The emergence of COVID-19 posed a significant global health challenge, necessitating rapid development of effective treatments. While vaccine development has been a priority, the need for effective therapeutic interventions remains crucial. The high mutation rate of SARS-CoV-2 necessitates strategies beyond targeting viral proteins, as drug resistance is a major concern. This study builds upon previous work demonstrating the potential of Sulindac, a non-steroidal anti-inflammatory drug (NSAID), in suppressing NF-kB, a key human transcription factor implicated in COVID-19 pathogenesis. This current research expands on this by examining 51 drugs with known efficacy against various diseases, evaluated using in silico methods for their potential to interact with key proteins involved in SARS-CoV-2 infection. A key consideration is the manipulation of protein-protein interactions between viral and host proteins, which are essential for viral replication and lifecycle completion. This research focuses on identifying potential drug targets within the human proteome, specifically targeting human transcription factors, including NF-kB, and its associated interactome. The research employed molecular dynamics simulation and other bioinformatics tools to assess potential drug-receptor interactions, generating data that could support the repurposing of the studied molecules for COVID-19 treatment after thorough clinical trials. This approach offers a potentially faster route to treatment compared to the development of entirely new drugs.

The study references numerous articles describing various approaches to drug discovery for COVID-19, including the use of affinity purification (AP) and proximity labeling-based strategies for receptor identification. Several studies highlighting the importance of specific SARS-CoV-2 protein domains and motifs in viral-host interactions are also cited. The authors highlight the work of Liu et al. [18] which identified methotrexate as a potentially useful drug. The challenges of viral mutation rates and the consequent development of drug resistance are also discussed, emphasizing the advantages of host-targeting drugs. The paper supports the need for a comprehensive virus-host proteome interaction atlas to identify cellular functions critical for viral processing, aiding the development of effective therapeutic strategies.

The study employed a combination of in silico techniques, including molecular docking and molecular dynamics simulations. Protein structures for five receptors (1R42, 4G3D, 6VW1, 6VXX, and 7MEQ) were retrieved from the RCSB PDB database. Fifty-one ligands from various classes (IKK complex inhibitors, IkB degradation inhibitors, etc.) were obtained from NCBI PubChem and ZINC databases. Ligands without available 3D structures were generated using CORINA. Virtual screening was performed to identify the top five ligands with the best docking scores for each receptor. Molecular docking studies utilized Autodock version 4.2.6, PyRx, and Open Babel tools for ligand preparation and docking. The grid box was adjusted to cover the entire protein molecule to allow comprehensive ligand binding site search. The top five docked complexes, based on RMSD value and binding energy, were selected for visualization using Maestro workspace and Discovery Studio. Molecular dynamics simulations (MDS) were performed using Desmond for 100 ns for each of the top five complexes. Protein preprocessing involved structure error checking, minimization, and water molecule removal. The simulation environment was set using the OPLS3e force field, and the system was neutralized by adding Na+ or Cl- ions. RMSD, RMSF, and protein-ligand contacts were analyzed to assess the stability of the complexes. Analysis of ligand properties (RMSD, radius of gyration, etc.) and secondary structure were performed. Protein-ligand contacts were analyzed using the PLIP tool and considered only if the interactions lasted more than 10% of the simulation time.

Molecular docking revealed that boswellic acid exhibited the strongest binding affinity to receptor 1R42 (-9.2 kcal/mol), while Pimecrolimus showed strong binding to receptor 4G3D (-9.1 kcal/mol), and GYY_4137 bound strongly to receptor 6VW1 (-10.3 kcal/mol). Detailed docking scores for the top five ligands against each receptor are provided in Table 3. Analysis of amino acid interactions revealed that key residues of the selected receptors played significant roles in the binding mechanism (Table 2). 3D visualizations of the top five interaction complexes for each receptor are presented (Figures 1-5). Molecular dynamics simulations (MDS) for 100 ns showed varying degrees of stability for the top five complexes. RMSD analysis indicated that all complexes reached equilibrium by the end of the simulation, although some showed more conformational changes than others (Figure 6). The RMSD values varied greatly, with pimecrolimus-4G3D having the highest RMSD, suggesting significant conformational changes. Protein RMSF analysis showed fluctuations in different protein regions (Figure 9). Protein-ligand contact analysis showed varying types and frequencies of interactions (Figure 10). Timeline charts of protein-ligand contacts revealed multiple contacts for certain residues in some complexes (Figure 11). Analysis of ligand properties revealed varying intramolecular hydrogen bonds, RMSD values, radius of gyration, molecular surface area, solvent-accessible surface area, and polar surface area (Figure 12). Secondary structure analysis showed varying proportions of alpha-helices and beta-strands for the different complexes (Figure 13). The study highlighted the stable binding of boswellic acid to human ACE2 and triamcinolone hexacetonide to TMPRSS2, suggesting their potential roles in intervening the ACE2 and TMPRSS2 pathways. The results also showed potential for NF-kB pathway inhibition through Pimecrolimus, potentially mitigating cytokine storms.

The findings suggest that targeting the NF-kB pathway, a key player in COVID-19 pathogenesis, might be a viable therapeutic approach. The results highlight several promising drug candidates that showed strong binding affinities and stable interactions with their respective protein targets during the molecular dynamics simulations. The observed stability of boswellic acid binding to ACE2 and triamcinolone hexacetonide to TMPRSS2 at a molecular level supports the potential for these drugs to intervene in the ACE2 and TMPRSS2 pathways. The results regarding NF-kB inhibition by pimecrolimus also hold significance, potentially alleviating the severity of cytokine storms often observed in COVID-19 patients. The research adds to the growing body of evidence supporting drug repurposing as a rapid and efficient strategy for addressing emerging infectious diseases. The combination of molecular docking and simulation analysis provide valuable insights into potential drug-receptor interactions. However, further in-vitro and in-vivo studies are necessary to confirm the efficacy and safety of these drug candidates.

This study provides compelling evidence for the potential repurposing of several existing drugs for COVID-19 treatment. The results highlight the potential of targeting host proteins, particularly NF-kB, as a means of mitigating viral effects. Further research, including preclinical and clinical trials, is warranted to validate these findings and assess the efficacy and safety of these drug candidates in humans. Future studies could focus on optimizing the identified drug candidates and exploring their synergistic effects with other antiviral therapies.

The study is limited by its reliance on in silico methods. While molecular docking and simulation provide valuable insights into potential drug-receptor interactions, these findings need to be validated through experimental studies, including in-vitro and in-vivo experiments. The 100ns simulation time, while a standard duration for these analyses, might not capture all possible long-term dynamics of the interactions. The study's focus on a limited number of receptors and ligands represents another limitation and could be expanded in future studies.