Trends Sci. 2026; 23(1): 11331

Malaria continues to be a pressing global health concern, with an estimated 249 million cases and more than 608,000 fatalities reported in 2022 [1,2]. The disease disproportionately affects sub-Saharan Africa, which accounts for 95% of malaria-related deaths, particularly impacting vulnerable populations such as young children and pregnant women [3]. Among the 6 Plasmodium species infecting humans [4,5], Plasmodium falciparum is the most severe [6], causing the majority of critical cases due to its rapid proliferation in red blood cells and its potential to trigger complications like cerebral malaria and organ dysfunction [7].

Efforts to combat malaria face significant obstacles due to the rise of drug-resistant strains of Plasmodium falciparum. Resistance has emerged against commonly used antimalarial drugs [8], including chloroquine [9], SP (Sulfadoxine-Pyrimethamine) [10], and even ACTs (artemisinin-based combination therapies) [11], which are widely regarded as the most effective current treatments. This resistance is primarily driven by genetic mutations in key parasite enzymes, including pfDHFR and pfDHODH, which are critical for parasite survival [12]. pfDHODH is essential for pyrimidine biosynthesis, a crucial pathway for the production of DNA and RNA, as the parasite is unable to utilize a pyrimidine salvage mechanism. [13]. Inhibitory compounds targeting the pfDHODH enzyme hold potential as antimalarial drugs because they disrupt a vital metabolic pathway essential for the parasite’s survival, by disrupting parasite replication by selectively interfering with this pathway [14]. Similarly, pfDHFR is a key enzyme involved in folate metabolism required for DNA synthesis [15]. While antifolate drugs like pyrimethamine have been effective in targeting pfDHFR, resistance mutations necessitate the development of novel inhibitors that can overcome these mutations while maintaining potency [16].

Quinazolinones have emerged as promising candidates for addressing drug resistance in malaria treatment nowadays. These heterocyclic compounds exhibit diverse pharmacological activities, including potent antimalarial effects [17,18]. Their structural adaptability allows for chemical modifications that enhance their activity against resistant strains of Plasmodium falciparum. Their structural flexibility allows for chemical modifications that enhance their activity against resistant strains of Plasmodium falciparum [19]. Remarkably, the presence of functional groups such as fluorine (F), chlorine (Cl), bromine (Br), methoxy (OCH₃), nitro (NO₂), and furan rings into the quinazolinone scaffold has been shown to significantly improve antimalarial efficacy. In vitro studies (Figure 1) have demonstrated that quinazolinone derivatives bearing these functional groups exhibit IC₅₀ values ranging from 1 - 3 μM against Plasmodium falciparum, highlighting their potential as effective antimalarial agents [20,21].

Figure 1 Quinazolinones with their antiplasmodial activities.

To accelerate the discovery of novel antimalarial compounds, in silico methods have become essential tools due to their efficiency and accuracy [22]. Among these, molecular docking, molecular dynamics simulation, and ADMET studies play critical roles in screening and optimizing the best candidate compounds [23]. Molecular docking analyzes the interactions between ligands and key amino acid residues within the target protein, as well as calculates binding energy to predict binding affinity [24]. Molecular dynamics simulation provides insights into the stability of protein-ligand complexes over time under physiological conditions, allowing evaluation of the flexibility and durability of these interactions [25]. ADMET studies assess the pharmacokinetic properties and safety profiles of drug candidates, which are vital for predicting their behavior in the human body [26].

Several studies summarized in Table 1 demonstrate a direct correlation between strong interactions at key amino acid residues, favorable binding energy values, and low IC₅₀ values obtained from in vitro assays. This strong correlation validates in silico study as a reliable predictor of biological activity. However, despite these advances, the number of in silico studies focused on quinazolinone derivatives as antimalarial agents remains limited. Expanding such computational investigations could offer valuable insights into their potential to inhibit malaria-related proteins and aid in identifying promising new drug candidates for further experimental validation.

Table 1 Computational study of quinazolinone derivatives against several receptors.

This study evaluated quinazolinone derivatives as potential pfDHFR and pfDHODH inhibitors through molecular docking, molecular dynamics simulations, and ADMET analysis. The multi-target approach and predictive analysis employed in this study have not been applied to quinazolinone research before, thus providing new insights into antimalarial drug resistance development. Therefore, the results of this study offer innovative solutions to address the growing challenge of antimalarial drug resistance.

Materials and methods

Materials

The 3-dimensional (3D) single-crystal structures of P. falciparum, i.e. pfDHFR (PDB: 1J3I) and pfDHODH (PDB:1TV5), were obtained from the Protein Data Bank (www.rcsb.org). WR99210 acted as the native ligand for both proteins, with thirty quinazolinones used as ligand models in the study. Computational analyses were carried out on a computer with an Intel® Xeon processor CPU E5-2650 [email protected] GHz. The study made use of GaussView 5.0, Gaussian 09W, Chimera 1.13.1, AutoDock 4.2, Discovery Studio 2019, and YASARA software.

Molecular docking analysis

The molecular docking protocol used in this study followed a previously reported procedure with slight modifications [30]. For the redocking analysis, the protein and ligand structures of pfDHFR (PDB: 1J3I) and pfDHODH (PDB: 1TV5) were prepared using Chimera. The 3D structures of quinazolinone derivatives (1 - 30) were generated with GaussView 5.0 and optimized using the DFT B3LYP 6-311G(d,p) method in Gaussian 09W, except for compound 26, which was optimized using the AM1 method. The 6-311G(d,p) basis set is not suitable for iodine, as it is designed only for atoms ranging from hydrogen (H) to chlorine (Cl) [31,32]. For optimizing compounds containing iodine, the AM1 method represents a possible alternative [33]. The optimized structures were saved in PDB format. Redocking was conducted with AutoDock4 using a grid box size of 40×40×40 ų and 100 runs of the Lamarckian Genetic Algorithm (LGA). The grid box coordinates were set to 27.665, 6.653, and 58.206 (x, y and z) for the pfDHFR protein and 38.204, 35.063, and 36.245 (x, y and z) for the pfDHODH protein. The method was considered valid for further docking analysis if the RMSD value was below 2 Å. The 2D structures of the quinazolinone derivatives are presented in Table 1. The docking of all compounds into the receptor binding sites was performed using the same parameters as those for redocking (grid map size and LGA). The visualization of protein-ligand interactions was carried out using Discovery Studio 2019, focusing on the pose with the lowest binding affinity.

Molecular dynamics simulation

Molecular dynamics simulations for the quinazolinone complexes were conducted using YASARA software. The `md_run.mcr` macro was applied for simulations involving pfDHFR (PDB: 1J3I), while `md_runmembrane.mcr` was utilized for pfDHODH (PDB: 1TV5). The system employed periodic boundary conditions and the AMBER14 force field. The simulation box dimensions were set to 100×100×100 ų, with conditions mimicking a physiological environment: Temperature at 310 K, pressure at 1 atm (NPT ensemble), pH 7.4, and 0.9% NaCl. The system underwent energy minimization using the steepest descent approach, followed by simulated annealing minimization, reaching a final density of 0.993 g/mL. A 250 ps equilibration phase was performed to stabilize the system before the production simulations, which ran for 100 ns with a timestep of 2.5 fs. Snapshots were saved every 100 ps. Following simulation, analysis was conducted using Yasara’s md_analyze.mcr macro to compute RMSD, RMSF, RoG (Radius of Gyration), SASA (Solvent Accessible Surface Area), and DSSP (Dictionary of Secondary Structure of Proteins) values. The BEcalculation.mcr macro was also used to evaluate MM-PBSA (Molecular Mechanics Poisson-Boltzmann Surface Area) binding energy.

Secondary structure analysis

The 3D protein model was validated through a secondary structure analysis utilizing the RAM (Ramachandran) RAM plot, which assesses the dihedral angles ψ (psi) and φ (phi) of amino acid residues. The RAM plot for the protein under study was acquired from the PDBSum online platform [34].

ADMET study

All compounds were converted into SMILES format using Discovery Studio and then uploaded individually to the pkCSM server [35] and the ProTox-3.0 server [36]. Physicochemical and ADMET properties were analyzed by selecting the ADMET option. The pkCSM server generated data for each compound, including Lipinski’s rule of 5, as well as details on absorption, distribution, metabolism, excretion, and toxicity.

Results and discussion

Molecular docking of quinazolinones against pfDHFR enzyme

Our study was began with the redocking of the native ligand WR99210 onto the pfDHFR enzyme structure from PDB ID 1J3I to validate the molecular docking protocol. The resulting RMSD value was 1.24 Å, which is well within the acceptable threshold of ≤ 2.0 Å for a reliable molecular docking procedure [37]. This indicates that the docking method successfully reproduced the experimental pose of the ligand. The binding energy of WR99210 was calculated to be −5.59 kcal/mol, with a binding constant of 79.47 μM, as shown in Table 2, and the superimposed docking pose is visualized in Figure 2(a). These results confirm the validity of the docking protocol and provide a basis for further studies on potential inhibitors targeting pfDHFR enzyme.

The redocking of the native ligand revealed key interactions with the pfDHFR enzyme. Hydrogen bond interactions were observed with Ile164 (2.20 Å), Ile14 (2.14 Å), and Asp54 (1.83 Å), while hydrophobic interactions involved Pro113, Ile113, Val45, Met55, Phe58, and Leu46. Additionally, a carbon-hydrogen bond was formed with Ser111 and Ile164. These interactions demonstrated the ligand’s strong affinity for the active site of the enzyme. The results highlight the importance of hydrogen bonds and hydrophobic interactions in stabilizing the ligand-enzyme complex, as visualized in Figure 2(b).

Figure 2 (a) The overlay of WR99210 as the native ligand, displayed in gray before the redocking process and in blue after redocking, (b) The depiction of chemical interactions between WR99210 and the active site of the pfDHFR enzyme.

The chemical structures of quinazolinones were optimized by using the DFT-B3LYP 6-311G(d,p) level of theory, a method commonly applied for quinazolinones [38]. Employing this computational approach is crucial for designing effective antimalarial inhibitors, as it accurately models molecular geometry and electronic properties important for binding. It precisely calculates hydrogen-bonding and binding energies, which are key to inhibitor efficacy; strong hydrogen bonds and optimal binding energies improve stability and bioactivity, helping prioritize promising compounds for drug development resistant to parasitic resistance [39,40]. In this study, compound 1 was served as a reference to assess the impact of substituent addition on the inhibitory activity of the proposed quinazolinone derivatives against the pfDHFR enzyme.

The docking results of quinazolinone derivatives against the pfDHFR enzyme are presented in Table 2. Compound 2 exhibits a key hydrogen bond with Ile164 (2.00 Å) and an additional hydrogen bond with Tyr170 (2.19 Å). Unfortunately, its binding energy is only −4.78 kcal/mol, which is not significantly better than the native ligand. To improve the binding energy, modifications were made to the quinazolinone scaffold by adding N-phenyl and C-phenyl groups with various substituents, including hydroxyl (OH), fluorine (F), chlorine (Cl), bromine (Br), methoxy (OCH₃), and nitro (NO₂).

The quinazolinone compounds were designed with phenyl substituents at key positions such as C-2 and N-1 as structure-activity relationship (SAR) studies consistently indicate that these groups are crucial for strong inhibitory activity against enzymes like pfDHFR in malaria. The phenyl group at the C-2 position enhances hydrophobic interactions and binding stability within the enzyme’s hydrophobic pocket, increasing the inhibitor’s potency, while a phenyl group at the N-1 position modulates the electronic and steric properties of the molecule and facilitates π-π stacking with amino acid residues near the active site, both of which are essential for high binding affinity and selectivity. Phenyl substituents were specifically chosen over heterocycles like pyridine or furan because heterocycles can introduce extra polarity or charge that may reduce affinity for the hydrophobic pocket or disrupt the stability of the enzyme-inhibitor complex; furthermore, heterocycles may alter the molecule’s electronic and steric characteristics in ways that decrease desired biological activity or cause unwanted side effects, with SAR and docking studies confirming that phenyl groups generally provide superior inhibitory activity and interaction stability. By placing phenyl groups at both the C and N positions, interactions with different regions of the enzyme’s active site are maximized, creating a synergistic effect and strengthening overall binding affinity; this approach is considered more effective than substituting only 1 position, as it ensures optimal inhibitory effects due to more comprehensive interactions [41-44].

This approach proved to be effective. However, only compounds 3, 5, and 12 demonstrated hydrogen bond interactions with key amino acid residues. Compound 3 forms a key hydrogen bond with Ile164 (2.10 Å) and an additional hydrogen bond with Ala16 (2.63 Å). It also exhibits hydrophobic interactions with Ile14, Ala16, Cys15, Ile164, and Leu40. This compound has a binding energy of −8.62 kcal/mol and a binding constant of 0.48 μM. Compound 5 forms a key hydrogen bond with Asp54 (1.82 Å) and hydrophobic interactions with Leu40, Leu46, Ala16, Ile14, Phe58, Ile164, Met104, and Ile112. Its binding energy is −8.46 kcal/mol with a binding constant of 0.62 μM. Meanwhile, compound 12 forms a key hydrogen bond with Asp54 (2.23 Å) and an additional hydrogen bond with Ser111 (2.88 Å). It also exhibits hydrophobic interactions with Leu40, Ala16, Ile14, Cys15, Ile164, as well as carbon-hydrogen bonds with Cys15 and Asp54. This compound has a binding energy of −8.49 kcal/mol and a binding constant of 0.59 μM. Based on these data, compound 3 is identified as the best inhibitor against pfDHFR due to its superior binding energy compared to the other compounds.

Table 2 The molecular docking results of quinazolinone derivatives in the active site of pfDHFR (PDB: 1J3I).

Molecular docking of quinazolinones against pfDHODH enzyme

Redocking was performed using the native ligand A26 against the PfHODH enzyme with the PDB ID 1TV5. The resulting RMSD value was 1.70 Å, with the superimposed pose shown in Figure 3(a). Additionally, the binding energy was calculated to be −5.06 kcal/mol, and the binding constant was determined to be 193.88 μM, as presented in Table 3. Key hydrogen bonds were observed with Met536 (2.54 Å), Arg265 (1.88 Å), and His (1.91 Å). Hydrophobic interactions were identified with Leu172, Cys175, Phe188, Cys184, and Val532, while Carbon-Hydrogen bonds were formed with Gly535 and Gly181. Furthermore, a halogen interaction (fluorine) was noted with Gly535. These interactions are visualized in Figure 3(b).

Figure 3 (a) The overlay of A26 as the native ligand, displayed in gray before the redocking process and in red after redocking, (b) The depiction of chemical interactions between A26 and the active site of the pfDHODH enzyme.

Docking of quinazolinone derivatives was performed at the same binding site as A26, the native ligand. The docking results for the pfHODH enzyme are summarized in Table 3. Compound 1 exhibited key hydrogen bond interactions with the amino acids Arg265 (2.41 Å) and His185 (2.05 Å), along with an additional hydrogen bond with Gly181 (2.30 Å). Furthermore, it demonstrated hydrophobic interactions with Val532 and Cys184, as well as a Pi-donor hydrogen bond with Cys184. This compound showed a binding energy of −5.54 kcal/mol, which is better than that of the native ligand, and a binding constant of 86.94 μM.

Modifications to the quinazolinone framework were implemented to enhance binding energy by incorporating N-phenyl and C-phenyl groups with various substituents, including hydroxyl (OH), fluorine (F), chlorine (Cl), bromine (Br), methoxy (OCH₃), and nitro (NO₂). All proposed compounds exhibited key interactions similar to the native ligand A26. Among these, compounds 2, 11, and 20 demonstrated the most promising results. Compound 2 showed a binding energy of −9.59 kcal/mol, with a binding constant of 0.09 μM and an RMSD of 0.26 Å. It formed crucial hydrogen bonds with Arg265 and His185, similar to the native ligand, and hydrophobic interactions with Ile263, Leu531, Val532, Cys175, Leu176, and Cys184. Additionally, it engaged in Carbon Hydrogen Bond, Pi-Donor Hydrogen Bond, and Pi-Sulfur interactions with Cys184. Compound 11 had a binding energy of −10.12 kcal/mol, a binding constant of 0.04 μM, and an RMSD of 0.87 Å. It shared hydrogen bonding with Arg265, similar to the native ligand, and hydrophobic interactions with Ile263, Val532, Phe188, Leu531, Phe171, Met536, Leu172, Leu176, Cys175, and Cys184. It also formed Carbon Hydrogen Bond and Pi-Donor Hydrogen Bond interactions with Cys184 and Gly181, as well as a Pi-Sulfur interaction with Cys184. Compound 20 exhibited a binding energy of −9.96 kcal/mol, a binding constant of 0.11 μM, and an RMSD of 0.24 Å. It formed hydrogen bonds similar to the native ligand with His185 and Arg265, and hydrophobic interactions with Ile263, Cys184, Val532, Leu531, and Leu172. It also engaged in Pi-Sulfur interactions with Cys184 and Cys175. These findings highlight the potential of these compounds as effective ligands due to their strong interactions with the target protein.

Table 3 The molecular docking results of quinazolinone derivatives in the active site of pfDHODH (PDB: 1TV5).

Molecular dynamics simulations against pfDHFR and pfDHODH protein

The molecular dynamics simulations provided valuable insights into the stability and binding efficacy of selected quinazolinone derivatives against the targets pfDHFR and pfDHODH. From the molecular docking studies, 3 promising compounds were identified for each protein target. RMSD analysis, presented in Figure 4, further elucidated the dynamic behavior of these protein-ligand complexes throughout the simulation period.

Notably, compound 12 exhibited a backbone RMSD of 3.07 ± 0.58 Å, which is higher than those of compound 3 (2.17 ± 0.23 Å), compound 5 (2.12 ± 0.28 Å), and the native ligand (2.16 ± 0.25 Å). Generally, an RMSD value below 3 Å indicates a stable protein-ligand complex, suggesting that compounds with lower RMSD values maintain more stable interactions during the simulation.

For pfDHODH inhibition, compound 20 showed an average RMSD of 2.05 ± 0.21 Å, which was lower than compound 2 (2.28 ± 0.25 Å), compound 11 (2.20 ± 0.13 Å), and the native ligand (2.33 ± 0.56 Å). These findings imply that compounds 3 and 5 provide greater structural stability when bound to pfDHFR, while compounds 20, 2, and 11 display comparable stability toward pfDHODH.

The relatively consistent and low RMSD values across these compounds highlight their potential as effective inhibitors for further antimalarial drug development [45]. Structural stability is critical because it reflects the ability of a compound to maintain a stable binding conformation over time, which often correlates with sustained inhibition and better therapeutic efficacy [46]. Therefore, these RMSD results reinforce the significance of dynamic stability as a key parameter guiding the selection of promising drug candidates targeting these enzymes.

Figure 4 RMSD of the backbone for (a) compound 3 (red line), compound 5 (yellow line), compound 12 (green line), and WR99210 (blue line) against pfDHFR and for (b) compound 2 (orange line), compound 11(grey line), compound 20 (purple line), and A26 (blue line) against pfDHODH.

The analysis of RMSF values for the protein-ligand complexes provided valuable insights into the flexibility and stability of the systems. RMSF analysis was also performed as shown in Figure 5. For pfDHFR, the RMSF values were as follows: Compound 3 exhibited an RMSF of 1.60 ± 1.10 Å, compound 5 showed an RMSF of 1.62 ± 1.11 Å, and compound 12 had an RMSF of 1.76 ± 1.43 Å, while the native ligand displayed an RMSF of 1.57 ± 1.04 Å. Notably, certain residues (1 - 3, 25 - 34, 43 - 45, 85 - 97, 115, 116, 132 - 139, and 230 - 233) demonstrated RMSF values exceeding 2 Å, indicating higher flexibility in these regions.

Figure 5 RMSF (a) compound 3 (red line), compound 5 (yellow line), compound 12 (green line), and WR99210 (blue line) against pfDHFR and (b) compound 2 (orange line), compound 11(grey line), compound 20 (purple line), and A26 (blue line) against pfDHODH.

In contrast, for pfDHODH, the RMSF values were: compound 2 had an RMSF of 1.41 ± 0.82 Å, compound 11 showed an RMSF of 1.29 ± 0.54 Å, compound 20 exhibited an RMSF of 1.36 ± 0.78 Å, and the native ligand displayed an RMSF of 1.59 ± 0.98 Å. Specific residues (159 - 161, 166, 171, 199, 203, 301, 302, 376 and 383 - 414) in pfDHODH also exhibited RMSF values greater than 2 Å, highlighting regions of increased flexibility within this enzyme. These findings suggest that although the ligands generally stabilize the protein structures, certain regions remain relatively flexible, which could influence binding dynamics and enzymatic function. Understanding these flexible regions is essential because they may affect the overall efficacy and mechanism of inhibition of these potential antimalarial compounds.

Further RAM analysis was also performed as shown in Figure 6. The pfDHFR complexed with compound 3 generated the most favored regions with 192 residues (88.1%), additional allowed regions with 26 residues (11.9 %), generously allowed regions with 0 residues (0%), and disallowed regions with 0 residues (0%). Meanwhile, the pfDHFR complexed with compound 5 generate the most favored regions with 185 residues (84.9%), additional allowed regions with 31 residues (14.2%), generously allowed regions with 2 residues (0.9%), and disallowed regions with 0 residues (0%). Moreover, the PfDHFR complexed with compound 12 generate the most favored regions with 191 residues (87.6%), additional allowed regions with 23 residues (10.6%), generously allowed regions with 2 residues (0.9%), and disallowed regions with 2 residues (0.9%). These results revealed that almost all regions of pfDHFR complexed with quinazolinones were located in allowed regions according to the RAM plot. The DSSP of each ligand in the active site of pfDHFR agreed with the RAM data that each helix, sheet, turn, coil, helix310, and helixPi fractions of pfDHFR remains stable during the 100 ns molecular dynamics simulations as shown in Figure 7.

The pfDHODH complexes with compounds 2, 11, and 20 exhibited favorable structural characteristics based on RAM plot analysis in Figure 6. Specifically, the complex with compound 2 showed 325 residues (87.8%) in the most favored regions, 41 residues (11.1%) in additional allowed regions, 3 residues (0.8%) in generously allowed regions, and 1 residue (0.3%) in disallowed regions. The complex with compound 11 had 326 residues (88.1%) in the most favored regions, 43 residues (11.6%) in additional allowed regions, and 1 residue (0.3%) in disallowed regions, with no residues in generously allowed regions. Meanwhile, the complex with compound 20 featured 318 residues (85.9%) in the most favored regions, 48 residues (13.0%) in additional allowed regions, 2 residues (0.5%) in generously allowed regions, and 2 residues (0.5%) in disallowed regions. These results indicate that nearly all regions of pfDHODH complexed with quinazolinones fell within allowed regions according to the RAM plot. Furthermore, the Dictionary of Secondary Structure of Proteins (DSSP) analysis for each ligand in the pfDHODH active site supported the RAM data, showing that the fractions of helices, sheets, turns, coils, helix310, and helixPi in pfDHODH remained stable throughout the 100 ns molecular dynamics simulations.

Figure 6 RAM plot for (a) compound 3, (b) compound 5, and (c) compound 12 against pfDHFR and for (d) compound 2, (e) compound 11, and (f) compound 20 against pfDHODH.

Figure 7 DSSP analysis for (a) compound 3, (b) compound 5, and (c) compound 12 against pfDHFR and for (d) compound 2, (e) compound 11, and (f) compound 20 against pfDHODH.

Figure 8 (a) RoG of the backbone and (b) SASA during 100 ns simulation for compound 3 (red line), compound 5 (yellow line), compound 12 (green line), and WR99210 (blue line) against pfDHFR.

Figure 9 (a) RoG of the backbone and (b) SASA during 100 ns simulation for compound 2 (orange line), compound 11(grey line), compound 20 (purple line), and A26 (blue line) against pfDHODH.

The compactness of the pfDHFR protein complexed with quinazolinones is effectively captured by the RoG values. As illustrated in Figure 8, the RoG values for compounds 3, 5, 12, and WR99210 were determined to be 19.28 ± 0.13, 18.92 ± 0.14, 19.25 ± 0.12 and 19.30 ± 0.13 Å, respectively. These values remained relatively stable and did not exhibit significant differences throughout the molecular dynamics simulations. Additionally, the SASA values, also presented in Figure 8, provide insight into the accessible surface area of the pfDHFR protein complex for each ligand. The SASA data for compounds 3, 5, 12, and WR99210 were found to be 13,444 ± 222, 13,016 ± 323, 13,447 ± 305 and 13,390 ± 346 Ų, respectively. These SASA values align with the stable RoG values, indicating that both metrics support the overall stability of the pfDHFR complexes during the simulations.

Additionally, the RoG values of pfDHODH protein complexed with quinazolinones of compound 2, 11, 20 and A26 were 21.30 ± 0.11, 21.20 ± 0.08, 21.20 ± 0.07 and 21.40 ± 0.11 Å, respectively. These values were not significantly different and relatively stable during the molecular dynamics simulations. On the other hand, SASA value represent the accessible surface area of pfDHODH protein complex for each ligand. As presented in in Figure 9(b), The SASA data of compound 2, 11, 20 and A26 were 18,431 ± 288, 18,537 ± 306, 17,987 ± 309, and 18,127 ± 308 Ų. These values were in line with the RoG value was stable during the molecular dynamics simulations.

The energy calculations for the pfDHFR complexes with compounds 3, 5, 12, and WR99210 provide valuable insights into their binding characteristics (Table 4). The potential energy of the receptor-ligand interactions (EpotRecept+) showed that WR99210 had the most favorable value at −338.92 ± 22.50 kJ/mol, followed by compound 3 at −153.63 ± 21.28 kJ/mol. The desolvation energy of the receptor (EsolvRecept+) highlighted significant differences, with WR99210 having the most negative value at −488.52 ± 23.66 kJ/mol, indicating strong interactions. The ligand’s potential energy (EpotLigand+) and desolvation energy (EsolvLigand-) also varied, reflecting distinct binding modes. Notably, the calculated binding energies revealed that WR99210 exhibited the most favorable binding energy at −141.81 ± 47.72 kJ/mol, while compound 12 had a positive binding energy, suggesting less favorable interactions. Overall, these energy calculations suggest that WR99210 forms the most stable complex with pfDHFR, followed by compounds 5 and 3, with compound 12 being less stable.

Table 4 MM-PBSA calculation for compounds 3, 5, 12, and WR99210 after molecular dynamics simulations in the active site of pfDHFR.

The energy calculations for the pfDHODH complexes with compounds 2, 11, 20, and A26 offer insights into their binding properties (Table 5). The potential energy of the receptor-ligand interactions (EpotRecept+) showed that compound 11 had the most favorable value at −99.25 ± 21.07 kJ/mol, while compound 2 and compound 20 had less favorable values. The desolvation energy of the receptor (EsolvRecept+) highlighted significant differences, with compound 11 having the most negative value at −67.21 ± 9.52 kJ/mol, indicating strong interactions. The ligand’s potential energy (EpotLigand+) and desolvation energy (EsolvLigand−) varied, reflecting distinct binding modes. Notably, the calculated binding energies revealed that all compounds had positive values, with A26 exhibiting the highest at 156.70 ± 28.77 kJ/mol, suggesting less favorable interactions compared to negative binding energies typically associated with stable complexes. Overall, these energy calculations suggest that while compound 2 and compound 11 shows favorable receptor-ligand interactions, the positive binding energies indicate that these complexes may not be as stable as those with negative binding energies.

Table 5 MM-PBSA calculation for compounds 2, 11, 20, and A26 after molecular dynamics simulations in the active site of pfDHODH.

The pharmacokinetic parameters of compounds 2, 3, 5, and 11 (Table 6) provide comprehensive insights into their drug-like properties and potential as therapeutic agents. In terms of drug-likeness, all compounds have similar molecular weights, with compound 11 being slightly heavier at 358.39 g/mol. The log P values, which indicate lipophilicity, are also comparable, ranging from 3.76 to 4.16, suggesting moderate lipophilicity. The number of hydrogen bond donors and acceptors varies slightly, with compounds 11 having an additional acceptor. The topological polar surface area (TPSA) is higher for compounds 11 and 12, which could affect their permeability and solubility. According to that criterias, none of the compounds show any violations of drug-likeness properties.

Regarding absorption, compounds 11 exhibits better water solubility compared to compounds 2, 3, and 5, which may enhance their bioavailability. The human intestinal absorption (HIA) is high for all compounds, indicating good absorption potential. Caco-2 permeability values suggest moderate permeability across cell membranes. In terms of distribution, the log BBB permeability values indicate that compound 11 may has slightly better brain penetration, although all compounds show limited CNS permeability.

Table 6 ADMET results of quinazolinones.