing a Tröger’s base derived from 2-(2-(pyridin-4-yl)ethyl)-4-amino-1,8-naphthalimide. Using confocal fluorescence microscopy, the authors demonstrated that Ru ^**II** -15 rapidly enters the cytoplasm of HeLa cells. Antiproliferative assays against HeLa cells showed that Ru ^**II** -15 is significantly cytotoxic (IC₅₀ = 4.9 ± 0.4 μM, AB assay and 24 h incubation) and exhibits showed good stability in a DMSO/water solution after luminescence. Ru ^**II** -15 also showed good anticancer activity against the HepG2 and HCT116 cell lines, reaching IC₅₀ values of 2.7 and 3.5 μM, respectively. Consequently, Ru ^**II** -15 could be used to develop new therapies (Shanmugaraju et al., 2018).

Mohan et al. (2022) synthesised another Tröger’s base from 2-(4-pyridyl)-4-amino-1,8-naphthalimide, which served as a precursor to a ruthenium(II) complex (Ru ^**II** -16) containing curcumin and p-cymene. Tröger’s base and Ru ^**II** -16 displayed fluorescence due to intramolecular charge transfer. Accordingly, using confocal fluorescence microscopy, cellular uptake of the Tröger’s base and Ru ^**II** -16 into the cytoplasm of cancer cells was demonstrated. Furthermore, Ru ^**II** -16 exhibited rapid cellular internalisation and good cytotoxicity against HeLa, HCT116, and HepG2 cancer cells, as assessed by the AB assay after 48 h of treatment. Nonetheless, the IC₅₀ values (HeLa, 2.0 ± 0.3; HCT116, 2.7 ± 0.1; and HepG2, 1.9 ± 0.1 μM) were much lower than those found for cisplatin (>13 μM), Tröger’s base (>100 μM), and Ru ^**II** -01 (>15 μM). This work demonstrated the versatility of Tröger’s bases as potential precursors for ruthenium(II) metallodrugs with potential therapeutic activity (Mohan et al., 2022) (Figure 11).

Efforts to enhance the biological performance of ruthenium(II) complexes through structural modifications have produced mixed results. In particular, biotin-functionalised derivatives exhibit low antiproliferative activity and show no clear correlation with SLC5A6 transporter expression, suggesting that the proposed cell-targeting mechanism is not convincingly demonstrated. Complexes incorporating Tröger bases, on the other hand, display micromolar cytotoxicity and efficient cellular internalisation. However, their biological activity has been evaluated in only a limited number of cell lines and primarily through single metabolic assays, which limits the ability to draw clear conclusions about their mechanism of action and therapeutic potential.

In 2017, a novel ruthenium(II) complex (Ru ^**II** -17) was published that, in addition to curcumin, contained the ligand 1,4,7-trithiacyclononane, used as a face-coating ligand instead of typical arene-based ligands. The in vitro cytotoxicity of Ru ^**II** -17 against PC-3 (unhealthy) and PNT-2 (healthy) cell lines was assessed using the AB assay after 48 h of treatment, with free curcumin as a reference. Ru ^**II** -17 showed no activity against PNT-2 cells, while free curcumin showed inhibition rates of approximately 65% and 35% at 80 and 50 μM concentrations, respectively. Ru ^**II** -17 did not alter PC-3 cell viability, while free curcumin showed a dose-dependent inhibition of approximately 73% at a concentration of 80 μM. The authors note that Ru ^**II** -17’s high photochemical stability and polarity modify curcumin’s biological activity. Besides, the DNA melting temperature assay showed that salmon sperm DNA in the presence of Ru ^**II** -17 exhibits a higher melting temperature. In contrast, the melting temperature remains unchanged in the presence of free curcumin. The increased DNA melting temperature evidenced the intercalation of small molecules into the DNA double helix (Henriques et al., 2017).

Pettinari et al. (2021) also proposed the neutral ligand 1,4,7-trithiacyclononane as an alternative to the facial protection conferred by arene ligands. According to the authors, incorporating the 1,4,7-trithiacyclononane ligand could improve the stability and solubility of the metal complexes, thereby increasing bioavailability and cellular penetration. To this end, they synthesised two cationic ruthenium(II) complexes with the 1,4,7-trithiacyclononane ligand derived from curcumin (Ru ^**II** -18) and bisdemethoxycurcumin (Ru ^**II** -18a). They subsequently exchanged the DMSO ligand for the PTA ligand, obtaining the compounds Ru ^**II** -19 and Ru ^**II** -19a. The MTS assay evaluated the cytotoxic potency of these compounds against 3 cell lines (A549, MCF-7, and HCT116). IC₅₀ values were obtained for the four compounds (A549, >500 μM; MCF-7 > 84.7 ± 12.8 μM; and HCT116, 68.0 ± 7.6 μM) higher than those found for curcumin (A549, >32.7 ± 5.0 μM; MCF-7 > 10.5 ± 0.4 μM; and HCT116, 7.1 ± 3.2 μM) and bisdemethoxycurcumin (A549, >21.6 ± 9.8 μM; MCF-7 > 8.4 ± 3.3 μM; and HCT116, 11.5 ± 1.8 μM). Furthermore, mass spectrometry studies allowed us to evaluate the interactions of the four complexes with the model protein, hen egg white lysozyme (HEWL). The results revealed relatively low reactivity between the four complexes and HEWL, consistent with their low cytotoxic activity. The low anticancer activity is likely related to the relatively high stability of these metal complexes (Pettinari et al., 2021) (Figure 12).

The incorporation of the 1,4,7-trithiacyclononane ligand has been proposed as a strategy to enhance the stability and solubility of ruthenium(II) complexes. However, the biological data reported for these systems reveal limited or even negligible cytotoxic activity relative to free curcumin. The complexes described exhibit significantly higher IC₅₀ values than the corresponding reference curcuminoids, suggesting that their increased stability may reduce biological reactivity. Furthermore, although some experiments indicate possible interactions with DNA, the absence of significant cytotoxic effects diminishes the biological relevance of these findings and their potential therapeutic applicability.

The planar ligands 2,2′-bipyridine and 1,10-phenanthroline are potential DNA-binding agents through intercalation or groove binding. Therefore, Srivastava et al. (2020) designed two novel ruthenium(II) complexes with curcumin, which contained 2,2′-bipyridine (Ru ^**II** -19) and 1,10-phenanthroline (Ru ^**II** -20) as auxiliary ligands. Moreover, the antimicrobial activity of both octahedral complexes was explored against a panel of ESKAPE pathogens (Escherichia coli, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, and Pseudomonas aeruginosa). The tested compounds exhibited significant inhibitory activity (MIC = 1 μg mL⁻¹) against a wide range of vancomycin-resistant (VRSA) and methicillin-resistant (MRSA) S. aureus strains. Moreover, both complexes exhibited high selectivity for bacterial cells over eukaryotic Vero cells (selectivity index, SI > 80). Ru ^**II** -19 also displayed potent antibiofilm activity (48% reduction at 10× MIC), outperforming levofloxacin and vancomycin against preformed S. aureus biofilms. Ru ^**II** -19 also acted synergistically with other FDA-approved drugs, which could be potentially valuable in combination therapy for treating infections. In vivo studies of murine S. aureus infection showed that Ru ^**II** -19 reduced bacterial counts to levels similar to those achieved with vancomycin, demonstrating its significant antimicrobial potential (Srivastava et al., 2019).

As we have seen, the β-diketone-based scaffold is a common synthetic intermediate as it can act as a chelating ligand for various metals. To evaluate the importance of this scaffold in biological activity, seven β-diketone-based ligands were used to construct ruthenium(II) complexes using 2,2′-bipyridine as the secondary ligand. On the other hand, lipophilicity correlates with cellular uptake. Ru ^**II** -19 was a lipophilic molecule with a log P_O/W = 0.75, placing it within the pharmacological range. The IC₅₀ values for the curcumin derivative (Ru ^**II** -19) against the HL-60 cell line with or without light activation (dark, 0.38 ± 0.007 µM; light, 0.34 ± 0.02 µM) exceeded the activity shown by free curcumin (dark, 4.1 ± 0.09 µM; light, <0.1 µM), cisplatin (dark, 3.1 ± 0.1 µM; light, 3.5 ± 0.6 µM), phenanthriplatin (dark, 1.8 ± 0.4 µM), and the other β-diketone complexes (except for the oxybenzone and avobenzone complexes). Nonetheless, Ru ^**II** -19 was less potent than in DU-145 or Mia PaCa-2 cells (IC₅₀ of 0.29 ± 0.05 and 0.77 ± 0.05 µM), and was surpassed by phenanthriplatin (IC₅₀ of 0.14 ± 0.01 and 0.15 ± 0.01 µM). In vivo toxicity in zebrafish embryos appears to correlate with lipophilicity, as Ru ^**II** -19 was toxic at 25 µM and had a positive log P_O/W value. Consequently, Ru ^**II** -19 displayed promising thermal stability and cytotoxicity, but was toxic in vivo (Ryan et al., 2021).

Renfrew et al. (2013) reported a ruthenium(II) complex based on the ligands tris(2-methylpyridine)amine and curcumin (Ru ^**II** -21). This compound was considerably more stable than its isoelectronic cobalt(III) analogue but showed little curcumin release in solution or in cells. The authors note that this remarkable stability may reduce the need for curcumin administration. In addition, the high IC₅₀ value of Ru ^**II** -21 obtained by MTT assay after 72 h of incubation in the DLD-1 cell line (52 ± 8 µM) relative to free curcumin (13 ± 2 µM) and its cobalt(III) analogue (39 ± 4 µM) indicates that these compounds induce cell death by different mechanisms (Renfrew et al., 2013).

Curcumin and its derivatives could overcome resistance to platinum(II)-based drugs. Moreover, ruthenium(II) as a central nucleus could inhibit tumour growth via multiple mechanisms of cell death. Based on the above, the synthesis of four ruthenium(II) complexes containing curcumin syringaldehyde (Ru ^**II** 20a and Ru ^**II** 20b) and bisdemethoxycurcumin (Ru ^**II** 20c and Ru ^**II** 20d) was reported. Cytotoxicity studies of the four compounds in the A2780 cell line, using carboplatin as a reference, were also reported (MTT assay at 24, 48, and 72 h of treatment). The IC₅₀ values were dependent on treatment time, with the best results found for Ru ^**II** 20b (3.35 ± 0.05 μM), Ru ^**II** 20c (0.50 ± 0.064 μM), and Ru ^**II** 20d (1.32 ± 0.11 μM) after 72 h of incubation, even surpassing the values found for free ligands (>100 μM) and carboplatin (9.52 ± 0.13 μM) in the same experimental conditions. Additionally, the correlation between IC₅₀ values and levels of the transcription factors FGF-2, MMP-9, and NF-κB was determined using principal component analysis (PCA). Ru ^**II** 20c exhibited the most remarkable cytotoxicity, and Ru ^**II** 20b was the most effective in targeting cell death mechanisms through MMP-9 and NF-κB signalling. Accordingly, the activity of all four complexes depended on treatment duration, as evidenced by the delayed modulation of MMP-9 and FGF-2. However, the consistent decrease in NF-κB following treatment with Ru ^**II** 20a and Ru ^**II** 20b demonstrated the compounds’ modulatory capacity across all exposure stages (Leskovská et al., 2022) (Figure 13).

A number of ruthenium(II) complexes described in these studies demonstrate noteworthy biological activities, including antimicrobial effects and micromolar cytotoxicity. Nevertheless, the interpretation of their therapeutic potential remains preliminary. In many cases, biological activity has been evaluated mainly through in vitro metabolic assays and a limited set of cellular models, which complicates the establishment of firm conclusions regarding their mechanisms of action and actual selectivity. In addition, certain compounds present important drawbacks, such as in vivo toxicity associated with excessive lipophilicity or overly high stability that may hinder the release of the active ligand. These factors highlight the need for more comprehensive mechanistic and pharmacological investigations.

3.4 Rhodium and iridium compounds

Although rhodium(III) complexes have shown little biological activity, their low toxicity and slow ligand release make them suitable systems for drug delivery. Hence, in 2017, a series of rhodium(III) complexes derived from pentamethylcyclopentadienate were reported. Among them, Rh ^**III** -01 stood out, which contains curcumin as a chelating agent. This compound exhibited cytotoxicity similar to that of free curcumin in A549 cells after 72 h of treatment (MTT assay). However, at 24 h, real-time cell migration assays showed a slow release of curcumin and lower activity. The other complexes containing ethylenediamine, acetylacetonate, or 1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane were inactive in cytotoxicity assays and proapoptotic and antimetastatic behaviour. X-ray fluorescence microscopy confirmed that Rh ^**III** -01 efficiently released curcumin into cells, while caspase activity was consistent with pre-apoptotic activity. Nevertheless, the authors note that curcumin-based compounds such as Rh ^**III** -01 promote cell proliferation and migration at low (non-cytotoxic) concentrations, which could increase metastasis, requiring higher doses to induce apoptosis (Markham et al., 2017).

The anticancer activity of curcumin-based half-sandwich organometallic complexes is widely investigated. In this regard, complexation studies of the [Rh(η⁵-C₅Me₅)(H₂O)₃]²⁺ derivative with curcumin led to [Rh(η⁵-C₅Me₅)(curcumin)(H₂O)]⁺. The stability of this species in solution was compared with that of the acetylacetone complex [Rh(η⁵-C₅Me₅)(acetylacetone)(H₂O)]⁺, which uses a simpler β-diketone ligand. The formation constant of [Rh(η⁵-C₅Me₅)(curcumin)(H₂O)]⁺ revealed stability similar to that of the acetylacetone complex. In this project, single crystals of Rh ^**III** -01 were also obtained by the reaction of deprotonated curcumin with sodium methoxide and the organometallic dimeric precursor [Rh(η⁵-C₅Me₅)(µ²-Cl)]₂. The in vitro cytotoxicity (MTT assay after 24 h) of free curcumin (IC₅₀ = 39.6 ± 2.4 µM) in the multidrug-resistant human colon adenocarcinoma cell line (COLO-320/MDR-LRP) was not enhanced by the formation of [Rh(η⁵-C₅Me₅)(curcumin)(H₂O)]⁺ (IC₅₀ = 34.8 ± 0.2 µM). Upon interaction with the HSA protein, the decomposition of the metal complexes and the release of the ligands were promoted by the binding of the metal (Mészáros et al., 2019).

Compared with platinum- or ruthenium-based metallodrugs, the study of the anticancer properties of rhodium and iridium complexes is underdeveloped. Consequently, a series of organometallic complexes of rhodium(III) and iridium(III) were designed with curcumin (Rh ^**III** -01 and Ir ^**III** -01) and several other curcuminoids (Rh ^**III** -01a-f and Ir ^**III** -01a-f). The in vitro activity of all compounds was evaluated against two human cancer cell lines (HepG2 and HeLa) and one healthy human cell line (HEK-293T). Rh ^**III** -01f exhibited significant anticancer activity (HepG2, IC₅₀ = 6.0 ± 0.2 µM and HeLa, IC₅₀ = 6.7 ± 0.9 µM) and remarkable selectivity (HEK-293T, IC₅₀ = 53.8 ± 6.0 µM) against the cell lines studied (MTT assay, 72 h of treatment). Agarose gel electrophoresis studies allowed us to evaluate the interaction of Rh ^**III** -01f with DNA, which was similar to that of cisplatin. These results support the design of rhodium(III) and iridium(III) curcuminoid-based complexes that could be useful in chemotherapy (Su et al., 2017).

In the same vein, a series of cationic half-sandwich complexes of rhodium(III) and iridium(III) were reported, containing 1,3,5-triaza-7-phosphaadamantane, as well as curcumin (Rh ^**III** -02 and Ir ^**III** -02) and bisdemethoxycurcumin (Rh ^**III** -02a and Ir ^**III** -02a) as bidentate chelators. The X-ray crystal structures of Rh ^**III** -02, Rh ^**III** -02a, and Ir ^**III** -02a confirmed the expected “piano stool” geometry. Except for Rh ^**III** -02, the remaining complexes were stable under pseudophysiological conditions and exhibited moderate cytotoxicity in A2780, A2780cisR, and HEK-293 cells, but lacked selectivity. Furthermore, IC₅₀ values in the same cell lines obtained by MTT assay after 72 h of treatment indicated that these compounds were less active than their free ligands (curcumin and bisdemethoxycurcumin) and their precursors Rh ^**III** -01, Ir ^**III** -01, Rh ^**III** -01c, and Ir ^**III** -01c (Pettinari et al., 2015) (Figure 14).

Curcumin-based rhodium(III) and iridium(III) complexes have attracted attention due to the relatively low toxicity of these metals and their potential use as drug delivery platforms. However, the reported biological results reveal several important limitations. In many cases, metal coordination does not significantly enhance cytotoxic activity compared with free curcumin, and some complexes even display lower activity or poor cellular selectivity. Moreover, the slow ligand release and the potential stimulation of cell proliferation at low concentrations raise concerns about their therapeutic efficacy. These observations indicate that more detailed mechanistic and pharmacological investigations are required to validate their potential as metallodrugs.

Curcumin exhibits chemopreventive properties against various tumours, but has low bioavailability. Consequently, Colombo et al. (2019) reported two new biscyclometallated complexes, containing curcumin (Ir ^**III** -03) and tetrahydrocurcumin (Ir ^**III** -03a) bidentately coordinated to iridium(III). Both complexes were rapidly absorbed by various human tumour cell lines (A549, HeLa, and T-89G) from micromolar solutions. Still, they showed weak cytotoxicity in the dark (MTT and CCK-8 assays after 48 h of treatment). According to time-lapse microscopy, Ir ^**III** -03a was photocytotoxic upon irradiation with visible light (2 s, 465–495 nm), inducing apoptosis after 2 h. Ir ^**III** -03a was luminescent in deoxygenated solution, generating singlet oxygen under aerated conditions, while Ir ^**III** -03 exhibited other non-radioactive decay pathways. The results showed that Ir ^**III** -03a could be used for photodynamic therapy (PDT) in combination with emission imaging microscopy. The studies demonstrated that Ir ^**III** -03a readily penetrated cells, concentrated in mitochondria, even at low concentrations, and was photocytotoxic after light irradiation, but not in the dark (Colombo et al., 2019).

On the other hand, the detection and inhibition of amyloid-β (Aβ) peptide aggregates are essential for the diagnosis and treatment of Alzheimer’s disease (AD). Therefore, Huang et al. (2021) reported a series of luminescent iridium(III) complexes based on 2-phenylbenzothiazole. Most of the metal complexes contained acetylacetone in their structure, except for Ir ^**III** -04, which was coordinated to curcumin. This compound penetrated the blood-brain barrier and stained amyloid plaques in the brains of AD mice after 10 days of treatment. Furthermore, Ir ^**III** -04 inhibited Aβ aggregation in vitro, reduced Aβ-induced cytotoxicity in living cells, and activated fluorescence upon binding to Aβ fibrils and oligomers. Hence, Ir ^**III** -04 could be used in vivo to label Aβ and inhibit its aggregation (Huang et al., 2021) (Figure 15).

Iridium(III) complexes incorporating curcumin have been explored for applications ranging from photodynamic therapy to diagnostic tools for neurodegenerative diseases. Despite these promising approaches, their therapeutic applicability remains limited. In photoactive systems, cytotoxic activity depends strictly on light irradiation, restricting their use to specific clinical contexts. Additionally, although certain complexes can cross the blood–brain barrier and detect amyloid-β aggregates, these findings are largely based on preliminary in vitro studies or animal models. Consequently, further investigations are necessary to confirm their efficacy and safety in more complex biological systems.

3.5 Palladium compounds

In 1998, the palladium(II) complex with curcumin was reported for the first time, along with its isostructural analogues of nickel(II), copper(II), and zinc(II). In addition to curcumin derivatives, metal complexes with other 1,7-diaryl-1,6-heptadiene-3,5-diones or 1,7-diarylheptanoids were also obtained. Overall, the metal complexes showed no significant activity against any of the microorganisms tested using the disk diffusion technique, as measured by inhibition zones. Interestingly, Pd ^**II** -01 exhibited greater bactericidal activity than PdII-01a-d against Aspergillus niger, Staphylococcus aureus, and Escherichia coli. This is likely due to the lack of significant interactions with the microorganisms’ cells (Krishnankutty et al., 1998). Twenty years later, Bhange et al. reported the use of Pd ^**II** -01 as an efficient and novel catalyst for forming carbon-carbon bonds in Heck and Suzuki coupling reactions. The application of Pd ^**II** -01 in organic catalysis is one of the first examples of using a natural ligand, such as curcumin, to obtain new molecules of biological interest (Bhange et al., 2018).

Conjugation of hexadecyl 2-phenylquinoline-4-carboxylate, palladium(II) acetate, and a curcumin derivative allowed the synthesis of Pd ^**II** -02 and Pd ^**II** -02a. For Pd ^**II** -02, pure curcumin was used and for Pd ^**II** -02a, 1,7-bis(4-(docosyloxy)-3-methoxyphenyl)hepta-1,6-diene-3,5-dione was used. The authors mention that the combination of two different ligands around palladium(II) enhances the cytotoxicity of metallodrugs. Treatment of DU-145 cells with Pd ^**II** -02 showed greater cytotoxicity than pure curcumin, while Pd ^**II** -02a exhibited a similar percentage of proliferation inhibition (MTS colourimetric test, 72 h exposure). In Pd ^**II** -02, the presence of the cyclopalladate ring prevents rapid dissociation in solution, favouring the compound’s ability to reach pharmacological targets. Meanwhile, in Pd ^**II** -02a, the activity is dominated by the presence of the aliphatic chains (Pucci et al., 2007; Fatima et al., 2008).

This research group reported compounds Pd ^**II** -03 and Pd ^**II** -03a, in which 2-phenylquinoline is functionalised with a terminal chiral substituent derived from a cholesteryl ester. The corresponding mononuclear complexes contain the same cucuminoids as Pd ^**II** -02 and Pd ^**II** -02a. Pd ^**II** -03 showed promising anticancer activity in a preliminary MTS assay (72 h of treatment) against two human prostate cancer cell lines (DU-145 and LnCaP). According to the authors, these multifunctional biomaterials simultaneously contain both the delivery component and the active ingredient, enabling the development of new, potentially effective anticancer therapies (Pucci et al., 2008) (Figure 16).

Palladium(II) complexes derived from curcumin have been investigated for both catalytic applications and potential biological activity. Nevertheless, the available evidence supporting their therapeutic relevance remains limited. In several reports, the observed antimicrobial or cytotoxic effects are only moderate and often comparable to those of free curcumin. Furthermore, most biological evaluations rely primarily on preliminary in vitro assays. As a result, more comprehensive mechanistic and pharmacological studies are needed to determine whether these systems are viable candidates for metallodrug development.

Li et al. (2018) reported eight palladium(II) complexes with different curcuminoids and evaluated their antitumor activity against A549, HeLa, and MCF-7 cells using the MTT assay after 48 h of incubation. Pd ^**II** -04c, Pd ^**II** -04e, and Pd ^**II** -04g with methoxy substituents showed greater antitumor activity than cisplatin and the compounds Pd ^**II** -04, Pd ^**II** -04b, Pd ^**II** -04d, and Pd ^**II** -04f with hydroxy substituents. The DCFDA assay showed that the Pd ^**II** -04c, Pd ^**II** -04e, and Pd ^**II** -04g complexes cause a strong ROS (reactive oxygen species) induction. Nonetheless, the presence of NAC (N-acetylcysteine) reduces ROS generation and improves the survival of HeLa cells treated with Pd ^**II** -04g, indicating that the antitumor effect is regulated by oxidative stress. Mechanistic studies show that Pd ^**II** -04g alters mitochondrial membrane potential, arrests the cell cycle in the S phase, and induces tumour cell apoptosis through a ROS-dependent pathway (Li et al., 2018).

Valentini et al. (2009) synthesised Pd ^**II** -05, which, in addition to pure curcumin, had 4,4′-dinonyl-2,2′-bipyridine as its second biologically active ligand. Pd ^**II** -05 inhibited cell growth in different hormone-independent human prostate cancer cell lines (DU-145, LnCaP, and PC-3) but not in the hormone-dependent cell line (LnCaP-SF). Pd ^**II** -05 induced apoptosis in prostate cancer cells through ROS production, JNK phosphorylation, decreased GSTP1, GSH depletion, and mitochondrial membrane depolarisation associated with decreased Bcl-2 and increased Bax. Furthermore, Pd ^**II** -05 did not activate PARP and was significantly more effective than pure curcumin. Pd ^**II** -05 is a highly stable species with enhanced biological activity, greater bioavailability, improved absorption, distribution, metabolism, excretion, etc. Therefore, Pd ^**II** -05 could aid in the development of molecules that circumvent GSH/GSTp-related detoxification, which is one of the causes of androgen-independent proliferation in prostate cancer and drug resistance (Valentini et al., 2009).

Karami et al. (2017) reported a palladium(II) organometallic complex (Pd ^**II** -06) containing two bioactive ligands (benzylamine and curcumin). Pd ^**II** -06 interacted with CT-DNA and BSA, cooperatively binding to DNA, likely via intercalation. Pd ^**II** -06 displaced MB from the MB-DNA complex, confirming intercalation as the most likely DNA-binding mode. Pd ^**II** -06 bound BSA at Sudlow site I in subdomain II via a static suppression mechanism. Pd ^**II** -06 displayed selective cytotoxicity against JURKAT and MCF-7 cell lines relative to normal peripheral blood mononuclear cells. The above confirms the correlation between DNA binding and the cytotoxicity of Pd ^**II** -06, indicating that this compound could be a promising candidate as an anticancer metallodrug (Karami et al., 2017) (Figure 17).

Some palladium(II) complexes containing curcumin exhibit notable cytotoxicity and suggest biological mechanisms involving oxidative stress or DNA interactions. However, the interpretation of these results is constrained by methodological limitations. Many studies rely on a restricted number of cellular models and predominantly on in vitro assays. In addition, the proposed mechanisms of action are often inferred indirectly from measurements of ROS generation or spectroscopic studies of biomolecular interactions. Consequently, more detailed biological and pharmacological investigations will be necessary to establish their efficacy, selectivity, and mechanism of action in more complex biological systems.

3.6 Platinum compounds

Curcumin alone does not reach target cells due to its low bioavailability and hydrolytic instability, whereas cisplatin undergoes rapid, uncontrolled hydrolysis. These drawbacks limit the chemotherapeutic applications of both cisplatin and curcumin. In this regard, Pt ^**II** -01, a platinum(II) diamine complex with curcumin, has been studied as a photoactivated chemotherapeutic agent. This compound is believed to release curcumin and platinum(II) diamine species, which are activated upon irradiation with visible light. Pt ^**II** -01 exhibited apoptotic photocytotoxicity in cancer cells (HaCaT, BT-474, T47D, and Hep3B) with IC₅₀ values between 12 and 18 µM (MTT assay, 4 h preincubation, 400–700 nm light, 10 J cm^–2). Furthermore, Pt ^**II** -01 remained passive in the dark with IC₅₀ values greater than 200 µM. Additionally, Pt ^**II** -01 interacted with CT-DNA and 5′-GMP, indicating the formation of platinum(II)-linked DNA adducts upon exposure to visible light. Platinum(II) titration and the Comet assay suggested the formation of Pt-DNA crosslinks. Fluorescence microscopy images showed the cytosolic localisation of curcumin, suggesting the potential for both chemotherapeutic and phototherapeutic action. Therefore, in vitro photodegradation of Pt ^**II** -01 allows the controlled generation of platinum(II) diamine species that inhibit DNA crosslinking/transcription and the release of curcumin, a PDT agent, at the site of action under visible light (Mitra et al., 2015).

Currently, controlled drug delivery systems used in cancer therapy offer a promising alternative. Especially those based on poorly soluble nanoparticles, which act directly on cancer cells, thereby increasing their bioavailability. In this regard, De Castro et al. (2020) prepared nanocolloids based on chitosan-pectin polysaccharides and used them as delivery systems for Pt ^**II** -01. Nanocolloids measuring approximately 100 nm were prepared via ultrasonication and layer-by-layer technology. The nanocolloids, highly stable in water, were activated by visible light, thereby increasing anticancer activity in HeLa cells through drug release. Light-activated Pt ^**II** -01 nanocolloids showed an IC₅₀ value in the nanomolar range (50 nM), while dark-activated Pt ^**II** -01 nanocolloids showed an IC₅₀ value in the micromolar range (75 μM). NMR metabolomic analysis revealed that light-activated Pt ^**II** -01 nanocolloids induced metabolic alterations comparable to the combined cisplatin and curcumin treatment. Hence, the hypothesis that photoactivation of Pt ^**II** -01 nanocolloids induces the release of curcumin and cisplatin species was confirmed. Light-activated Pt ^**II** -01 nanocolloids induce a cellular metabolic stress response due to a combined synergistic effect of cisplatin and curcumin (De Castro et al., 2020).

Additionally, a series of cationic platinum(II) complexes based on curcumin and triflate counterions were designed. The main difference between the compounds was the secondary ligand used: propane-1,3-diamine (Pt ^**II** -02), pyridin-2-yl-methanamine (Pt ^**II** -03), 2,2′-bipyridine (Pt ^**II** -04), and 4,4′-dinonyl-2,2′-bipyridine (Pt ^**II** -05). Pt ^**II** -02 to Pt ^**II** -04 complexes addressed the main disadvantages of free curcumin, including low bioavailability, photodegradability, and rapid oxidation. However, Pt ^**II** -05 underwent only partial degradation. Pt ^**II** -02 to Pt ^**II** -05 complexes did not exhibit cytotoxic activity against MDA-MB-231 cells (IC₅₀ > 200 µM). The antioxidant properties were evaluated spectrophotometrically, and no changes were observed after coordinating free curcumin with platinum(II), nor were there any differences related to the presence of the secondary ligand. Experimental and theoretical studies determined the mechanism of antioxidant action. The in vitro inhibitory activity of bacterial growth against Staphylococcus aureus and Escherichia coli showed that Pt ^**II** -02 to Pt ^**II** -05 complexes had greater activity than free curcumin (disk diffusion technique). Bacterial growth inhibition was enhanced by increasing aromaticity and lipophilicity with secondary ligands. Consequently, careful selection of secondary ligands enables modulation of the lipophilic properties of platinum(II) complexes, which are necessary to enhance antibacterial activity (Caligiuri et al., 2022).

As we have seen, curcumin degrades rapidly in biological media, exhibiting low bioavailability. On the other hand, cisplatin is labile and generates nonspecific interactions as it travels to target cells. In this regard, Mitra et al. (2017) synthesised three platinum(II) complexes based on curcumin. As in the previous report, they modified the auxiliary ligands: in Pt ^**II** -01, they used two molecules of ammonia; in Pt ^**II** -06, they used ethylenediamine; and in Pt ^**II** -07, they used 1R,2R-(−)-1,2-diaminocyclohexane. Pt ^**II** -01 exhibits photodegradation upon exposure to visible light, as previously reported by this research group. This property of Pt ^**II** -01 led to the controlled generation of platinum(II) diamine species in Pt ^**II** -06 and Pt ^**II** -07, which were used as DNA crosslinkers, and to the release of curcumin, a potential photosensitising agent. Pt ^**II** -01 was the most photocytotoxic complex, with IC₅₀ values ranging from 13 to 19 μM in the HaCaT, BT-474, T-47D, and Hep3B cell lines (MTT assay, 4 h preincubation, 400–700 nm light, 10 J cm^–2). Pt ^**II** -01 also showed 15-fold lower cytotoxicity in the dark (IC₅₀ > 200 μM), a property essential for the search for potential PDT agents. Pt ^**II** -06 and Pt ^**II** -07 were less photocytotoxic, with IC₅₀ values ranging from 22 to 60 μM under the same experimental conditions, likely due to their rigid diamine structure, which reduced cellular uptake. The IC₅₀ values in the standard human cell line (HPL1D, IC₅₀ = 33–57 μM) demonstrate the three complexes’ selective activity toward cancer cells. Furthermore, all three complexes formed covalent platinum(II)-DNA adducts upon exposure to light and treatment with CT-DNA, and cytoplasmic localisation and ROS-mediated apoptotic cell death were observed in visible light. In this work, the authors demonstrated that these complexes can deliver curcumin and a platinum(II)-diamine species into cancer cells and monitor their uptake using visible light (Mitra et al., 2017) (Figure 18).

Curcumin-based platinum(II) complexes have been proposed as promising strategies to overcome some of the limitations associated with both free curcumin and conventional platinum drugs such as cisplatin. In particular, approaches involving photochemical activation or controlled-release systems have attracted considerable attention. Nevertheless, much of the available evidence is still derived from in vitro experiments. In addition, the cytotoxic effects often depend on specific experimental conditions, such as irradiation with visible light, which may restrict their clinical applicability. Consequently, further studies evaluating stability, selectivity, and efficacy in more complex biological models are necessary to clarify their real therapeutic potential.

In 2019, three platinum(II) complexes were synthesised, two containing curcumin (Pt ^**II** -08 and Pt ^**II** -09) and one with caffeine. The cytotoxic activity of these compounds was preliminarily evaluated using the MTT assay against five cancer cell lines (A375, A549, MCF-7, SK-OV-3, and SW-620) and one non-cancer cell line (MCF-10A). Pt ^**II** -08 and Pt ^**II** -09 were moderately cytotoxic and selective after 24 h of incubation with the cells, especially against MCF-7, SW-620, and MCF-10A. Pt ^**II** -09 presented IC₅₀ values in the low micromolar range (MCF-7, IC₅₀ = 2.5 ± 0.3 μM; SW-620, IC₅₀ = 4.1 ± 1.0 μM; and MCF-10A, IC₅₀ = 4.3 ± 1.9 μM). Furthermore, Pt ^**II** -08 and Pt ^**II** -09 can be activated by visible-light irradiation, releasing curcumin. Interaction studies with DNA revealed distinct behaviours. Pt ^**II** -08 tends to intercalate between nucleobase pairs, inducing oxidative degradation, while Pt ^**II** -09 binds in grooves through electrostatic forces. These complexes enabled photosensitization, generating a more efficient system surpassing its precursors, curcumin and cisplatin, in cytotoxicity and selectivity. Moreover, preincubation studies in the cellular medium reduced the effectiveness of the coordination complexes (Censi et al., 2019) (Figure 19).

Along the same lines, Upadhyay et al. (2019) designed three platinum(II) complexes with mixed ligands. 1,10-Phenanthroline (Pt ^**II** -10, Pt ^**II** -10a) and 5-biotin-1,10-phenanthroline (Pt ^**II** -11) were used as photocytotoxic agents targeting cancer cells, while curcumin (Pt ^**II** -10, Pt ^**II** -11) and diglucosyl-curcumin (Pt ^**II** -10a) were used as anticancer drugs and photosensitizers. Photocytotoxicity was assessed using an MTT assay with HeLa, HepG2, MDA-MB-231, A549, and normal HPL1D cells. These complexes showed photoinduced apoptotic cell death under visible light (IC₅₀ = 6.2–27.4 µM, 400–700 nm) and minimal cytotoxicity in the dark (IC₅₀ = 43.9–95.2 µM), mainly in HeLa, HepG2, and A549 cancer cells. Besides, the complexes were active in MDA-MB-231 cells resistant to several drugs (IC₅₀ = 8.6–25.0 µM). Pt ^**II** -10a and Pt ^**II** -11, which contain biotin and glycosyl groups, exhibited greater selectivity for cancer cells than for normal cells, making them potential photochemotherapeutic agents. The formation of ROS and the mechanism of apoptotic cell death were studied using the DCFDA and Annexin-V/FITC-PI assays, respectively. ICP-MS studies demonstrated their preferential localisation in the cell nucleus. The complexes showed a high propensity for DNA binding and light-activated ROS generation, as indicated by DNA photocleavage studies (Upadhyay et al., 2019).

Four platinum(II) complexes were prepared, each containing 4,4′-bisdimethoxyazobenzene as the primary ligand and acetylacetone derivatives with methyl, pyrenyl, and anthracenyl substituents, as well as curcumin (Pt ^**II** -12), as auxiliary ligands. These organometallic azo complexes exhibited activity in PDT, with IC₅₀ values of 8.0–14.2 μM under visible light (400–700 nm, 10 J cm⁻²) in HaCaT cells (MTT assay), whereas they were less cytotoxic in the dark (IC₅₀ values of 45.0–61.9 μM). The photoactive base β-diketonate acts as a photocytotoxin and exhibits photochemotherapeutic activity, while the platinum(II) species acts as a transcription inhibitor. The complexes photocleave plasmid DNA, forming 75% nicked circular DNA, and generate singlet oxygen and hydroxyl radicals as ROS, which cause apoptotic cell death under visible light. The pyrenyl and anthracenyl complexes are appropriate DNA binders with intercalative properties, surpassing the activity of Pt ^**II** -12 (Raza, et al., 2016) (Figure 20).

A number of platinum(II) complexes derived from curcumin exhibit micromolar cytotoxicity and interesting photoactive properties. Despite these encouraging results, most biological data originate from in vitro studies and conventional metabolic assays. In many cases, the observed activity is strongly dependent on irradiation with visible light, which limits the range of possible therapeutic applications. Additional investigations focusing on stability, selectivity, and performance in more complex biological systems will therefore be essential to determine their actual therapeutic relevance.

3.7 Silver and gold compounds

Syed et al. (2015) reported the synthesis of Ag ^**I** -01 and preliminary antibacterial activity studies against four strains (Staphylococcus aureus, Escherichia coli, Bacillus subtilis, and Bacillus cereus). Results obtained using the agar well diffusion method showed that curcumin inhibited all strains tested, especially S. aureus. Contrary to expectations, Ag ^**I** -01 did not exhibit greater activity than free curcumin. The authors state that the exchange of protons for silver(I) ions on the phenolic rings of curcumin could have decreased the antibacterial activity. Based on these findings, the authors presented biologically improved models as future recommendations (Syed et al., 2015).

Similarly, Au ^**I** -01 was synthesised and used in anti-arthritic studies in rats, evaluating its ability to inhibit inflammation in the contralateral paw by measuring paw volume, anatomical changes in the affected limb, comparing radiographs, and serum acid phosphatase levels during the disease, as well as serum gold levels at different doses. The results obtained in this animal model on the second day showed that Au ^**I** -01 (30 mg/kg) inhibited paw inflammation by 68.75%, while aspirin (200 mg/kg) inhibited it by 49.56%. Nonetheless, aspirin at the same concentration as AuI-01 (30 mg/kg) did not provide any protection. Radiographs of the affected areas showed destructive lesions in the ankle joints and tarsal and metatarsal bones, as well as periosteal changes. The inflammation of the surrounding tissue decreased with Au ^**I** -01 treatment. The reduction in serum acid phosphatase levels in the Au ^**I** -01-treated group was highly significant, suggesting that this compound may act by inhibiting lysosomal acid phosphatase. This was not observed in animals treated with aspirin, indicating that the two compounds have different mechanisms of action. Serum gold levels, measured by atomic absorption spectrometry, showed that Au ^**I** -01 is absorbed in the gastrointestinal tract and is dose-independent (Sharma et al., 1987) (Figure 21).

Metal complexes based on silver(I) and gold(I) have also been investigated in combination with curcumin for their potential biological activities. However, the reported results reveal several important limitations. In the case of the silver(I) complex, coordination with curcumin did not enhance antibacterial activity compared with the free ligand, suggesting that metal binding may alter functional groups responsible for biological activity. The gold(I) complex, on the other hand, demonstrated anti-inflammatory effects in an animal model. Nevertheless, the available evidence remains limited, with few detailed mechanistic and toxicological studies, highlighting the need for further investigation before therapeutic relevance can be established.

3.8 Cadmium and mercury compounds

Cd ^**II** -01 and Hg ^**II** -01 were prepared along with their isostructural lead(II), tin(II), and calcium(II) derivatives. In all of them, curcumin behaved as a monobasic bidentate ligand in which the aforementioned metal ions replaced the enol proton. Antibacterial (Bacillus cereus) and antifungal (Aspergillus flavus, Aspergillus heteromorphus, Aspergillus niger, and Penicillium verruculosum) studies were conducted using the well-diffusion method. The metal complexes exhibited higher activity than free curcumin and the reference drug (nystatin). Interestingly, Hg ^**II** -01 was the most active against A. flavus, although the tin(II) derivative showed the most significant activity against the remaining microorganisms (Pallikkavil et al., 2013).

Similarly, Cd ^**II** -02 and its isostructural copper(II) and zinc(II) derivatives were obtained. The cytotoxic activity of free curcumin and its three binuclear hydroxo complexes was investigated in vitro. The cancer cell lines chosen were HCT116, HepG2, HeLa, and MCF-7 (SRB assay). Interestingly, free curcumin was more effective as an anticancer agent than its metal complexes, matching the activity of the reference drug doxorubicin. The authors mention that the presence of an oxo-lating source during the synthesis of the metal complexes inhibits the synergistic effect between curcumin and the metal centre, decreasing curcumin’s reducing potential. Furthermore, they suggest that the size of the coordinated metal ion could influence the cytotoxic activity of the metal complex [IC₅₀ value, copper(II) 0.57 Å < zinc(II) 0.60 Å < cadmium(II) 0.78 Å] by blocking the central carbon atom of curcumin, which contains a labile hydrogen (Khalil et al., 2014) (Figure 22).

Some curcumin-derived metal complexes have shown enhanced antimicrobial activity relative to the free ligand. Even so, these results should be interpreted cautiously because the biological evaluation is limited and often relies on qualitative techniques such as agar diffusion assays. In cytotoxicity experiments, free curcumin frequently exhibited greater activity than its corresponding metal complexes, suggesting that coordination with certain metals may reduce its biological effectiveness. More detailed studies analysing SAR and the underlying mechanisms are therefore required to explain these differences.

3.9 Rare earth compounds and others

In recent years, curcumin-derived metal complexes have attracted considerable interest among scientists for the design of useful metallodrugs for the treatment of various diseases. Subhan et al. (2014) used both experimentally prepared curcumin and curcumin extracted from turmeric to synthesise a series of rare-earth complexes, including yttrium(III), lanthanum(III), cerium(III), and europium(III). In addition to transition metal complexes, such as chromium(III) and palladium(II). Curcumin metal complexes (Cr ^**III** -01, Y ^**III** -01, La ^**III** -01, Ce ^**III** -01, Eu ^**III** -01, Pd ^**II** -01) were characterised by IR and UV–Vis. Curcumin, in its deprotonated form, coordinated bidentately with these metal ions. The antibacterial study by the disk diffusion method indicated that these complexes exhibited moderate antibacterial activity against Klebsiella pneumoniae, Proteus spp., Pseudomonas sp., and Escherichia coli. The DNA-binding study using agarose gel electrophoresis showed that these complexes interacted with genomic DNA. Eu ^**III** -01 showed appropriate DNA cleavage activity, while Cr ^**III** -01 exhibited the best antibacterial activity. Y-01 and Pd ^**II** -01 showed little activity against these bacteria (Subhan et al., 2014).

Mawani et al. (2014) reported the first homogeneous complexes of curcumin with trivalent rare-earths (europium(III), gadolinium(III), and terbium(III)). The cytotoxicities of Eu ^**III** -01, Gd ^**III** -01, and Lu ^**III** -01 were investigated in MG-63 cells after 48 h of treatment using the MTT assay to evaluate their potential as antiosteoporotic agents. The three metal complexes exhibited cytotoxicity close to that of cisplatin (EC₅₀, Eu ^**III** -01, 61.6 µM; Gd ^**III** -01, 77.0 µM; Lu ^**III** -01, 99.3 µM; and cisplatin, 42.7 µM) but lower than that of curcumin (EC₅₀ = 22.1 µM). The results are not particularly promising, as a possible dissociation of curcumin from lanthanide(III) ions could limit the use of these compounds in the treatment of osteoporosis (Mawani and Orvig, 2014).

The mixed ligand-based complexes (Zr ^**IV** -02, Rh ^**III** -03, Th ^**IV** -01, UO _**2** ^**IV** -01, and UO _**2** ^**IV** -02), including curcumin and (E)-1-(3,4-dimethylphenyl)-2-(4-hydroxy-3-methoxycyclohex-3-en-1-ylidene)ethan-1-one exhibited different geometries and were biologically active. Antibacterial (Staphylococcus aureus, Bacillus subtilis, Pseudomonas aeruginosa, and E. coli) and antifungal (Aspergillus fumigatus, Penicillium italicum, Syncephalastrum racemosum, Alternaria alternata, and Candida albicans) assays were performed using agar diffusion with chloramphenicol (antibacterial) and terbinafine (antifungal) as positive controls. The free ligands were equally or more active than some of their metal complexes, except for UO _**2** ^**IV** -01 and UO _**2** ^**IV** -02, which were moderately antifungal, especially against A. fumigatus, P. italicum, and S. racemosum. The variation in efficacy was directly related to the microorganisms’ impermeability (Arun et al., 2008) (Figure 23).

The synthesis of curcumin complexes with rare-earth elements and other metal ions has considerably expanded the structural diversity of these systems. Nevertheless, many of the reported compounds display only moderate biological activity or even lower activity than the free ligand. In addition, much of the available evidence is based on preliminary microbiological or cytotoxic assays together with indirect techniques for studying DNA interactions. Potential ligand dissociation and the limited scope of biological evaluation make it difficult to draw definitive conclusions regarding their suitability as metallodrug candidates.

4 Concluding remarks

The development of curcumin-derived coordination complexes with second- and third-row transition metals has generated considerable interest in bioinorganic chemistry due to the potential to combine the ligand’s biological properties with the physicochemical characteristics of the metal centre. In general, metal coordination can enhance relevant curcumin properties, such as chemical stability, solubility, and bioavailability, and can also modulate its electronic, lipophilic, and photophysical properties. These characteristics have led to the exploration of applications in diverse biomedical areas, including anticancer, antimicrobial, and anti-inflammatory activity, molecular diagnostics, and photodynamic therapy. However, reported results show that metal complexation does not always lead to a significant improvement in biological activity. In numerous cases, the complexes exhibit activities comparable to or even lower than those of the free ligand, suggesting that metal coordination may alter key functional groups responsible for curcumin’s biological activity. Furthermore, the high stability of some complexes can limit the release of the active ligand or decrease its biological reactivity, thus reducing its therapeutic efficacy.

Another relevant aspect is that most available studies are based on in vitro assays, mainly using metabolic assays such as MTT, MTS, or AB, and rely on a limited number of cell lines or microorganisms. This makes it difficult to draw firm conclusions about the mechanisms of action and biological selectivity of these systems. In addition, some complexes have significant limitations, such as irradiation dependence in photoactive systems, potential proliferative effects at low concentrations, or toxicity problems associated with lipophilicity or the metal itself. Despite these limitations, several studies show promising results. Some ruthenium, palladium, platinum, or iridium complexes exhibit cytotoxicity in the micromolar range, relevant antimicrobial activity, or photoactive properties useful for photodynamic therapy. Furthermore, certain systems have demonstrated the ability to interact with key biomolecules, including DNA, serum proteins, and amyloid aggregates, opening up prospects for both therapeutic and diagnostic applications.

In this context, one of the main challenges is to establish clearer structure-activity relationships and to gain a deeper understanding of the molecular mechanisms responsible for biological activity. To achieve this, it will be necessary to complement chemical studies with more comprehensive biological evaluations, including more complex cell models, detailed mechanistic studies, pharmacokinetic analyses, and in vivo assessments. In conclusion, although curcumin metal complexes represent a promising platform for the design of new metallodrugs, their development still requires more rational structural optimisation and more rigorous biological validation. The integration of molecular design strategies, controlled-release systems, and advanced mechanistic studies could allow the full therapeutic potential of these systems to be realised in biomedical applications in the future.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Glossary

5′-GMP 5′-guanosine monophosphate A375 human melanoma cell line A549 adenocarcinomic human alveolar basal epithelial cell line A2780 human ovarian cancer cell line A2780cisR cisplatin-resistant human ovarian cancer cell line AB Alamar blue Aβ amyloid-β ABTS 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid AD Alzheimer’s disease ATP adenosine triphosphate Bax Bcl-2-associated X protein Bcl2 B-cell lymphoma 2 B-DNA form deoxyribonucleic acid BT-474 human breast cancer cell line ε molar absorption coefficient ESI electrospray ionization CD3CN deuterated acetonitrile COLO-205 human colorectal adenocarcinoma cell line COLO-320 human colorectal adenocarcinoma multidrug resistant expressing ABCB1 (MDR1)-LRP cell line CCK8 cell counting kit-8 CP8 cisplatin-resistant human ovarian cancer cell line CT-DNA calf thymus DNA D2O heavy water DCFDA 2′,7′-dichlorofluorescein diacetate DFT density functional theory DLD-1 human colorectal adenocarcinoma cell line DMEM Dulbecco’s modified eagle medium DMSO dimethyl sulfoxide DPPH 1,1-diphenyl-2-picrylhydrazyl DPPIV dipeptidyl peptidase IV DU-145 human prostate carcinoma cell line EC50 half maximal effective concentration ESI-MS electrospray ionisation mass spectrometry FAO Food and Agriculture Organisation FDA Food and Drug Administration FGF-2 soluble basic fibroblast growth factor FITC-PI fluorescein isothiocyanate-propidium iodide FT-ICR-MS Fourier transform ion cyclotron resonance mass spectrometry GLP-1 glucagon-like peptide 1 GSH glutathione GSTP1 glutathione S-transferase P1 HaCaT human keratinocyte cell line HCT116 human colorectal carcinoma cell line HeLa human cervical carcinoma cell line HEK-293 human embryonic kidney 293 cell line HEK-293T human embryonic kidney 293T cell line Hep3B human hepatocellular carcinoma cell line HepG2 human liver cancer cell line HL-60 human promyelocytic leukemia cell line HEWL hen egg white lysozyme HPL1D human peripheral lung epithelial cell line HPLC high-performance liquid chromatography HRMS high-resolution mass spectrometry HSA human serum albumin IC50 half-maximal inhibitory concentration ICP-MS inductively coupled plasma mass spectrometry IR infrared spectroscopy 3J and 4J coupling constant to three or four bonds JNK c-Jun N-terminal kinases JURKAT human T lymphocyte cell line Ki inhibitory constant λmax maximum wavelength log PO/W logarithm of the octanol-water partition coefficient LnCaP human prostate cancer cell line (lymph node carcinoma of the prostate) LnCaP-SF LNCaP steroid-free m/z mass-to-charge ratio M-1 cm-1 molar inverse centimetre inverse MB methylene blue MCF-7 breast cancer cell line (michigan cancer foundation-7) MCF-10A normal breast epithelial cell line MDA-MB-231 human breast cancer cell line (M. D. Anderson-metastatic breast-231) MG-63 human osteosarcoma cell line Mia PaCa-2 human pancreatic carcinoma cell line MIC minimum inhibitory concentration μM micromolar MMP-9 matrix metalloproteinase-9 MRSA methicillin-resistant Staphylococcus aureus MS mass spectrometry MTS 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NAC N-acetylcysteine NF-kB nuclear factor kappa-light-chain-enhancer of activated B cells nm nanometer NMR nuclear magnetic resonance NRF2 nuclear factor erythroid 2-related factor 2 QTOF quadrupole time-of-flight P27 cyclin-dependent kinase inhibitor 1B P53 tumor protein p53 gene PARP poli(ADP-ribosa) polimerasa PC-3 human prostate cancer cell line PCA principal component analysis PDT photodynamic therapy pH hydrogen potential PNT-2 human prostate cell line PTA 1,3,5-triaza-7-phosphaadamantane RAPTA ruthenium arene PTA RAPTA C [RuII(η6-p-cymene)Cl2(pta)] ROS reactive oxygen species RPMI 1640 Roswell Park Memorial Institute 1640 medium SCXRD single-crystal X-ray diffraction SI selectivity index SK-OV-3 human ovarian cancer cell line SLC5A6 solute carrier family 5 member 6 (gene) SRB sulforhodamine B SW-620 human colorectal adenocarcinoma cell line T-47D human breast cancer cell line T-89G human glioblastoma cell line UHPLC ultra-high-performance liquid chromatography U-87 Uppsala 87 malignant glioma UV–Vis ultraviolet-visible spectroscopy Vero African green monkey kidney cell line VRSA vancomycin-resistant Staphylococcus aureus WHO World Health Organisation.

Footnotes

Edited by: Craig C. McLauchlan, Illinois State University, United States

Reviewed by: Mahmood Ahmed, University of Education Lahore, Pakistan

Mahima Samanth, Vijaya Vittala Institute of Technology, India