Full Text:
Dengue virus (DENV) and Zika virus (ZIKV) belong to the Flaviviridae family and are transmitted by Aedes mosquitoes. These two viruses pose significant risks to public health on a global scale due to their widespread distribution and serious health impacts [1]. DENV is found in over 100 countries and affects approximately 390 million people annually, with about 96 million cases manifesting clinical symptoms [2]. Severe consequences such as dengue hemorrhagic fever and dengue shock syndrome can lead to substantial illness and mortality, especially in tropical and subtropical regions. The ZIKV outbreak in the USA in 2015–2016 also raised global concerns. While ZIKV typically causes mild illness, it is linked with serious birth defects including microcephaly and neurological conditions such as Guillain-Barré syndrome [3]. Factors like temperature, humidity, and precipitation influence the population of Aedes mosquitoes leading to increased cases and outbreaks of the diseases in regions including the USA, Southeast Asia, and the Pacific Islands. Currently, effective treatments for these viruses are lacking [4]. The highly contagious nature and serious medical effects of these diseases underscore the urgent need for effective antiviral therapies [5].
DENV and ZIKV share a similar protein composition, consisting of three structural proteins (Capsid (C), Membrane (M), and Envelope (E)) and seven non-structural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5). Among these, NS5 is notable for its dual functions as an RNA-dependent RNA polymerase (RdRp) and a methyltransferase (MTase) [6]. The NS5 MTase domain methylates the viral RNA cap by utilising S-adenosyl-L-methionine (SAM) to produce N-7-methylguanosine and 2'-O-methyladenosine. These methylation processes are essential for flaviviruses replication, making NS5 MTase an attractive target for the development of antiviral drugs. Key areas of the NS5 MTase crystal structure include the SAM-binding pocket, cap-binding site, and positive RNA-binding site. Inhibitors like sinefungin and S-adenosylhomocysteine (SAH) have targeted these areas but face challenges due to low cellular permeability and lack of selectivity [7]. The NS2B/NS3 protease complex is essential for viral polyprotein synthesis. Together, the NS2B cofactor and NS3 protease enhance proteolytic activity by cleaving viral polyproteins at different sites, generating proteins required for viral replication [8]. Moreover, the NS3 protein also functions as a helicase, unwinding double-stranded viral RNA intermediates in an ATP-dependent manner during replication [9]. The NS3 helicase ensures that the viral RNA template is accessible and properly aligned for the NS5 RdRp to synthesize new viral RNA [10]. Due to similarities in NS3 helicase domains across flaviviruses, it represents a promising target for broad-spectrum antiviral drugs that could be effective against multiple viruses.
This study investigates the inhibitory potential of prodigiosin, a natural red pigment, on key proteins of DENV and ZIKV using molecular docking and molecular dynamics simulations. Prodigiosin, composed of three connected pyrrole rings, is biosynthesized by certain bacterial strains including Serratia [11]. Recently, a strain of Serratia nematodiphila has been identified in chrysanthemum rhizospheric soil [12], producing substantial amounts of prodigiosin. Previous research highlights prodigiosin's potent antimicrobial activity against diverse pathogenic bacteria and fungi, along with its documented induction of apoptosis in cancer cells, suggesting potential anti-cancer properties. Furthermore, prodigiosin exhibits immunosuppressive and anti-inflammatory effects, which could be beneficial in treating autoimmune diseases and inflammation [13].
Given its broad spectrum of biological activities, we hypothesize that prodigiosin could effectively inhibit viral proteins critical for key viral processes. Therefore, this study aims to conduct a comprehensive in-silico analysis to evaluate whether prodigiosin can inhibit viral replication by binding to the active sites of NS5 methyltransferase (MTase), NS2B/NS3 protease, and NS3 helicase. In-silico methods, such as molecular docking and molecular dynamics simulations, provide a powerful approach to predict and analyze the molecular interactions between antiviral drugs and viral proteins. These techniques enable detailed examination of binding affinities and the stability of protein-ligand complexes, critical for understanding the inhibitory mechanisms of potential antiviral agents. By exploring the molecular interactions between prodigiosin and selected non-structural proteins, we aim to gain insights into how prodigiosin may impede viral replication. Successful binding to these active sites could potentially reduce viral load in infected individuals and hinder pathogen dissemination. The findings from this study could lay the groundwork for the development of novel antiviral therapies targeting DENV and ZIKV, offering promising strategies to control and mitigate the devastating impact of these pathogens.
Figure 1. Summary of the methods and tools used in this study. This schematic outlines the workflow of the study, starting with the drug-likeness and ADMET analysis of prodigiosin using the SwissADME server. Ligands and control compounds were retrieved from the PubChem database and prepared using PyMOL and AutoDockTools. Six viral proteins were obtained from the RCSB PDB and processed with Biovia Discovery Studio and AutoDockTools. Molecular docking studies using AutoDock Vina assessed the binding affinities of prodigiosin to viral proteins, identifying the best-interacted pairs. These pairs were then subjected to molecular dynamics simulations using the Desmond package under the Schrödinger suite, evaluating stability and interactions through parameters like RMSD, RMSF, radius of gyration, and solvent-accessible surface area.
Drug-likeness and ADMET analysis of the ligand
A summary of the computational methods and tools employed in this study has been provided in Figure 1. Prodigiosin, the primary ligand in this study, was subjected to drug-likeness and ADMET (absorption, distribution, metabolism, excretion, and toxicity) analysis to evaluate its potential as a drug candidate. Drug-likeness and ADMET properties are vital for determining the viability of a compound in drug development. The SwissADME server (http://www.swissadme.ch/) [14], that predicts drug-like features, and pharmacokinetic properties of small molecules, was used to predict the drug-likeness of prodigiosin based on several criteria including Lipinski’s five-criterion rule [15], Ghose’s rule [16], Veber’s rule [17], Muegge’s rule [18], topological polar surface area (TPSA) and the number of rotatable bonds. The SMILES format of prodigiosin was retrieved from the ChEMBL database (https://www.ebi.ac.uk/chembl/) and used as input. For ADMET analysis, the pkCSM server [19] was utilized to analyze the ADMET properties of prodigiosin using the same SMILES format as input.
Ligand retrieval and preparation
The 3D structure of prodigiosin was obtained in SDF format from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) with PubChem CID 135455579. Ribavirin 5’-triphosphate (PubChem CID: 122108) and Chloroquine (PubChem CID: 2719) were used as control ligands against NS5 MTase of DENV and ZIKV, respectively, and were also retrieved from PubChem in 3D-SDF format. These SDF files of ligand and controls were converted to PDB format using PyMol v2.5.8 [20] (https://www.pymol.org/) and subsequently to PDBQT format by adding gasteiger charges and merging non-polar hydrogens using AutoDockTools v1.5.6 [21] (https://autodocksuite.scripps.edu/adt/).
Protein retrieval and preparation
Six different proteins were used as receptors for prodigiosin: three from DENV including NS2B/NS3 protease (PDB ID: 2FOM), NS5 methyltransferase (PDB ID: 2P41), and NS3 helicase (PDB ID: 2BMF), and three from ZIKV including NS3 helicase (PDB ID: 5JMT), NS2B/NS3 protease (PDB ID: 5YOD), and NS5 methyltransferase (PDB ID: 5ULP). The 3D structures of these proteins were obtained from the RCSB Protein Data Bank (https://www.rcsb.org/) in PDB format. Preparation of proteins involved removing water molecules and adding polar hydrogen atoms and Kollman charges using Biovia Discovery Studio 2021 [22] (https://www.3ds.com/products/biovia/discovery-studio) and AutoDockTools v1.5.6. The prepared proteins were saved in PDBQT format for the molecular docking studies. The docking grid box dimensions, set to x: 40, y: 40, z: 40, were determined using AutoDockTools v1.5.6 and remained consistent across all proteins, as detailed in Table 1.
The docking of prodigiosin and control ligands with the viral proteins was performed using Autodock Vina v1.1.2, which estimates binding energy between ligand and receptor molecules using the Lamarckian Genetic Algorithm.[23,24]. Docking was performed with a search space volume of 27000 ų and an exhaustiveness of 8 for enhanced results. The software calculates the receptor and ligand’s interacting energy and adjusts the complexes’ poses [25]. Input for AutoDock Vina v1.1.2 included PDBQT format files of ligands and proteins. Post-docking interactions of the best-docked protein-ligand complexes were analyzed and visualized using Biovia Discovery Studio 2021. The CASTp 3.0 server was utilized to predict binding pockets for selected proteins [26].
Molecular dynamics (MD) simulation
MD simulations were conducted to predict the dynamic properties and structural stability of the protein-ligand complexes under physiological conditions, evaluating the strength of prodigiosin’s binding to the target receptors. Desmond package under the Schrödinger suit was employed to analyze the MD simulation of the best-docked protein-ligand complex for 100 ns each [27]. The Protein preparation wizard was used to pre-process the protein-ligand complexes prior to simulation [28]. An orthorhombic-shaped boundary box was assigned for each complex SPC water model with the interval of (10 × 10 × 10 Å3). The salt concentration was maintained at 0.15 M by arbitrarily selecting and dispersing Na+ and Cl- ions throughout the solvated system. The OPLS3e force field was applied to reduce and relax the system [29]. Afterwards, 300.0 K temperature and 1.01325 bar pressure were used to complete the constant pressure-constant temperature (NPT) ensemble [30,31]. Each complex was first allowed to relax, and then the final analyses were conducted at 100 ps intervals by applying an energy level of 1.2 [32]. After completion of the final production, various dynamics analyses including root mean square deviation (RMSD), root mean square fluctuation (RMSF), radius of gyration (rGyr), solvent accessible surface area (SASA), and protein-ligand interactions, were calculated to assess the stability and flexibility of the complexes.
Drug-likeness and ADMET properties
Prodigiosin demonstrated favorable properties in drug-likeness and ADMET analysis. It complied with Lipinski's rule of five and passed other criteria including Ghose's, Veber's, and Muegge's filters. It exhibited a bioavailability score of 0.55 and a topological polar surface area (TPSA) of 53.17 Ų. Table 2 lists the drug-likeness properties of prodigiosin. Regarding the ADMET analysis, prodigiosin demonstrated a high rate of human intestine absorption and positive Caco-2 permeability (Table 3). Importantly, it does not function as a substrate or inhibitor of P-glycoprotein (I and II), suggesting minimal drug-drug interaction potential. Moreover, the ligand exhibited a moderate blood-brain barrier (BBB) permeability of 0.115 (Log BB) and a low central nervous system (CNS) permeability of -2.794 (Log PS), indicating limited brain penetration. In terms of metabolism, prodigiosin was found to be a substrate for CYP3A4 and an inhibitor for CYP1A2 and CYP2C9. The total clearance of prodigiosin was estimated to be 1.048 (log ml/min/kg). Additionally, prodigiosin did not exhibit AMES toxicity, hepatotoxicity, or skin sensitization, indicating a relatively low risk of adverse effects.
Prodigiosin was docked against three viral proteins of DENV (NS2B/NS3 Protease, NS3 helicase, and NS5 methyltransferase) and three viral proteins of ZIKV (NS2B/NS3 protease, NS3 helicase, and NS5 MTase) to identify the best matches as inhibitors based on binding affinity. Since the lowest binding energy predicts a strongest binding affinity between a ligand and a target receptor, the protein with the lowest binding energy for each virus was selected for further analysis. Among the DENV proteins, prodigiosin exhibited binding energies of -7.2 kcal/mol (NS2B/NS3 protease), -6.3 kcal/mol (NS3 helicase), and -7.6 kcal/mol (NS5 MTase). For ZIKV, binding energies were -6.7 kcal/mol (NS2B/NS3 protease), -7.4 kcal/mol (NS3 helicase), and -7.7 kcal/mol (NS5 MTase) (Table 4). Notably, prodigiosin exhibited the strongest binding affinity against NS5 MTases of both DENV and ZIKV. To compare the results, Ribavirin 5’- triphosphate, a known inhibitor of DENV NS5 MTase [33], was docked as a positive control. The binding energy of Ribavirin 5’- triphosphate with DENV NS5 MTase was -7.9 kcal/mol. Additionally, chloroquine, also a significant inhibitor of ZIKV NS5 MTase [34,35], was docked as a control, resulting in a binding energy of -6.5 kcal/mol. The detailed molecular docking results are presented in Table 4, and the interactions of all protein-ligand complexes are depicted in Figure 2, Figure 3, and Figure 4. The interactions between prodigiosin and key viral proteins of DENV and ZIKV, as illustrated in Figure 2, reveal significant hydrogen bonding with amino acids of NS2B/NS3 protease and NS3 helicase (GLY153, ARG225, ASP120, ASP602), along with other bonds including pi-alkyl, alkyl, pi-anion, pi-cation, pi-sigma etc. Figure 3 demonstrates the bonding of DENV NS5 MTase with prodigiosin through hydrogen and hydrophobic bonds with residues GLY81, LYS105, VAL132, and ILE147. These residues were also noted during the interaction with the control, Ribavirin 5’-triphosphate. Similarly, Figure 4 illustrates ZIKV NS5 MTase and prodigiosin interactions compared to the control ligand chloroquine. GLY81 of ZIKV NS5 MTase interacted with prodigiosin via hydrogen bonds, while GLY83 interacted through pi-sigma bonds. Hydrophobic bonds involving PHE133, VAL132, ILE147, and LYS105 were also observed. These findings underscore the diverse binding capabilities of prodigiosin, supporting its potential as a broad-spectrum antiviral agent through its multifaceted interactions with critical viral proteins. Additionally, CASTp 3.0 was used to validate the binding of the ligand in the active site of NS5 MTases of both viruses, and the list of the amino acids present in the binding site of the protein is given in Table 5.
Figure 2. Three-dimensional (3D) and two-dimensional (2D) interactions between prodigiosin and viral proteins: (A) DENV NS2B/NS3 protease, (B) DENV NS3 helicase, (C) ZIKV NS2B/NS3 protease, and (D) ZIKV NS3 helicase.
Molecular dynamics (MD) simulation
MD simulations were performed to investigate the stability and dynamics of the protein-ligand complexes between prodigiosin and the target proteins under specific physiological conditions. Since prodigiosin demonstrated the highest binding affinity against DENV NS5 MTase and ZIKV NS5 MTase in molecular docking analysis, these proteins with their ligand prodigiosin (CID: 135455579) were selected for MD simulation. Additionally, the complexes of the control ligands Ribavirin 5’-triphosphate (CID: 122108) and Chloroquine (CID: 2719) of DENV NS5 MTase and ZIKV NS5 MTase respectively, were also subjected to MD simulations for comparative analysis. This inclusion allows us to validate our simulation approach and compare the binding behavior of prodigiosin with these established inhibitors. Various parameters were analyzed to understand the conformational changes in the proteins upon ligand binding, including root mean square deviation (RMSD), root mean square fluctuation (RMSF), solvent-accessible surface area (SASA), radius of gyration (rGyr), and protein-ligand interactions.
Figure 3. Interactions between DENV NS5 MTase and ligands: (A) Three-dimensional (3D) representation of bonds between DENV NS5 MTase and prodigiosin, (B) Two-dimensional (2D) representation of bonds between DENV NS5 MTase and prodigiosin, (C) 3D representation of bonds between DENV NS5 MTase and Ribavirin 5’-triphosphate (control), and (D) 2D representation of bonds between DENV NS5 MTase and Ribavirin 5’-triphosphate (control).
Figure 4. Interactions between ZIKV NS5 MTase and ligands: (A) Three-dimensional (3D) representation of bonds between ZIKV NS5 MTase and prodigiosin, (B) Two-dimensional (2D) representation of bonds between ZIKV NS5 MTase and prodigiosin, (C) 3D representation of interactions between ZIKV NS5 MTase and the control ligand chloroquine (CID: 2719), and (D) 2D representation of interactions between ZIKV NS5 MTase and chloroquine.
RMSD measures the deviation of a protein's backbone structure in a protein-ligand complex relative to its initial conformation, providing insights into the protein’s stability, dynamic properties, and conformational changes during the simulation [36,37]. Figure 5A presents the RMSD analysis of the four protein-ligand complexes: DENV NS5 MTase-prodigiosin (CID 135455579), DENV NS5 MTase-ribavirin 5’- triphosphate (CID 122108), ZIKV NS5 MTase-prodigiosin (CID 135455579) and ZIKV NS5 MTase-chloroquine (CID 2719). Both DENV and ZIKV NS5 MTases demonstrated remarkable stability when bound to prodigiosin. The low RMSD values of 1.28 Å for the DENV NS5 MTase-prodigiosin complex and 1.41 Å for the ZIKV NS5 MTase-prodigiosin complex indicate minimal structural deviations and strong interactions during the simulation. In contrast, chloroquine (CID 2719), a positive control for the ZIKV NS5 MTase, exhibited a higher RMSD of 2.26 Å when bound to ZIKV NS5 MTase, indicating realtively less stability and greater structural fluctuations compared to prodigiosin. Interestingly, the DENV NS5 MTase-Ribavirin 5’-triphosphate (CID 122108) complex showed an average RMSD of 1.26 Å, similar to the DENV NS5 MTase-prodigiosin complex, suggesting comparable stability and effective binding. Among the four complexes, the highest RMSD values recorded were 1.66 Å, 1.69 Å, 1.87 Å, and 3.57 Å, respectively, and the lowest RMSD values were 0.96 Å, 0.79 Å, 0.85 Å, and 0.89 Å, respectively.
RMSF evaluates local fluctuations in the protein chain when ligands interact with its specific residues. In this study, RMSF analysis was conducted on the NS5 MTase proteins of DENV and ZIKV to understand their structural flexibility upon binding with prodigiosin (CID 135455579), or the control ligands Ribavirin 5’-triphosphate (CID 122108) and chloroquine (CID 2719) for DENV and ZIKV NS5 MTases respectively (Figure 5B). Prodigiosin exhibited distinct effects on both viral proteins. DENV NS5 MTase showed RMSF values of 0.70 Å with prodigiosin (CID 135455579) and 0.68 Å with Ribavirin 5’-triphosphate (CID 122108), implying similar structural stability in both complexes. Conversely, for ZIKV NS5 MTase, prodigiosin (CID 135455579) resulted in an average RMSF value of 0.78 Å, indicating relatively low local fluctuations and stable binding, in contrast to chloroquine (CID 2719), which showed an average RMSF value of 1.14 Å. Furthermore, analysis revealed that ZIKV NS5 MTase exhibited the highest fluctuations at residues 29-52 and 98-106 when bound to chloroquine, while achieving greater stability with prodigiosin. For DENV NS5 MTase, similar fluctuations were observed regardless of whether the ligand was prodigiosin or Ribavirin 5’-triphosphate. Both proteins regardless of any ligands demonstrated stability in the region of residues 140-240, with the highest fluctuations noted at the N- and C-terminal regions of the proteins, likely due to the inherent flexibility of N- and C- terminals domains in the proteins.
Figure 5. Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) analyses of NS5 MTase proteins: (A) RMSD values for the NS5 MTase proteins in four complexes: ZIKV NS5 MTase with prodigiosin (CID 135455579) (blue), ZIKV NS5 MTase with chloroquine (CID 2719) (orange), DENV NS5 MTase with prodigiosin (CID 135455579) (ash), and DENV NS5 MTase with ribavirin 5’-triphosphate (CID 122108) (yellow). (B) RMSF values for the same protein-ligand complexes, with the color scheme corresponding to the same complexes as in (A).
Figure 6. Radius of Gyration (rGyr) and Solvent Accessible Surface Area (SASA) analysis of protein-ligand complexes: (A) The radius of gyration (rGyr) values for the four complexes are shown, with ZIKV NS5 MTase – CID 135455579 (prodigiosin) in blue, ZIKV NS5 MTase – CID 2719 (chloroquine, control) in orange, DENV NS5 MTase – CID 135455579 (prodigiosin) in ash, and DENV NS5 MTase – CID 122108 (ribavirin 5'-triphosphate, control) in yellow. (B) The Solvent Accessible Surface Area (SASA) for NS5 MTase proteins in the four complexes is depicted, utilizing the same color coding as in (A) to represent the respective complexes.
The spatial arrangement of atoms within a protein-ligand complex with respect to its axis is known as the radius of gyration, or rGyr. The rGyr analysis was performed to assess the compactness of the four protein-ligand complexes in our study (Figure 6A). For the DENV NS5 MTase complex with prodigiosin (CID 135455579), the average rGyr was 4.60 Å, slightly higher than the average rGyr of 4.44 Å observed with the control ligand Ribavirin 5'-triphosphate (CID 122108). In the case of the ZIKV NS5 MTase – prodigiosin (CID 135455579) complex, the average rGyr value was 4.44 Å, indicating a stable and compact structure. This was comparable to the rGyr value of the ZIKV NS5 MTase – chloroquine (CID 2719) complex, which was 4.43 Å, suggesting that prodigiosin maintains the structural integrity of the ZIKV NS5 MTase complex similarly to the known inhibitor chloroquine.
The SASA measures the area of a protein's surface that can interact with a solvent, which is essential for understanding protein stability, folding, and interactions in various solvent environments. Hence, we calculated SASA for our selected four complexes and the values are presented in Figure 6B. Prodigiosin (CID 135455579) induced a more compact and less solvent-exposed structure in both complexes of DENV and ZIKV NS5 MTases, with relatively low average SASA values of 296.84 Ų and 284.76 Ų, respectively. However, the ZIKV NS5 MTase-chloroquine (CID 2719) complex exhibited a significantly higher SASA value of 500.30 Ų compared to the ZIKV NS5 MTase-prodigiosin (CID 135455579) complex, suggesting that prodigiosin binding resulted in a considerable reduction of solvent-exposed surface area. In contrast, ribavirin 5'-triphosphate (CID 122108), the documented inhibitor of DENV NS5 MTase, exhibited a SASA value of 221.68 Ų when bound to DENV NS5 MTase, indicating a similar effect as prodigiosin on the compactness and solvent exposure of DENV NS5 MTase.
Protein-ligand contact analysis
Protein-ligand interactions and the architecture of the protein-ligand complexes were analyzed via the simulation interaction diagram (SID) during the 100 ns MD simulation. The interactions between the proteins and ligands in our four complexes are shown in Figure 7. These interactions encompassed various bond types, including ionic bonds, hydrogen bonds, non-covalent bonds, and water bridges. Prodigiosin (CID: 135455579) made interactions with the ZIKV NS5 MTase receptor at PRO113 and LEU126, with interaction fractions (IF) of 0.33 and 0.13 respectively (Figure 7A). In contrast, the control ligand chloroquine (CID: 2719) interacted with ZIKV NS5 MTase at GLU6, TYR25, GLU99, PRO108, HIS110, GLU111, ASP146, GLU149, ARG175, ARG200, ARG201, and ALA265, with IF values ranging from 0.01 to 0.06 (Figure 7B). Prodigiosin (CID: 135455579) also showed interactions with DENV MTase at HIS110 and LYS181, with IF values of 0.25 and 0.03 respectively (Figure 7C). Conversely, Ribavirin 5’-triphosphate (CID 122108) interacted with the protein at LYS29, ARG84, HIS110, GLU149, LYS181, and ARG212, with IF values ranging from 0.3 to 2.6 (Figure 7D).
Figure 7. Bar charts presenting the interactions of protein-ligand complexes during the 100 ns simulation. (A) Interactions between ZIKV NS5 MTase and prodigiosin (CID 135455579), (B) Interactions between ZIKV NS5 MTase and chloroquine (CID 2719). (C) and (D) Interactions between DENV NS5 MTase and prodigiosin (CID 135455579) and DENV NS5 MTase and ribavirin 5'-triphosphate (CID 122108), respectively.
DENV and ZIKV continue to pose significant threats to public health globally, necessitating the urgent development of effective antiviral treatments. Despite extensive research efforts, the clinical success of antiviral candidates has been limited by issues such as efficacy and toxicity. For example, balapiravir, which was first studied for Hepatitis C, showed limited effectiveness against the DENV in the preliminary study but was unsuccessful in clinical trials [38]. Likewise, NITD008, an adenosine analogue, demonstrated potential effectiveness against the Zika virus but was not further developed because of in-vivo toxicity issues [39]. Moreover, the unique characteristics of viral proteins often require tailored inhibitors specific to each virus. These challenges underscore the critical need for novel, broad-spectrum antiviral drugs. Drug development, however, is a complex process that demands a multifaceted approach and significant time investment. Nonetheless, recent advancements in Computer-Aided Drug Design (CADD) have significantly transformed the drug discovery landscape, particularly in identifying potential antimicrobial agents. In this study, prodigiosin, a microbial tripyrrole pigment, was evaluated using computational tools to assess its suitability as an antiviral drug against DENV and ZIKV.
Our initial assessment evaluated prodigiosin's drug-likeness to ascertain whether the pigment meets essential physicochemical and pharmacokinetic criteria. Prodigiosin met the stringent criteria set by Lipinski's, Ghose's, Veber's, Muegge's, and Egan’s rules, indicating optimal molecular weight, hydrogen bond donors and acceptors, lipophilicity, TPSA, and molar refractivity. These properties are essential for ensuring adequate oral bioavailability and distribution within the body.
The ADMET analysis further explored key insights into prodigiosin’s pharmacokinetic properties, determining how the drug might behave in the body from entry to exit [40]. Caco-2 permeability and human intestine absorption rate are two important factors checked for absorption in the ADMET profile of a drug [41]. Notably, prodigiosin demonstrated high human intestinal absorption and positive Caco-2 permeability, suggesting efficient gastrointestinal absorption, a favorable characteristic for oral administration. Importantly, prodigiosin did not act as a substrate or inhibitor of P-glycoprotein, reducing the risk of drug-drug interactions. This implies an important advantage as drug interactions can lead to reduced efficacy or increased toxicity of co-administered medications. Prodigiosin's moderate BBB permeability and low CNS permeability indicate limited brain penetration, reducing the risk of adverse effects on the central nervous system. This property is particularly important for antiviral drugs as it reduces the risk of neurotoxicity. On the other hand, metabolism of drugs is vital as it converts them into active forms and detoxifies them for elimination. A broad family of heme-containing monooxygenase enzymes known as cytochrome P450s (CYPs) is responsible for the body's first-pass metabolism of medications and foreign substances. There are several subtypes of mammalian CYP enzymes and in this study, 6 of them were assessed. In our study, prodigiosin appeared to function as a substrate for CYP3A4 while inhibiting CYP1A2 and CYP2C9 enzymes. This suggests that although prodigiosin would undergo metabolism by common liver enzymes, its inhibitory effects on CYP enzymes could potentially lead to drug interactions or side effects [42]. The absence of AMES toxicity, hepatotoxicity, or skin sensitization indicates a low risk of adverse effects of prodigiosin.
Molecular docking studies confirmed prodigiosin's efficacy in interacting with multiple targets, highlighting its potential to inhibit key viral pathways. In this study, it was shown found that prodigiosin had strong binding affinity against all the different proteins of both DENV and ZIKV and the binding affinity values were lower than -6 kcal/mol. Notably, prodigiosin displayed a strong binding affinity for the NS5 MTases of both DENV and ZIKV, comparable to or even exceeding that of established inhibitors like chloroquine and Ribavirin 5'-triphosphate. This suggests that the NS5 MTases are particularly promising targets for therapeutic intervention. Given NS5 MTase's role in viral RNA capping, which is essential for replication, prodigiosin's inhibition of this enzyme could disrupt viral proliferation and underscore its potential as a broad-spectrum antiviral agent. The reliability of the docking results was ensured by CASTp 3.0, a powerful tool for identifying, defining, and quantifying the topological and geometric characteristics of protein structures. The analysis confirmed that prodigiosin would accurately bind within the NS5 MTase active site. These findings can guide future efforts in structure-based drug design and optimization.
We performed MD simulations to gain deeper insights into the stability, dynamics, and interactions of prodigiosin with NS5 MTases of DENV and ZIKV, reinforcing the molecular docking results. Prodigiosin demonstrated significant stability with both DENV and ZIKV NS5 MTases, as indicated by low RMSD values, which reflect consistent interactions with the target proteins over the simulation period. This stability is vital for preserving the structural integrity of the protein-ligand complexes, potentially hindering protein function and thereby disrupting viral replication. RMSF analysis corroborated these results, showing minimal local fluctuations in both viral proteins, particularly at key binding residues. Additionally, the rGyr and SASA values supported the compact and stable nature of the prodigiosin-bound complexes with lower solvent exposure, suggesting that prodigiosin binding stabilizes the protein structure and potentially reduces its susceptibility to degradation. Besides, the reduction in solvent-exposed surface area indicates a more stable protein-ligand complex, which is favorable for drug efficacy. The protein-ligand contact analysis highlighted specific interactions including hydrogen bonds, non-covalent bonds, and water bridges between prodigiosin and multiple amino acid residues of NS5 MTases, further validating the stability and specificity of prodigiosin binding and supporting its inhibitory potential.
Throughout the study, prodigiosin demonstrated comparable or superior binding affinity, stability, and interactions with both DENV and ZIKV NS5 MTases compared to established inhibitors. These results strengthen the potential advantages of prodigiosin over known inhibitors, particularly in terms of binding stability and targeted interactions. Overall, the MD simulation results combined with the molecular docking studies suggest that prodigiosin is a promising candidate for further development as a broad-spectrum antiviral agent against the flaviviruses. However, more in vitro and in vivo tests are necessary to confirm the utility of the prodigiosin against the target proteins in drug design.
Based on the comprehensive analysis conducted in this study, prodigiosin has demonstrated promising potential as a broad-spectrum inhibitor against important DENV and ZIKV proteins. Through computational approaches including molecular docking and molecular dynamics simulations, prodigiosin exhibited strong binding affinities with NS5 methyltransferases of both viruses, essential for viral replication. This interaction was supported by stable protein-ligand complexes observed in MD simulations, suggesting effective inhibition mechanisms. Furthermore, prodigiosin met drug-likeness criteria and showed favorable ADMET properties, indicating its potential as a safe and effective therapeutic candidate. These findings underscore the utility of prodigiosin as a promising lead compound for further development as an antiviral agent against DENV and ZIKV infections. However, the study's reliance on in-silico analyses necessitates further experimental validation, including in-vitro and in-vivo studies, to confirm prodigiosin’s efficacy and safety in biological systems. Future research should also investigate prodigiosin's mechanism of action in greater detail and evaluate its potential synergy with existing antiviral drugs. Overall, prodigiosin represents a promising candidate for further development, contributing to the urgent need for effective treatments against flaviviral infections.
Data availability statement
All data are available within the article.
Funding
No financial supports are provided to the author(s) for the research, authorship and publication of this article.
Declaration of competing interest
The author(s) declare that there are no potential conflicts of interest.
Author contributions
TJH and MSHB conceived the idea. TR and MSHB designed and conducted the study. TR and MSHB wrote the first draft. TJH revised the manuscript.
[1] R.S.M. Godoy, L. dos S. Felix, A. da S. Orfanó, B.A. Chaves, P.M. Nogueira, B. dos A. Costa, A.S. Soares, C.C.A. Oliveira, R. Nacif-Pimenta, B.M. Silva, A.P. Duarte, M.V.G. de Lacerda, W.M. Monteiro, N.F.C. Secundino, P.F.P. Pimenta, Dengue and Zika virus infection patterns vary among Aedes aegypti field populations from Belo Horizonte, a Brazilian endemic city, PLoS Negl. Trop. Dis. 15 (2021) e0009839. https://doi.org/10.1371/journal.pntd.0009839.
[2] Y. Li, S. Wu, Dengue: what it is and why there is more, Sci. Bull. 60 (2015) 661–664. https://doi.org/10.1007/s11434-015-0756-5.
[3] F. Krauer, M. Riesen, L. Reveiz, O.T. Oladapo, R. Martínez-Vega, T.V. Porgo, A. Haefliger, N.J. Broutet, N. Low, W.Z.C.W. Group, Zika Virus Infection as a Cause of Congenital Brain Abnormalities and Guillain–Barré Syndrome: Systematic Review, PLOS Med. 14 (2017) e1002203. https://doi.org/10.1371/journal.pmed.1002203.
[4] C.N. Farfan-Morales, C.D. Cordero-Rivera, J.F. Osuna-Ramos, I.E. Monroy-Muñoz, L.A. De Jesús-González, J.E. Muñoz-Medina, A.M. Hurtado-Monzón, J.M. Reyes-Ruiz, R.M. Del Ángel, The antiviral effect of metformin on zika and dengue virus infection, Sci. Rep. 11 (2021) 8743.
[5] R. Sarker, A.S.M. Roknuzzaman, Md.A. Haque, Md.R. Islam, E.R. Kabir, Upsurge of dengue outbreaks in several WHO regions: Public awareness, vector control activities, and international collaborations are key to prevent spread, Health Sci. Rep. 7 (2024) e2034. https://doi.org/10.1002/hsr2.2034.
[6] V.J. Klema, M. Ye, A. Hindupur, T. Teramoto, K. Gottipati, R. Padmanabhan, K.H. Choi, Dengue Virus Nonstructural Protein 5 (NS5) Assembles into a Dimer with a Unique Methyltransferase and Polymerase Interface, PLOS Pathog. 12 (2016) e1005451. https://doi.org/10.1371/journal.ppat.1005451.
[7] W. Song, H. Zhang, Y. Zhang, Y. Chen, Y. Lin, Y. Han, J. Jiang, Identification and Characterization of Zika Virus NS5 Methyltransferase Inhibitors, Front. Cell. Infect. Microbiol. 11 (2021) 665379. https://doi.org/10.3389/fcimb.2021.665379.
[8] A. Wahaab, B.E. Mustafa, M. Hameed, N.J. Stevenson, M.N. Anwar, K. Liu, J. Wei, Y. Qiu, Z. Ma, Potential role of flavivirus NS2B-NS3 proteases in viral pathogenesis and anti-flavivirus drug discovery employing animal cells and models: a review, Viruses 14 (2021) 44.
[9] K.E. Du Pont, M. McCullagh, B.J. Geiss, Conserved motifs in the flavivirus NS3 RNA helicase enzyme, Wiley Interdiscip. Rev. RNA 13 (2022) e1688.
[10] S. Xu, Y. Ci, L. Wang, Y. Yang, L. Zhang, C. Xu, C. Qin, L. Shi, Zika virus NS3 is a canonical RNA helicase stimulated by NS5 RNA polymerase, Nucleic Acids Res. 47 (2019) 8693–8707.
[11] N.R. Williamson, H.T. Simonsen, R.A.A. Ahmed, G. Goldet, H. Slater, L. Woodley, F.J. Leeper, G.P.C. Salmond, Biosynthesis of the red antibiotic, prodigiosin, in Serratia: identification of a novel 2-methyl-3-n-amyl-pyrrole (MAP) assembly pathway, definition of the terminal condensing enzyme, and implications for undecylprodigiosin biosynthesis in Streptomyces, Mol. Microbiol. 56 (2005) 971–989. https://doi.org/10.1111/j.1365-2958.2005.04602.x.
[12] M.N.A. Chowdhury, T.J. Hossain, S.A. Zedny, M.S.H. Bappi, M. Rahman, M.M. Hossen, I.H. Nafiz, R. Islam, Multifaceted Plant Growth Promoting Traits and Abiotic Stress Resistance Abilities Exhibited by Chrysanthemum Rhizobacteria, Curr. Appl. Sci. Technol. (2023) e0259128–e0259128. https://doi.org/10.55003/cast.2024.259128.
[13] M.M. Anwar, C. Albanese, N.M. Hamdy, A.S. Sultan, Rise of the natural red pigment ‘prodigiosin’ as an immunomodulator in cancer, Cancer Cell Int. 22 (2022) 419. https://doi.org/10.1186/s12935-022-02815-4.
[14] A. Daina, O. Michielin, V. Zoete, SwissADME: a free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules, Sci. Rep. 7 (2017) 42717. https://doi.org/10.1038/srep42717.
[15] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings, Adv. Drug Deliv. Rev. 23 (1997) 3–25.
[16] A.K. Ghose, V.N. Viswanadhan, J.J. Wendoloski, A Knowledge-Based Approach in Designing Combinatorial or Medicinal Chemistry Libraries for Drug Discovery. 1. A Qualitative and Quantitative Characterization of Known Drug Databases, J. Comb. Chem. 1 (1999) 55–68. https://doi.org/10.1021/cc9800071.
[17] D.F. Veber, S.R. Johnson, H.-Y. Cheng, B.R. Smith, K.W. Ward, K.D. Kopple, Molecular Properties That Influence the Oral Bioavailability of Drug Candidates, J. Med. Chem. 45 (2002) 2615–2623. https://doi.org/10.1021/jm020017n.
[18] N. Guex, M.C. Peitsch, SWISS‐MODEL and the Swiss‐Pdb Viewer: An environment for comparative protein modeling, Electrophoresis 18 (1997) 2714–2723. https://doi.org/10.1002/elps.1150181505.
[19] D.E.V. Pires, T.L. Blundell, D.B. Ascher, pkCSM: Predicting Small-Molecule Pharmacokinetic and Toxicity Properties Using Graph-Based Signatures, J. Med. Chem. 58 (2015) 4066–4072. https://doi.org/10.1021/acs.jmedchem.5b00104.
[20] W.L. DeLano, Pymol: An open-source molecular graphics tool, CCP4 Newsl Protein Crystallogr 40 (2002) 82–92.
[21] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility, J. Comput. Chem. 30 (2009) 2785–2791. https://doi.org/10.1002/jcc.21256.
[22] D.S. BIOVIA, Discovery Studio Visualizer, v20. 1.0. 19295, (2019).
[23] O. Trott, A.J. Olson, AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading, J. Comput. Chem. 31 (2010) 455–461. https://doi.org/10.1002/jcc.21334.
[24] AutoDock Vina 1.2.0: New Docking Methods, Expanded Force Field, and Python Bindings | Journal of Chemical Information and Modeling, (n.d.). https://pubs.acs.org/doi/10.1021/acs.jcim.1c00203 (accessed June 6, 2024).
[25] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function, J. Comput. Chem. 19 (1998) 1639–1662. https://doi.org/10.1002/(SICI)1096-987X(19981115)19:14<1639::AID-JCC10>3.0.CO;2-B.
[26] J. Dundas, Z. Ouyang, J. Tseng, A. Binkowski, Y. Turpaz, J. Liang, CASTp: computed atlas of surface topography of proteins with structural and topographical mapping of functionally annotated residues, Nucleic Acids Res. 34 (2006) W116–W118.
[27] S. Release, 3: Desmond molecular dynamics system, DE Shaw research, New York, NY, 2017, Maest.-Desmond Interoperability Tools Schrödinger N. Y. NY (2017).
[28] B. Goyal, D. Goyal, Targeting the Dimerization of the Main Protease of Coronaviruses: A Potential Broad-Spectrum Therapeutic Strategy, ACS Comb. Sci. 22 (2020) 297–305. https://doi.org/10.1021/acscombsci.0c00058.
[29] K. Roos, C. Wu, W. Damm, M. Reboul, J.M. Stevenson, C. Lu, M.K. Dahlgren, S. Mondal, W. Chen, L. Wang, R. Abel, R.A. Friesner, E.D. Harder, OPLS3e: Extending Force Field Coverage for Drug-Like Small Molecules, J. Chem. Theory Comput. 15 (2019) 1863–1874. https://doi.org/10.1021/acs.jctc.8b01026.
[30] G. Kalibaeva, M. Ferrario, G. Ciccotti, Constant pressure-constant temperature molecular dynamics: a correct constrained NPT ensemble using the molecular virial, Mol. Phys. 101 (2003) 765–778. https://doi.org/10.1080/0026897021000044025.
[31] G.J. Martyna, Remarks on ‘“Constant-temperature molecular dynamics with momentum conservation,”’ Phys. Rev. E 50 (1994) 3234–3236. https://doi.org/10.1103/PhysRevE.50.3234.
[32] T.A. Bouback, S. Pokhrel, A. Albeshri, A.M. Aljohani, A. Samad, R. Alam, M.S. Hossen, K. Al-Ghamdi, M.E.K. Talukder, F. Ahammad, Pharmacophore-based virtual screening, quantum mechanics calculations, and molecular dynamics simulation approaches identified potential natural antiviral drug candidates against MERS-CoV S1-NTD, Molecules 26 (2021) 4961. https://doi.org/10.3390/molecules26164961.
[33] D. Benarroch, M.-P. Egloff, L. Mulard, C. Guerreiro, J.-L. Romette, B. Canard, A structural basis for the inhibition of the NS5 dengue virus mRNA 2′-O-methyltransferase domain by ribavirin 5′-triphosphate, J. Biol. Chem. 279 (2004) 35638–35643. https://doi.org/10.1074/jbc.m400460200.
[34] C. Li, X. Zhu, X. Ji, N. Quanquin, Y.-Q. Deng, M. Tian, R. Aliyari, X. Zuo, L. Yuan, S.K. Afridi, X.-F. Li, J.U. Jung, K. Nielsen-Saines, F.X.-F. Qin, C.-F. Qin, Z. Xu, G. Cheng, Chloroquine, a FDA-approved Drug, Prevents Zika Virus Infection and its Associated Congenital Microcephaly in Mice, eBioMedicine 24 (2017) 189–194. https://doi.org/10.1016/j.ebiom.2017.09.034.
[35] W. Song, H. Zhang, Y. Zhang, Y. Chen, Y. Lin, Y. Han, J. Jiang, Identification and characterization of Zika virus NS5 methyltransferase inhibitors, Front. Cell. Infect. Microbiol. 11 (2021) 665379. https://doi.org/10.3389/fcimb.2021.665379.
[36] R. Pawara, I. Ahmad, S. Surana, H. Patel, Computational identification of 2,4-disubstituted amino-pyrimidines as L858R/T790M-EGFR double mutant inhibitors using pharmacophore mapping, molecular docking, binding free energy calculation, DFT study and molecular dynamic simulation, Silico Pharmacol. 9 (2021) 54. https://doi.org/10.1007/s40203-021-00113-x.
[37] S. Ghosh, S. Das, I. Ahmad, H. Patel, In silico validation of anti-viral drugs obtained from marine sources as a potential target against SARS-CoV-2 Mpro, J. Indian Chem. Soc. 98 (2021) 100272. https://doi.org/10.1016%2Fj.jics.2021.100272.
[38] N.M. Nguyen, C.N.B. Tran, L.K. Phung, K.T.H. Duong, H. le A. Huynh, J. Farrar, Q.T.H. Nguyen, H.T. Tran, C.V.V. Nguyen, L. Merson, A randomized, double-blind placebo controlled trial of balapiravir, a polymerase inhibitor, in adult dengue patients, J. Infect. Dis. 207 (2013) 1442–1450.
[39] Y.-Q. Deng, N.-N. Zhang, C.-F. Li, M. Tian, J.-N. Hao, X.-P. Xie, P.-Y. Shi, C.-F. Qin, Adenosine Analog NITD008 Is a Potent Inhibitor of Zika Virus, Open Forum Infect. Dis. 3 (2016) ofw175. https://doi.org/10.1093/ofid/ofw175.
[40] J. Lin, D.C. Sahakian, S.M. De Morais, J.J. Xu, R.J. Polzer, S.M. Winter, The role of absorption, distribution, metabolism, excretion and toxicity in drug discovery, Curr. Top. Med. Chem. 3 (2003) 1125–1154.
[41] S. Hua, Physiological and pharmaceutical considerations for rectal drug formulations, Front. Pharmacol. 10 (2019) 489933.
[42] Z.B. Zakia Bibi, Role of cytochrome P450 in drug interactions., (2008). https://www.cabidigitallibrary.org/doi/full/10.5555/20093011768 (accessed June 18, 2024).
