Searching for new candidates for antimalarial therapy: physicochemical and in vitro evaluation of gold(I) and silver(I) N-heterocyclic carbene complexes Authors: William Castro, Hector R Rangel, Liseth Garibaldi, Wilma Rojas-Campos, Antonio José Hernández, Joe Berroteran, Oriel Sanchez-Velasco, María Cristina Goite, Miguel A Díaz, David Coll, Mariana Hidalgo Published: 2026 ABSTRACT AbstractBackgroundThe global fight against malaria is currently hindered by the rapid emergence and spread of multi-drug-resistant Plasmodium falciparum strains. This growing resistance has compromised the efficacy of frontline treatments and has led to the disease continuing to claim thousands of lives each year. Consequently, a new effective and non-toxic anti-malarial drug is urgently needed. In this context, metal–NHC (N-heterocyclic carbene) complexes have diverse physicochemical properties that could enhance biological activity and lead to the development of a new drug to combat malaria. On this regard, this study examined the biological activity of two metal–NHC complexes and their NHC precursor in two strains of the parasite, Plasmodium (3D7 and FCR-3), with specific targets.MethodsThe synthesis of all metal–NHC complexes was carried out under nitrogen atmosphere using Schlenk techniques. We studied the biological activity of two metal–NHC complexes and their NHC precursor in two Plasmodium strains (3D7 and FCR-3) and the selectivity index. Additionally, the possible interactions of these metal compounds with specific targets were evaluated. These targets included reactive oxygen species (ROS), albumin (BSA), Fe(III) PPIX, and β-hematin. The evaluation also included lipophilicity and ADMET properties. These properties were evaluated through diverse physical and spectroscopic methods.ResultsThe IC50 values of the NHC precursor (L) and the two metal–NHC complexes (1 and 2) were similar in P. falciparum chloroquine-sensitive and -resistant strains. They showed a higher selectivity index than chloroquine and do not cause hemolysis in red blood cells. Additionally, the compounds have a dose-dependent effect on the progression of the P. falciparum erythrocytic cycle, resulting in the accumulation of ring and trophozoite forms at high concentrations in the FCR-3 strain. These compounds may alter the pH of acidic organelles in the cytosol of the parasite, such as the digestive vacuole and the accessory vacuoles. Compounds L, 1 and 2 are capable to generate reactive oxygen species (ROS).ConclusionsPreliminary studies reported in this paper suggest that the two metal–NHC complexes 1 and 2 interact with targets such as the inhibition of β-hematin and the production of reactive oxygen species (ROS). They also have lipophilic and physicochemical profiles that correlate with their antiplasmodial efficacy. Both complexes exhibit moderate activity, indicating potential for further evaluation. A plausible explanation is that this behavior arises from the nature of both the ligand and the metal center. Accordingly, these initial findings provide a basis for the further exploration of new antimalarial compounds.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12936-026-05944-7.Keywords: N-heterocyclic carbine, Metal–NHC complexes, Antimalarial metallodrugs, Malaria targets Background The main malaria control strategies are vector control, the introduction of the RTS/S/AS01 and R21/Matrix-M malaria vaccines, and chemotherapy [1]. However, resistance has become a major problem, as evidenced of this by the increasing ineffectiveness of existing inexpensive drugs such as chloroquine [2, 3]. Although the aforementioned antimalarial therapies are generally effective, there is an increasing number of patients whose infecting parasite strains is resistant to these treatments. This is not only detrimental to the person’s health, but also allows them to remain as a reservoir, maintaining transmission of the resistant parasite strain. Since 2016, there has been increasing resistance to the first-line treatment for malaria, artemisinin-based combination therapies (ACTs), which has been confirmed in five countries in the Greater Mekong Subregion (Cambodia, Lao People’s Democratic Republic, Myanmar, Thailand and Vietnam) [4]. The spread of resistant strains to other regions could pose a major public health challenge and threaten important recent advances in malaria control [5]. These cases, which are resistant to several available combination therapies, are reservoirs that challenge parasite elimination and contribute to the persistence of residual malaria [6]. This has prompted researchers to explore and to develop new and more effective treatments in the control of the disease [7] and this has been driven by the understanding of the mechanism of action of current drugs, that depending on the compound could have different pathways, such as interfering with hemozoin formation in the vacuole (e.g., chloroquine, quinine, mefloquine, and other alkaloids) [8], or targeting with the dihydrofolate pathway (e.g., pyrimethamine, sulfadoxine, proguanil) [9], or interfering with the mitochondrial electron transport chain (e.g., atovaquone) [10] or triggering oxidative stress (e.g., artemisinin and derivatives, primaquine) [11]. Therefore, determining the localization of the compounds is crucial to understanding their potential mechanism of action. Metal–NHC (N-heterocyclic carbene) complexes are widely studied for their versatile structural properties and applications, based on great stability, ease of modulation of the electronic properties and excellent σ-donating capacity. In this sense, Ag(I)-NHC complexes have shown exceptional antimicrobial efficacy, including activity against a broad spectrum of both Gram-positive and Gram-negative bacteria, demonstrate in first place by Youngs and his co-workers [12]. These complexes have emerged as improved antibacterial agents due to their low toxicity in humans and slow release of silver leading to sustained biological activity. Another metal center of great interest is gold, which plays an important role towards the development for new M(I)–NHC drugs with different molecular targets in malaria [13]. Hemmert and co-workers described the first examples of Au(I)–NHC derivatives with antileishmanial properties, they synthesized the mononuclear cationic and neutral gold(I) complexes containing functionalized quinolones, the compounds presented high selectivity index for promastigote and additionally, antiproliferative activity against amastigote was observed with IC50 values below 1 mM. These results opened the doors for the investigation of this type of compounds in parasitic diseases. NHCs and their complexes—particularly with gold(I) and silver(I)—are emerging as promising, and highly potent agents against multi-resistant P. falciparum malaria parasites by modulating characteristics such as lipophilicity and electronic properties, excellent results can be found [14–16]. These compounds are often designed as hybrids, combining the NHC with established drugs like artemisinin to target mitochondrial thioredoxin reductase in the parasite. A recent example of this is provided by a series of hybrid molecules consisting of a dihydroartemisinin derivative and gold(I)-NHC complexes investigated by Hemmert and Benoit-Vical. All hybrid molecules tested so far have shown activity against P. falciparum [17]. On this regard, when designing this kind of hybrid systems or metal–NHC complexes, the lipophilic/hydrophilic balance, as well as the electronic properties and rigidity of the NHC ligand, are key factors to consider. Nowadays specific NHC frameworks have attracted considerable attention and have been more widely employed than other ligands, as exemplified by the pincer carbene 2,6-bis(3-methylimidazolin-2-yliden-1-yl)pyridine dibromide (Fig. 1, L), due to its amphiphilic nature, strong σ-donor character, and the structural stability and rigidity it imparts upon coordination. Therefore, further exploration of its complexes is warranted to evaluate their potential therapeutic applications. Structure of L (2,6-bis(3-methylimidazolin-2-yliden-1-yl)pyridine dibromide), complex 1 [Ag(C13H13N5)]2Br2, and complex 2 [Au2(C13H13N5)Cl2] Due to their low level of toxicity, good activity in the thioredoxin reductase system compared with auronofin, their distribution coefficient at pH 7 and P < 0, and growing scientific interest in this family of compounds as antimalarials, we decided to evaluate the biological properties of [Ag(L)]2Br2 (complex 1) and [Au2(L)Cl2] (complex 2)—both previously reported by our group [18]—against malaria parasites and their specific targets. We have studied the antiproliferative in vitro activity of these compounds (L, complex 1 and complex 2) against 3D7 and W2 strains (CQ-sensitive and CQ-resistant strains, respectively) and the possible interactions of these metal compounds with specific targets such as reactive oxygen species (ROS), albumin (BSA), Fe(III) PPIX and β-hematin, lipophilicity and ADMET properties through diverse physical and spectroscopy. Methods General procedures All solvents were purified by standard procedures immediately prior use, chloride hemin, buffers and solvents were purchased from Sigma-Aldrich. Ligand (L) and silver(I)- and gold(I)-N-heterocyclic-carbene complexes (complex 1 and complex 2) were synthesized according to literature [18]. Fluorescence measurements were carried out using a Perkin Elmer LS45 spectrophotometer with a pulsed xenon lamp. UV–vis spectra were recorded in a Thermo Scientific model Evolution 260 Bio spectrophotometer using 1 cm path length cells at 298 K, using the corresponding solvent, and Zeiss Axioscop microscope equipped with a digital camera and microplate reader Spectra Max to quantify. Biological evaluation Parasite culture Parasitized, Percoll gradient-concentrated cells, supplemented RPMI 1640 medium (25 mM HEPES, 2 mM l-glutamine, 25 mM sodium bicarbonate, 3.25 μM reduced glutathione, 50 μg/mL gentamicin, 36 μM hypoxanthine) and adjusted to 5% hematocrit and 5% parasitemia. Subsequently, positive controls and parasite cultures were exposed to the compounds at 1, 3, and 5 µM. The cultures were monitored. The cells were kept at 37 °C in a humidified CO2 incubator. Cytotoxicity assays and selectivity index Cytotoxicity was evaluated in Vero cells using the same dilution protocol as for the antiplasmodial assays. The plating of cells was done in 100 μL on 96-well plates, with seeding at 104 cells per well. After a 24-h incubation at 37 °C and 5% CO2, the cells were exposed to different concentrations of the compounds (100 μL per well) for an additional 48 h. Cell viability was measured using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Sigma-Aldrich). Briefly, 100 μL of MTT solution (0.5 mg/mL in MEM) was added to each well. Thereafter, the plate was incubated for 1 h at 37 °C under 5% CO2 atmosphere. Next, the cells and MTT crystals were dissolved in 100 μL of DMSO using a plate shaker. Absorbance was measured at 540 nm using the μQuant from BioTek. Cell proliferation was calculated using GraphPad Prism software (version 7; GraphPad Software, San Diego, CA, USA) from a minimum of three independent experiments. Culture synchronization The different stages of the parasite were separated and concentrated using a discontinuous Percoll gradient [19]. The procedure was performed by adding one volume of a washed blood culture suspension in PBS pH 7.3 (10.40 mM disodium phosphate dihydrate, 3.15 mM monopotassium phosphate, 116.32 mM sodium chloride) and four volumes of isotonic Percoll. The suspension was then subjected to centrifugation at 1300×g for 15 min at 22 °C. This Percoll gradient yielded two bands, the denser one concentrating the ring-stage parasitized erythrocytes was collected. The parasitic stage was confirmed by microscopic examination using Giemsa-staining smears. Hoechst and acridine orange staining RBC Acridine orange (2.6 µM, in PBS) was used to stain viable Plasmodium sp., as this fluorophore accumulates in acidic compartments is sensitive to pH gradient collapse [20]. For this purpose, a fraction of the culture was subjected to simultaneous incubation with nuclear counterstaining using a solution of Hoechst 33,258 dye in 2 µM PBS in a darkened environment at room temperature for 15 min. The results were measured by epifluorescence microscopy. Staining of 1 mL of asynchronous P. falciparum 3D7 strain culture (2% hematocrit, ≅3% parasitaemia) with 0.5 μM L, 1, and 2 in PBS was performed. After 5 min of incubation (at 37 °C, humidified atmosphere with 5% CO2), WGA-Alexa 633 at ≤1 μg/mL (1:2 dilution from the original stock) and 5 μg/mL Hoechst 33,342 (Invitrogen, Thermo Fisher Scientific) were added followed by incubation for 10 min. Wheat germ agglutinin-Alexa 633 was purchased from Thermo Fisher Scientific. Hoechst 33,342 (Invitrogen, Thermo Fisher Scientific) is a lipophilic DNA-binding fluorescent dye. Following the WGA-Alexa 633 and Hoechst 33,342 protocols, samples were washed to remove unbound material. Hemolysis assay The hemolytic activity of the compounds was evaluated using human red blood cells. EDTA-treated tubes were used to collect blood. The tubes were then spun at 1500×g for 10 min to separate the erythrocytes. The pellets were then washed three times with PBS (pH 7.4). Then, the erythrocytes were resuspended to a final hematocrit of 2%. Aliquots of the RBC suspension were incubated with 1, 3 and 5 μM of the compounds at 37 °C for 1 h. RBCs were used as negative controls, and were incubated with PBS alone. Positive controls were treated with 0.1% Triton X-100 to induce complete hemolysis. Following incubation, samples were centrifuged at 1000×g for 5 min, and the absorbance of the supernatant was measured at 540 nm using a microplate reader to quantify the hemoglobin released. Hemolysis (%) was calculated using the following formula: This assay allowed assessment of the compound’s cytotoxicity toward erythrocytes and the determination of safe concentration ranges for further studies [21]. Assessment of the mechanisms of action of antimalarial drugs Evaluation of ROS using DAB Aliquot from the different treatments and control samples from the synchronized P. falciparum cultures was collected at 10 and 22 h of incubation. The samples were smeared onto slides and fixed with methanol. A 2.8 mM diaminobenzidine (DAB) solution (Sigma D5905) with 3% hydrogen peroxide in 0.05 M Tris buffer (pH 7.6) was used for the staining. The solution was incubated in the dark for 6 min. The detection of reactive oxygen species in the form of DAB-oxidation precipitates was carried out using bright-field microscopy, and images were captured with a Zeiss Axioscop microscope equipped with a digital camera [22]. Protein binding studies Fluorescence measurements were performed on a Perkin Elmer LS45 with a pulsed xenon lamp, using a quartz cell with a path length of 1 cm (3 mL volume). The excitation wavelength was 278 nm with 10 nm slits; emission spectra were recorded in the 300–800 nm wavelength range with 5 nm slits and 0.1 s integration time. The maximum fluorescence emission of BSA at λex = 280 nm was at 343 nm. The fluorescence intensity of BSA in the absence and presence of compound was measured by keeping the BSA concentration constant (1 × 10−5 M) while varying the compound concentration from 0 to 1 × 10−4 M (ri = [complex]/[BSA]) = 0–5. The absorption titration experiment was performed by keeping the BSA concentration constant (14 µM) and varying the concentrations of the compounds (0–6 µM). This was done by adding different concentrations of the compound to fixed volume of BSA solution. Samples were equilibrated for 8 min before each measurement. Titrations were performed manually using a micropipette. The change in absorbance of the characteristic band of BSA at λmax = 280 nm was recorded after each titrant addition. Interaction with ferriprotoporphyrin and β-hematin formation As previously described, the association constant of L and metal–NHC complexes 1 and 2 with ferriprotoporphyrin IX (Fe(III)PPIX) was determined [23]. A 4 μM solution of hemin was prepared, and the absorbance was recorded at 402 nm. Each compound was titrated a reference cell containing 40% (v/v) DMSO, 0.020 M Tris, (pH 7.5) to correct for its absorbance. The data were fitted to a 1:1 complexation model using nonlinear least squares fitting in strict accordance with the procedure outlined by Egan et al. [24]. This process yielded the conditional association constant. The conversion of hemin to β-hematin in acidic acetate solutions was investigated using infrared spectroscopy, employing the technique devised by Egan et al. [24]. Domínguez et al.’s protocol was followed when performing the IC50 of β-hematin formation in a buffer assay [25]. In brief, a solution of hemin (50 μL, 4 mM) dissolved in DMSO was distributed in 96-well microplates. The hemin complex was dissolved in DMSO and added in triplicate to the test wells (50 μL each) for a final concentration of 0–20 mM per well. Controls contained water and DMSO. The addition of acetate buffer (100 μL, 0.2 M, pH 4.4) initiated β-hematin formation. The plates were then subjected to an incubation process at 37 °C for 48 h, thereby enabling the culmination of the designated reaction. Thereafter, the plates were subjected to a centrifugation process at 4000 RPM for 15 min. After discarding the supernatant, the pellet was washed twice with DMSO (200 μL), then dissolved in NaOH (200 μL, 0.2 N). The solutions were then diluted 1:2 with NaOH (0.1 N), and the absorbance was measured at 405 nm. The results were expressed as a percentage of β-hematin formation inhibition. A method proposed by Egan et al. [26] and modified by Martínez et al. [27] was used to study the formation of β-hematin at water/n-octanol interface. Hemin was dissolved in a 0.1 M NaOH solution to produce hematin, followed by the addition of acetone to achieve an acetone/water ratio of 4:6 (v/v). The final solution contained 15 mg hematin/mL. A sample of hematin in acetone was introduced at the interface between n-octanol (2 mL) and aqueous acetate buffer (5 mL, 8 M; pH 4.9) at 37 °C for 2 h. The pellet was collected and washed with DMSO and ethanol. Finally, the pellet was dissolved in 0.1 M NaOH for spectrophotometric quantification. To assess the capacity of the compounds to impede β-hematin formation, the requisite amount of the compound was dissolved prior to the addition of the hematin solution at the interface. All experiments were performed in triplicate. In silico (DFT calculations) The starting structures of the compounds were sketched and optimized by molecular mechanics with the Discovery Studio Visualizer package [28]. For chloroquine, its neutral form was selected for calculation because charged species are not permeable through the cell membrane [29]. Subsequent quantum mechanical calculations were performed with the Gaussian09 package, in the context of the density functional theory, using the hybrid functional B3LYP and the 6-31 + G* basis set for non-metallic elements, and the LANL2DZ pseudopotential for the metals, and empirical dispersion corrections were added using Grimme’s D3 method [30–34]. Stability tests of the wave function for the singlet and triplet states of the molecules in vacuum were performed, and in all cases the singlet state was found to be the most stable. The generated wave functions were stored into checkpoint files and then read as an initial guess for subsequent geometry optimizations in vacuum, water, and n-octanol dielectric environments via the solvation model based on density (SMD) [35]. In all cases, the existence of minimum energy structures was verified through vibrational frequency calculations. The calculation of thermodynamic parameters was carried out at a temperature of 310.15 K to approximate biological conditions. The n-octanol/water partition coefficient (P) is used as an arbitrary and conventional measure of lipophilicity for different substrates [36]. To estimate the affinity of these compounds with the lipophilic environment of the cell membrane, these coefficients are calculated using the solvation-free energies of the compounds at biological temperature: where [S]k represents the molar concentration of the substrate S in the solvent k in equilibrium condition, ΔGsolv(S → k, T) is the free energy of solvation in k of the species at temperature T, R is the constant of the ideal gas and GS → k(T) is the molar free energy of S in solvent k at temperature T and 1 atm of pressure, which is calculated from the electronic energy, the repulsion energy between the nuclei, the energy vibrational correction of the 0 point and the thermodynamic corrections of translation, rotation and vibration programmed in the Gaussian09 package. If the concentration of S in n-octanol exceeds the corresponding one in water, P > 1, while in the contrary case, 0 < P ≤ 1. Another intuitive measure of lipophilicity is the base 10 logarithm of the partition coefficient: If log(P) > 0, the equilibrium concentration of the substrate will be higher in n-octanol, and if log(P) < 0 the concentration will be higher in water. According to Bader [37], the boundary of a molecule can be defined by an electron density isosurface. This methodology has been used previously for various purposes [38, 39]. Wavefunction files (*.wfn) were generated from DFT calculations to generate electron density grids using the AIM-UC software [40]. This program builds atomic basins from the grid and enables their integration, providing both the electron population and the volume associated with each atom in atomic units. Partition coefficient (log P) The water/n-octanol partition coefficients were determined by the stir-flask method [41, 42] and carried out in our previous research [18]; a UV–Vis calibration curve for each complex in n-octanol-saturated water in the 2–50 μM range was prepared. A mixture of 5 mL n-octanol and 5 mL water (each saturated in the other) was stirred for 30 min at 27 °C, after addition of the sample. The pH was maintained at pH 7.2; 4.9 and 2.9 (0.1 M Triz-HCl; 0.1 M Citrate and 0.1 M Citrate respectively). After reaching equilibrium, the phases were separated by centrifugation. Finally, the measurement of the drug concentration in each phase was determined spectrophotometrically to determine P values (P = [drug in n-octanol]/[drug in water]). These experiments were performed in triplicate [43]. Druglikeness of compounds: an in silico ADMET study Based on Lipinski’s Rule of five, the drug-likeness properties of the compounds L, 1 and 2 were determined through the SwissADME web tool, a free online program available at http://www.swissadme.ch/ (accessed on 15 February 2025) [44]. The physicochemical and pharmacokinetic parameters of the compounds that were analyzed include their molecular weights, partition coefficient (W log P) method, number of hydrogen bond acceptors (HBA) and donors (HBD), topological polar surface areas (TPSA), water solubility and gastrointestinal absorption [45–47]. Results and discussion Biological evaluation We determined dose–response profiles of L, complex 1, and complex 2 against the blood stage of CQ-sensitive 3D7 strain and CQ-resistant FCR-3 strains. The antiplasmodial activity (IC50) and cytotoxicity (CC50) of the compounds in Vero cell line and their selectivity indices were determined (Table 1). The reference antiplasmodial compounds chloroquine (CQ) and artemisinin (ART) were included as controls. The results showed that the IC50 values against P. falciparum ranged from 1.4 to 6.0 µM for the ligand L and complexes 1 and 2 (Table 1), indicating that L is a bioactive substance with an IC50 value of less than 5 µM in both strains. The incorporation of the gold(I) center in 2 slightly improved the antiplasmodial efficacy compared to L and to silver complex 1. Table 1In vitro anti-malarial activity of the compounds on P. falciparum chloroquineS and P. falciparum chloroquineR strains, Vero cell cytotoxicity, resistance index (RI), selectivity index (SI) and hemolytic activityCompoundsAntiplasmodial activityCytotoxicity activityHemolytic activity (g/dL)P. falciparum chloroquineS strain (3D7)IC50 (μM)P. falciparum chloroquineR strain (FCR-3)IC50 (μM)Resistance indexVero cell lineCC50 (μM)Selective index (SI)P. falciparum chloroquineS (3D7)Selective index (SI)P. falciparum chloroquineR (FCR-3)L4.93.80.820 × 103≫1000≫10001.61Complex 16.05.80.921 × 103≫1000≫10001.65Complex 221.40.721 × 103≫1000≫10001.63CQ0.01490.6241.6134≫10002161.60ART0.0170.00170.1130≫1000≫10001.58Resistance index (RI) = IC50 (FCR-3)/IC50 (3D7)Selectivity index (SI) = CC50 (Vero cell)/IC50 (P. falciparum strain)S sensitive, R resistant In vitro anti-malarial activity of the compounds on P. falciparum chloroquineS and P. falciparum chloroquineR strains, Vero cell cytotoxicity, resistance index (RI), selectivity index (SI) and hemolytic activity Resistance index (RI) = IC50 (FCR-3)/IC50 (3D7) Selectivity index (SI) = CC50 (Vero cell)/IC50 (P. falciparum strain) S sensitive, R resistant It is noteworthy that while CQ and ART displayed higher antiplasmodial activity, all tested compounds exhibited a resistance index (RI) below 1, suggesting that they can effectively overcome the CQ resistance shown by the strain (FCR-3). Furthermore, the selectivity index (SI) for both sensitive and resistant strains were remarkably high, comparable to the reference compound CQ, and with values even higher than metal complexes reported in the literature [14, 16]. Based on these findings from Table 1, further studies will focus on studying the effect of complexes and identifying possible targets on the stages of Plasmodium. Effect of compounds in the viability of Plasmodium intraerythrocytic cycle The evaluation of the parasitostatic effects of drugs with potential antimalarial activity is reported to be linked to the improvement of the host’s immune response, as they delay the progression of infection and lead to the expression of plasmodial antigens in the early stages of infection, which enhances the acquired immune response. CQ and ART have been reported to drastically reduce parasitaemia during initial infection, which would also reduce the immune system’s ability to interact with parasitized cells. The compounds under investigation (L, complex 1 and complex 2) appear to have a dose-dependent effect on the progression of the P. falciparum erythrocytic cycle, resulting in the accumulation of rings and trophozoite forms at high doses in FCR-3 strain. In particular, complex 2 reduces the percentage of schizonts at a concentration of 1 µM, and as the concentration increases, the proportion of rings and trophozoites increases. This behavior is similar to what occurs with CQ and ART, but only at a high concentration (effect of complex 2 in Fig. 2 and effect of L and 1 in Figs. S1 and S2). Effect of complex 2 in the accumulation of P. falciparum erythrocytic stages (FCR-3 strains). Line shows the percentage of different stages, rings (blue), trophozoites (red), schizonts (yellow) forms observed after 20 h incubation at different concentrations (chloroquine 9 nM and artemisinin 9 nM). Data collected by microscopic inspection of Wright’s-stained blood smears and compared with controls untreated. The percent of each group accounts for the fraction of cells showing every parasite stage out of a total of 1000 erythrocytes. Results are representative of two experiments All compounds decreased parasitaemia up to 20 h and then showed a progressive decrease in parasitized cells. This dual parasitostatic and parasiticidal profile suggests that L, complexes 1 and 2 may act through different molecular targets compared to those of antimalarial drugs [8–11]. Consequently, determination of the subcellular location of the compounds is essential to further provide an indication of the possible mechanism of action. Cellular localization studies Hoechst and Acridine Orange (AO) staining of parasitized erythrocytes were used to assess local accumulation of compounds in pRBC (parasitized red blood cells) during infection. Hoechst staining is specific for chromatin state and DNA conformation. An increase in fluorescence intensity indicates chromatin condensation and DNA fragmentation within the parasite. Additionally, the AO dye was used to determine the effect of the compounds on the integrity of parasite’s acidic organelles, such as lysosomes and cytosolic invaginations of the parasitophorous vacuole. It also targets the main digestive vacuole, the site of hemoglobin degradation; this process occurs primarily during the trophozoite stage, where the most intense proteolysis takes place [48]. In order to observe the morphological changes in the erythrocytes due to the influence of the tested compounds, a light microscope (DIC microscopy) was used. After 20 h of incubation, P. falciparum reached the schizont stage. The nuclei of P. falciparum schizonts in pRBC are observed as bright blue dots, while the acidic compartments are assessed by the red emission band (650 nm) localized in a defined area. This is consistent with the trapping of AO in the digestive vacuole upon protonation. Concerning to evaluated compounds (L, complexes 1 and 2), significant changes were observed not only in the parasite morphology but also in the fluorescence intensity of bright blue dots (Fig. 3). This increase indicates chromatin condensation and DNA deformation in the parasite. Complex 2 induced the most significant changes, suggesting that this gold complex triggers stronger chromatin condensation than either L and complex 1. These modifications are seen to be greater than in the respective reference compounds (CQ and ART). However, it should be noted that DNA is not the main target of these conventional antimalarials. Nuclear morphological observation of Pf3D7 using Hoechst and Acridine Orange stain after treated with increasing concentration of compounds evaluated for 72 h at 100× (oil immersion) magnification and differential interference contrast (DIC). A Non-parasitized RBC (NpRBC, untreated), B parasitized RBC (pRBC, untreated), C parasitized RBC with L [3.5 µM], D parasitized RBC with complex 1 [5 µM], E parasitized RBC with complex 2 [2.5 µM], F parasitized RBC with artemisinin [0.01 µM] and G parasitized RBC with chloroquine [0.01 µM]. Hoechst (blue) was excited at 364 nm, acridine orange (red) at 650 nm. Red scale bar = 5 μm Additionally, changes in red fluorescence can be observed (Fig. 3C–E), indicating pH fluctuations, although the increase in intensity is greater for CQ, which is known for its potent pH-altering effect. These results suggest that this compound may alter the pH of acidic organelles in the cytosol of the parasite, such as the digestive vacuole and the accessory vacuoles. However, as with CQ, there are other mechanisms of parasite developmental inhibition independent of deprotonation of acidic organelles, and it cannot be state unambiguously that pH changes is the primary mechanism of action of compound [49]. In addition, all evaluated compounds and reference drugs, distorted forms of pRBC. Notably, complex 2 caused rupture of schizonts, demonstrating a more potent interaction with Plasmodium. Consequently, due to these results it was essential to evaluated the hemolytic activity of the complexes in unparasitized red blood cells to determine their toxicity toward healthy host cells. Hemolytic activity on unparasitized erythrocytes The percentage of lysis of uninfected erythrocytes (NpRBCs) by the complexes was performed with a dual purpose: to assess whether the compounds cause lysis, leading to serious side effects, and to determine the cytotoxic potency against NpRBCs, which is critical to the determination of a preliminary drug’s safety profile. The assay showed no differences in extracellular hemoglobin content between compound-treated and vehicle control (0.5% v/v DMSO)-treated erythrocytes. No hemolytic effect of any of the compounds at the concentrations tested (Table 1). This suggests that L, complex 1 and complex 2 exhibit favorable hemocompatibility, as they remain non-toxic to healthy red blood cells at concentrations up to 5 μM. Assessment of the mechanisms of action of antimalarial drugs Assessing ROS generation with DAB in pRBC A key mechanism of action for several antimalarial drugs is the induction of ROS and consequent oxidative stress to kill the malaria parasite. To investigate this pathway, an intracellular 3,3′-Diaminobenzidine (DAB) was used. DAB assay is a colorimetric indicator developed for in situ ROS imaging. It reacts with H2O2 under peroxidase catalysis to form an insoluble brown precipitate that can be imaged microscopically at the tissue or cellular level. Healthy erythrocytes (NpRBC) exhibited a uniform brown color throughout the cytoplasm, that corresponds to an intermediate DAB oxidation product which has high affinity for heme groups. In untreated parasitized erythrocytes (pRBC, Fig. 4A), the cytoplasm was uniformly stained, similar to NpRBC, whereas the parasite body present in the cytoplasm remained unstained, except for discrete black-to-brown aggregates of heterogeneous size, corresponding to ROS production. In the control, after 22 h of culture, it was possible to observe some precipitate inside the parasite, corresponding to the basal level of reactive oxygen species that react with DAB and H2O2 (Fig. 4). Evaluation of reactive oxygen species with DAB in pRBC. Oxidized DAB in bright field after irradiation at 455–495 nm, showing the dark brown (oxidized) DAB product to 10 and 22 h. The arrows at the periphery of the parasitophorous vacuole indicate the oxidized product (granules). A Parasitized RBC (pRBC, untreated), B L [5 µM], C complex 1 [5 µM], D complex 2 [5 µM], E artemisinin [0.01 µM], F chloroquine [0.01 µM]. Red scale bar = 5 μm Maintaining redox metabolism in the parasite involves several key components, one of which is the ability to defend against oxidative attacks arising from haemoglobin metabolism. These oxidative attacks result in the formation of ROS and the release of toxic ferriprotoporphyrin IX [50]. The decrease in ROS observed in the positive control with CQ is consistent with that reported by Loria et al. [51], establishing that this drug accumulated into the parasite food vacuole, where it inhibits the peroxidative breakdown of heme. Thus, one of its mechanisms of antimalarial action consists in the inhibition of the degradation and detoxification pathway of the heme group mediated by hydrogen peroxide. In contrast, compounds L, 1 and 2 produced precipitates at a concentration of 5 µM as early as 10 h of culture. First, microscopically, it was observed that the precipitates increased in density and diameter over control (arrowheads). Furthermore, a rearrangement of these precipitates (Fig. 4, open arrows in the different compounds) can also be observed in the membrane of the parasitophorous vacuole of the parasite. A pattern that coincides with those observed for the ART control. It is important to acknowledge that, while the mitochondrion in Plasmodium sp. is the classical organelle associated with maintaining redox homeostasis of the parasite, the apicoplast also plays a crucial role. This organelle maintains high metabolic activity in the synthesis of fatty acids and parasitic components, which requires a high demand for antioxidant protection. In addition some key enzymes have been found in this process such as glutathione reductase and thioredoxin reductase [52], suggesting that the oxidative stress induced by our complexes could be targeting these vital metabolic hubs. Interaction protein binding studies (BSA) Serum albumin serves as a delivery system for lipophilic compounds in the biomedical field. The basis of its use in biomedical applications is the similarity to human serum albumin (HSA), a fully biocompatible protein. Bovine serum albumin (BSA) is generally employed as its human homologue and can be studied via fluorescence spectroscopy, in particular, tryptophan (Trp) and tyrosine (Tyr) are responsible for its intrinsic fluorescence. A qualitative analysis of the binding between the compounds and BSA can be performed by monitoring the quenching of the Trp-213 residue [53]. The tryptophan residue is the only one excited when the excitation wavelength is 280 nm. However, if the excitation wavelength is shifted to 275 nm, both the tryptophan and the tyrosine residues will be excited. For this reason, 280 nm was employed for the excitation of BSA in the experiment, so as to rule out the contribution of tyrosine residues [54]. Fluorescence spectra were recorded from 300 to 500 nm with the excitation wavelength at 280 nm (Fig. 5). The fluorescence intensity of BSA was measured in the absence and presence of L, complexes 1, 2 and CQ as reference control, and all experiments were performed at 298 K. The Stern–Volmer equation describes the quenching property (Eq. 1): where KSV Stern–Volmer is the quenching constant. Effect of complex 1 on fluorescence spectra of BSA (T = 298 K, pH: 7). Curve (1): 0.4 μM BSA; curves (2–14): 0.4 μM BSA in the presence various concentrations of complex 1 (from 0.0 to 10 μM) at increments of 10 μL; blue line shows the emission spectrum of complex 1 only The modified Stern–Volmer equation was used (Eq. 2) [55]: Kb represents the binding (association) constant, and n is the number of binding sites. As shown in Table 2, it was found that the binding ratio of the BSA to L and complex 1 was approximately 1:1. In contrast; no significant changes in fluorescence intensity were observed upon the addition of complex 2 indicating that this complex does not interact with BSA under the experimental conditions tested. Table 2Interaction of compounds with BSA and hemin respectively, at 298 KCompoundInteraction with BSAInteraction with heminKsv (104 L mol−1)Number of biding sites (n)Kb (103 L mol−1)Log KCI50HA (µM) in bufferCI50HA (mM) in interfaceL2.93 ± 0. 210.98 ± 0.0121.5 ± 0.63.44 ± 0.07˃10009.5 ± 0.6Complex 14.32 ± 0.990.98 ± 0.0232.3 ± 1.33.67 ± 0.14˃25013.3 ± 0.6Complex 2–––3.99 ± 0.09˃2501.44 ± 0.2CQ9.21 ± 0.240.94 ± 0.0352.1 ± 0.25.31 ± 0.09242.15 ± 1.252.21 ± 0.7ARTNDNDNDND˃100090.29 ± 0.9Ksv Stern–Volmer quenching constant, n number of binding sites for binding of compound to BSA, Kb association constant, CI50HA Concentration of drug required to inhibit 50% of heme aggregation, ND not determinedData are expressed as mean ± SD of three measurements Interaction of compounds with BSA and hemin respectively, at 298 K Ksv Stern–Volmer quenching constant, n number of binding sites for binding of compound to BSA, Kb association constant, CI50HA Concentration of drug required to inhibit 50% of heme aggregation, ND not determined Data are expressed as mean ± SD of three measurements For L and complex 1, the BSA fluorescence intensity decreases significantly; a slight shift of Trp-213 emission maximum (~340 nm) towards lower wavelengths (blue shift) was observed. This shift in the emission maximum wavelength indicates the formation of an adduct between BSA and the compounds, which promotes a more hydrophobic microenvironment around Trp-213 [56]. The Stern–Volmer plots remain linear across the entire range of BSA/compound molar ratios studied (Fig. 5). Ksv values are lower for L and 1 than those obtained for CQ, indicating weaker interactions with BSA. However, they can be compared with the chemotherapeutic drugs and potential antibacterial agents reported in the literature [57–59]. These results suggests that L and complex 1 are not transported by BSA as effectively as CQ to reach the site of biological action in the human body [60]. Interaction with ferriprotoporphyrin and inhibition of β-hematin During the pathogenic blood stage, the malaria parasite catabolizes host hemoglobin as its primary source of amino acids, releasing free heme groups, highly toxic to its organism. To overcome this, the parasite has a detoxification pathway transforming the free heme group into an insoluble, inert compound known as hemozoin (malaria pigment), which is one of the main targets for antimalarial drugs such as quinolines [61]. The inhibition of free heme aggregation by quinoline-type antimalarial agents has been the subject of several hypotheses. It has been suggested that the primary mode of action involves a direct interaction between the drug and ferriprotoporphyrin IX (FP) [62]. In this study, the interaction of heme with compounds L, 1 and 2 was evaluated spectrophotometrically by following the Soret band [63]. This band is characterized by an absorption maximum at 402 nm, which is indicative of the presence of ferriprotoporphyrin IX (Fe(III)PPIX) and its characteristic π-π* transitions within the porphyrin macrocycle [23]. A strictly analogous fit was obtained in the case of log K for the tested compounds. These values range between 3.44 and 3.99 (Table 2) which is similar to that of mefloquine (log K = 3.90). This indicates that the compounds interact moderately with hematin under the experimental conditions employed [23]. These results suggest that the compounds may have different targets, one of which is the inhibition of the formation of hemozoin through an interaction with ferriprotoporphyrin. Hemozoin formation is an essential process for Plasmodium parasite survival, making it an attractive target for drug development-often referred to as “malaria’s Achilles heel” from several authors. β-hematin, the synthetic analogue, is a heme dimer produced by reciprocal coordinates bonds between the carboxylic acid groups of the protoporphyrin-IX ring and the iron center of an adjacent heme molecule. Heme aggregation inhibition activity (HAIA) has been extensively reported in the literature, in the present study, it was evaluated using two distinct procedures: the aqueous buffer and the lipid-water interphase methods [25, 27]. The aqueous buffer method showed that these compounds are unable to inhibit the formation of hemozoin under these conditions (acetate buffer, pH 5.0), while CQ showed a value of 242.15 µM, which is consistent to other authors and our own previous findings [64, 65]. In the lipid-water interphase method adapted from the technique reported in 2009, significant differences were observed in the IC50HA values obtained. This method represents a more realistic biomimetic condition where β-hematin assembles rapidly and spontaneously, and the overall aggregation process is much faster (60 min). Hematin is carefully introduced near the interface after the drug has equilibrated between the two phases [26]. The activity trend underwent a radical transformation compared to the outcomes of aqueous buffer assay. In the three tested compounds, inhibition of heme aggregation near the interface was observed. While the values of L and 1 were found to be comparable, the IC50HA for complex 2 was tenfold lower, showing a potency comparable to that of CQ. The discrepancy found between both methods may be attributed to the fact that the interface environment is better suited to metal complexes, as previously reported [65]. These findings provide a plausible explanation for the antiplasmodial activity of these metal complexes, suggesting that they can inhibit aggregation, although they likely target additional pathways beyond inhibiting β-hematin. Lipophilicity The study of lipophilicity is fundamental to drug design and development [66]. Lipid affinity facilitates absorption across cell membranes, which are mainly composed of phospholipid bilayers. This property is related to the ability of a drug to efficiently distribute into tissues that are lipid-rich environments, such as cell membranes or the parasite itself [67]. Therefore, lipophilicity must be carefully considered and optimized during the drug development process to ensure that therapeutic agents are effective and safe. Lipophilic properties are related to molecular size or volume and the partition coefficient (log P) [68]. In this study both computational tools and experimental models were used to estimate these parameters. Atom-in-molecule (AIM) theory posits that molecular systems consist of regions in space where a set of atomic nuclei exchange charge and momentum, resulting in an electron density distribution, ρ, with well-defined topological features that are transferable between chemical environments. In this approach, space is partitioned into regions separated by zero-flux surfaces (ZFS) defined by any point in space tangential to the electron density gradient ∇ρ. The existence of a ZFS between any pair of bound atoms A and B implies that the set of all surfaces related to atom A delineates its volume, known as the atomic basin ΩA. Due to the variable and complicated shapes of atomic basins, integration through atomic basins is performed by numerical approaches implemented in AIM analysis packages. In principle, the interatomic surfaces extend indefinitely in space. Therefore, the molecular space must be delimited by an envelope with the practical purpose of solving volume integrals over atomic basins. In the case of the partition coefficient obtained from ab initio calculations, the concentration of the substance in both phases remains constant with time under equilibrium conditions. Fundamentally, the partition coefficient (Pow), is the equilibrium constant for the water-n-octanol transition. In this equation, [A]x denotes the molar concentration of A in the solvent x, ΔTGΘ represents the standard free energy for the water ⟶ n-octanol transition, the gas constant is denoted by R and the absolute temperature (310.15 K for physiological conditions) by T The logarithm of Pow is an intuitive measurement for the preferential phase, i.e., the phase which the most molecules of A are found: Both parameters-molecular volume and log Pow, were calculated for L, and complex 1 and 2. CQ and ART were included as reference compounds to validate the method, and the obtained results are comparable to those reported in the literature. Molecular volume varies according to electron density. Table 3 shows the volumes both in a vacuum and in the presence of a solvent. Considering volume as an intrinsic molecular property, it is crucial to observe how the volume is affected by the surrounding medium, such as a lipid membrane. Our results indicate that this property is slightly influenced by the electrostatic effects of the solvent and the values are in accordance with their molecular weight. Regarding the partition coefficient, the values obtained for the reference compounds align to those reported in the literature, notably CQ exhibited the highest lipophilicity, validating our computational approach. Table 3DFT and Stir-flask method analysis of molecular volumes and partition coefficients of the compounds studiedCompoundDensity functional theory calculationsStir-flask partition coefficient method (log P)Molecular volume (Å3)Log(Pow)pH 3pH 5pH 7VacuumWaterOctanolL288.6290.5290.1−2.1861−2.09 ± 0.29−1.07 ± 0.04−0.67 ± 0.02Complex 1578.6581.0580.13.6112−0.60 ± 0.03−0.59 ± 0.04−0.54 ± 0.07Complex 2367.2372.3370.74.7781−0.97 ± 0.020.54 ± 0.030.52 ± 0.03CQ436.5436.4436.83.7169−1.72 ± 0.41−0.73 ± 0.110.57 ± 0.07ART342.7342.7342.91.5856−1.65 ± 0.36−0.37 ± 0.011.27 ± 0.17 DFT and Stir-flask method analysis of molecular volumes and partition coefficients of the compounds studied Among the tested compounds, complex 2—the one with the largest volume—exhibits the most lipophilic behavior, while the one with the smallest volume, ligand L, demonstrates a preference for a hydrophilic environment. Consequently, similar to the reference compounds, the metal complexes tend to prefer the n-octanol phase. This evidence indicates that both substances are capable of crossing biological membranes and spreading effectively throughout the lipid tissues of the body. The stir-flask method was used to determine n-octanol/water partition coefficient. This method is applicable to solutions under constant temperature, pressure and pH. It is particularly suitable to pure material dispersed between two pure immiscible solvents when the solute concentration in both phases exceeds 0.01 mol/L. More specifically, the method involves bringing a test substance into equilibrium with n-octanol and water in a stirred reactor followed by the measurement of the substance’s concentrations in each phase. The partition coefficient (Pow) is then calculated as the ratio of the equilibrium concentration in the organic phase to that in the aqueous phase [41]. The log P values obtained for CQ and ART were used to validate the method used in this study [69, 70]. At pH 7.0 the log P value of both references is positive (0.6 and 0.35 respectively). Meanwhile, at pH 5.0 and pH 3.0, the log P values became negative, which confirms that the lipophilic behavior of these substances is highly dependent on the pH of the medium. Similarly, complex 2 exhibits hydrophilic behavior at pH 3.0 but shifts toward a preference for lipophilic environment at pH 5.0 and 7.0. In contrast, for complex 1 and L, the log P is negative across all studied pH values, indicating a consistent affinity for the aqueous phase regardless of the medium’s acidity. These findings suggest that these compounds may have difficulty when crossing biological membranes [71–73], potentially resulting in lower bioavailability compared to complex 2. The pH-dependent lipophilicity of complex 2 suggests a more versatile pharmacokinetic profile, potentially enhancing its ability to penetrate lipid bilayers under physiological conditions. While lipophilic behavior in acidic conditions seems to favor antiplasmodial activity, it has not been possible to establish a direct linear correlation between the values of molecular volume, partition coefficient and biological activity. These findings suggest that the ability of these compounds to circumvent parasite resistance may not rely on a single physicochemical property, but rather on a complex interplay of multiple factors and molecular targets. ADMET analysis of the compounds As part of a preliminary pharmacokinetic study, some of the physicochemical and ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) properties pharmacokinetic properties of the compounds were evaluated using the SwissADME online tool to assess their potential viability [44, 74]. We compare the properties of the ligand L and complexes 1 and 2 with two well-established antimalarials, CQ and ART. Table 4 summarizes the ADMET profiles of the studied compounds. Table 4ADMET-related properties of compoundsCompoundsLComplex 1Complex 2CQARTPhysicochemical propertiesMW (g/mol)401.10854.10704.11515.86282.33Log P (W log P)−5.68−2.992.882.762.39Hba121105Hbd00070LRo5 Viol01130TPSA (Å2)30.5165.2232.61203.3053.99Pharmacokinetic propertiesLog Sw (ESOL)−4.82−7.84−6.29−1.59−3.42GI absorptionLowLowHighLowHighCYP1A2 inhibitorNoNoNoNoYesCYP2C19 inhibitorNoNoNoNoNoCYP2C9 inhibitorNoNoNoNoNoCYP2D6 inhibitorNoNoNoNoNoCYP3A4 inhibitorNoNoNoNoNoMW molecular weight, log(Pow) partition coefficient calculated by W log P method, Hba number of hydrogen bond acceptors, Hbd number of hydrogen bond donors, LRo5 Viol Lipinski’s rule violations, TPSA topological polar surface area, Log Sw water solubility, GI absorption, gastrointestinal absorption ADMET-related properties of compounds MW molecular weight, log(Pow) partition coefficient calculated by W log P method, Hba number of hydrogen bond acceptors, Hbd number of hydrogen bond donors, LRo5 Viol Lipinski’s rule violations, TPSA topological polar surface area, Log Sw water solubility, GI absorption, gastrointestinal absorption The evaluated compounds exhibited some significant differences in their profiles. The first is the difference in molecular weight, the metal complexes exceed the upper limit of 500 g/mol established by the Lipinski’s Rule of Five (LRo5) [75]. While a higher molecular weight can sometimes indicate a lower likelihood of good absorption and oral permeability [76], it is important to note the reference drug CQ also exceeds this threshold. This suggests that for this class of compounds, a slight deviation from the Ro5 mass limit does not necessarily preclude therapeutic potential. The partition coefficient (W log P), a key indicator of lipophilicity, which may relate to ease of biological membrane penetration, revealed contrasting profiles among the compounds. L and 1 showed negative values, indicating high hydrophilicity. In contrast, complex 2, CQ and ART exhibited positive values, reflecting high lipophilicity. These findings correlate strongly with the lipophilicity values (log P) obtained experimentally by the stir-flask method in an n-octanol/buffer system at pH 7.0. This correlation validates both the computational and experimental approaches, further highlighting the superior lipophilicity for 2 compared to the other evaluated compounds. The number of hydrogen bond acceptors (HBA) and number of hydrogen bond donors (HBD) for L, 1 and 2 are within the limits of LRo5. This indicates a favorable potential for membrane permeation and intestinal absorption [77–80]. While complexes 1 and 2 show one violation of Lipinski’s rule (molecular weight), L does not show any violation. Overall, these parameters indicate a favorable drug-like profile compared to ART and CQ. Intestinal absorption and the ability to cross the blood–brain barrier (BBB) are linked to the Topological Polar Surface Area (TPSA) [81–83]. Compounds L, 1 and 2 have low TPSA values, it has been suggested that compounds with values below 140 Å2 generally have good oral absorption. While all tested compounds showed relatively low aqueous solubility, complex 2 stands out for its predicted high gastrointestinal (GI) absorption, a profile remarkably similar to that of ART. This combination of low TPSA and favorable absorption suggests that complex 2 possesses the most promising pharmacokinetic potential among the new complexes [84]. Biotransformation via enzymatic metabolism is a critical factor in determining a drug’s fate within the body. The Cytochrome P450 (CYP450) system is the major hepatic enzyme complex involved in phase 1 metabolism (oxidation) and therefore represents a key pharmacokinetic target [85]. In humans, the 57 CYP genes from 17 families have been identified, but only the CYP1, CYP2, CYP3, and CYP4 families are involved in drug metabolism, mediating the biotransformation of more than 90% of medications that go through phase I metabolism [86, 87]. Importantly, the inhibition of these CYP isoforms may lead to the inefficient or toxic effects of drugs [88]. Notably, none of the tested compounds are predicted to be an inhibitor of the CYP isoenzymes that were analyzed. This safety profile is similar to that of CQ, suggesting a low risk of metabolic interference or enzyme-related toxicity. Conclusions The IC50 values of compounds L, 1 and 2 were comparable across both chloroquine-sensitive (3D7) and chloroquine-resistant (FCR-3) P. falciparum strains, ranging from 1.46 to 6 μM. These results demonstrate that the compounds are evading the parasite’s resistance mechanism, complex 2 exhibited a minor resistance index (RI) by FCR-3 strain (0.7). Additionally, the compounds have higher selectivity indices than CQ and did not induce hemolysis in red blood cells, confirming their low systemic toxicity. The compounds have a dose-dependent effect on the progression of the P. falciparum erythrocytic cycle, resulting in the accumulation of ring and trophozoite stages at high doses in FCR-3. This dual parasitostatic and parasiticidal effect may be attributed to the compounds’ ability to act on multiple targets. Such mechanisms include alteration of pH in acidic organelles, such as the digestive vacuole and the accessory vacuoles, as well as the generation of reactive oxygen species (ROS) that disrupt parasite metabolism. Regarding transport and interaction, it was observed that while L and complex 1 could be transported by albumin, (not as effectively as CQ), complex 2 showed not significant interaction with this biomolecule. All three compounds interact with ferriprotoporphyrin IX, with complex 2 possessing the highest affinity constant (K). These findings correlate with the inhibition of β-hematin formation at the lipid-water interface, and are in agreement with those found for its biological activities. Furthermore, pH-dependent log P modifications were established: the evaluated compounds and the reference drugs exhibit a hydrophilic behavior at low pH and increased lipophilicity at physiological pH, which favors their biological activity. In summary, these metal–NHC complexes exhibit moderate antiplasmodial activity, positioning them among the most important in the literature for this type of compounds. Their lipophilicity, protein interaction, inhibition of hemozoin formation, and ROS production are linked to the nature of the carbene ligand. These ligands interact with metal centers primarily through strong σ-donation and, to a lesser extent, through π-back-donation. These results highlight the potential of metal–NHC architectures in the search for new antimalarial compounds. Future work will focus on expanding cytotoxicity assays in human cell lines to further establish safety. Subsequently, the most promising candidates may be evaluated in murine models to assess their in vivo efficacy and pharmacokinetic profiles. Supplementary Information Additional file 1. Author contributions William Castro: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing—original draft, Writing—review and editing. Hector R. Rangel: Investigation, Methodology. Liseth Garibaldi: Investigation, Methodology. Wilma Rojas: Investigation, Methodology. Joe Berroteran: Investigation, Methodology, Validation, Data curation, Visualization, Software. Oriel Sanchez-Velasco: Investigation, Methodology—review and editing. Maria Cristina Goite: Investigation, Methodology. Miguel A. Díaz: Investigation, Methodology. David Coll: Investigation, Methodology. Mariana Hidalgo: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing—original draft. Funding The study was funded by Fondo Nacional de Ciencia Tecnología e Innovación de Venezuela (FONACIT), Abordaje Epidemiológico de la Malaria en Zonas Endémicas de Venezuela Project Nro. 2024PGP445. Availability of data and materials No datasets were generated or analysed during the current study. All data generated or analysed during this study are included in this published article [and its supplementary information files]. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. William Castro and Mariana Hidalgo have contributed equally to this work. References Associated Data Supplementary Materials Additional file 1. Data Availability Statement No datasets were generated or analysed during the current study. All data generated or analysed during this study are included in this published article [and its supplementary information files].