Discovery of genistein derivatives as potential SARS-CoV-2 main protease inhibitors by virtual screening, molecular dynamics simulations and ADMET analysis

Jiawei Liu, Ling Zhang, Jian Gao, Baochen Zhang, Xiaoli Liu, Ninghui Yang, Xiaotong Liu, Xifu Liu, Cheng Yu, Yu ChengFrontiers in Pharmacology 2022 · DOI

TL;DR

This study leverages computational methods to discover novel genistein derivatives as potential inhibitors of the SARS-CoV-2 main protease (MPro), a key enzyme for viral replication. By combining virtual screening, molecular dynamics simulations, and ADMET predictions, the researchers identified compound 11 as a promising candidate with strong binding affinity, favorable drug-like properties, and good predicted safety profiles.

What Problem Does This Paper Address?

The ongoing threat of COVID-19, driven by the continuous mutation of SARS-CoV-2, necessitates the development of effective therapeutic agents. While vaccines have been crucial, the emergence of new variants and the limited efficacy of existing treatments highlight the urgent need for novel antiviral drugs. The SARS-CoV-2 main protease (MPro) is a critical enzyme for viral polyprotein processing and is considered a prime target for drug development. However, identifying potent and safe inhibitors from vast chemical libraries is a significant challenge. This paper addresses this by employing a computational drug discovery pipeline to screen a large library of genistein derivatives, aiming to find compounds that can effectively inhibit MPro.

Methods

The researchers employed a multi-stage computational approach:

  • Virtual Screening: A library of 9,614 genistein derivatives was screened against the SARS-CoV-2 MPro using two molecular docking software: LeDock and AutoDock Vina. This standard approach helps to quickly identify potential binders from a large chemical space. The use of two different software aims to reduce false positives by leveraging different algorithms.
  • Normalization and Ranking: To reconcile the results from the two docking programs and improve the reliability of the ranking, the docking scores were normalized. This process allowed for a more robust selection of the top candidates.
  • Molecular Dynamics (MD) Simulations: The top 20 compounds identified from the normalized docking were subjected to 200 ns MD simulations. This technique is crucial for assessing the stability of the protein-ligand complex over time and provides insights into the dynamic interactions, which are more informative than static docking poses.
  • Binding Free Energy Calculation: The MM/PBSA method was used to calculate the binding free energies between the protein and the top compounds, providing a quantitative measure of their binding affinity.
  • Drug-likeness and ADMET Prediction: Standard computational tools (DruLiTo, Molinspiration, and pkCSM) were used to predict key drug-like properties (e.g., Lipinski's rule compliance), biological activity against various targets, and Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) profiles. These analyses are essential for assessing the potential of a compound as a viable drug candidate.

Datasets

  • ZINC20 Library: A subset of approximately 9,614 genistein derivatives was sourced from the ZINC20 database, a widely used resource for drug discovery.
  • SARS-CoV-2 MPro Crystal Structure (PDB ID: 6LU7): This structure, complexed with a peptide-like inhibitor N3, was used as the target protein for docking and simulation studies.

Key Insights and Ideas

  • Synergistic Docking: Combining two distinct molecular docking programs (LeDock and AutoDock Vina) and normalizing their results proved effective in identifying promising candidates and reducing the likelihood of false positives. This highlights the importance of using multiple computational tools for robust screening.
  • Genistein as a Promising Scaffold: The study reinforces genistein's potential as a lead compound for antiviral drug discovery, demonstrating that its derivatives can be modified to achieve potent inhibition of SARS-CoV-2 MPro.
  • Chirality Matters: The significant difference in binding affinity between closely related isomers (compounds 2 and 3) underscores the critical role of molecular stereochemistry in drug-target interactions.
  • Hydrophilicity Trade-off: Compounds with more sugar moieties (e.g., 15, 16, 17) showed lower normalized affinities, suggesting that excessive hydrophilicity can sometimes hinder binding to hydrophobic protein sites. This points to a delicate balance needed in designing inhibitors.
  • Compound 11's Multifaceted Potential: Compound 11 emerged as a standout candidate, exhibiting strong binding, favorable drug-like properties, predicted protease inhibitory activity, and excellent ADMET characteristics, making it a strong contender for further investigation.

Key Results

  • Virtual Screening Success: Out of 9,614 genistein derivatives, 20 were selected based on normalized molecular docking scores.
  • Stable Interactions: Molecular dynamics simulations indicated that compounds 2, 4, 5, 11, 13, 14, 17, and 18 exhibited the most stable binding interactions with SARS-CoV-2 MPro, with binding free energies exceeding -50 kJ/mol.
  • Promising Drug-like Properties: Several compounds, including compound 11, demonstrated acceptable drug-like properties according to Lipinski's rule and other physicochemical assessments.
  • Predicted Bioactivity: Compounds 11 and 12 showed high predicted inhibitory activity against MPro, suggesting their potential as COVID-19 inhibitors.
  • Favorable ADMET Profile: Compound 11 exhibited excellent predicted ADMET properties, including high absorption and low toxicity, making it a particularly attractive candidate.

Limitations and Future Work

  • In Silico Nature: The study relies entirely on computational methods. Experimental validation of the predicted binding affinities, inhibitory activities, and ADMET properties is crucial.
  • Lipinski's Rule Deviations: While many compounds exceeded the molecular weight limit of 500 Da in Lipinski's rule, the authors note that this doesn't preclude them from being drug candidates, especially with advancements in pharmaceutical technology. However, further investigation into overcoming potential absorption issues might be necessary.
  • Experimental Validation: The authors explicitly state that compound 11 has "great promise for further research," implying that the next steps involve experimental synthesis and testing of this compound and other promising candidates.
  • Mechanism of Action Nuances: While interactions were analyzed, a deeper understanding of the precise binding mechanisms and potential off-target effects would be beneficial.

Summary updated

Abstract

<b>Background:</b> Due to the constant mutation of virus and the lack of specific therapeutic drugs, the coronavirus disease 2019 (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) still poses a huge threat to the health of people, especially those with underlying diseases. Therefore, drug discovery against the SARS-CoV-2 remains of great significance. <b>Methods:</b> With the main protease of virus as the inhibitor target, 9,614 genistein derivatives were virtually screened by LeDock and AutoDock Vina, and the top 20 compounds with highest normalized scores were obtained. Molecular dynamics simulations were carried out for studying interactions between these 20 compounds and the target protein. The drug-like properties, activity, and ADMET of these compounds were also evaluated by DruLiTo software or online server. <b>Results:</b> Twenty compounds, including compound <b>11</b>, were screened by normalized molecular docking, which could bind to the target through multiple non-bonding interactions. Molecular dynamics simulation results showed that compounds <b>2</b>, <b>4</b>, <b>5</b>, <b>11</b>, <b>13</b>, <b>14</b>, <b>17</b>, and <b>18</b> had the best binding force with the target protein of SARS-CoV-2, and the absolute values of binding free energies all exceeded 50 kJ/mol. The drug-likeness properties indicated that a variety of compounds including compound <b>11</b> were worthy of further study. The results of bioactivity score prediction found that compounds <b>11</b> and <b>12</b> had high inhibitory activities against protease, which indicated that these two compounds had the potential to be further developed as COVID-19 inhibitors. Finally, compound <b>11</b> showed excellent predictive ADMET properties including high absorption and low toxicity. <b>Conclusion:</b> These in silico work results show that the preferred compound <b>11</b> (<b>ZINC000111282222</b>), which exhibited strong binding to SARS-CoV-2 main protease, acceptable drug-like properties, protease inhibitory activity and ADMET properties, has great promise for further research as a potential therapeutic agent against COVID-19.
BibTeX
@inproceedings{liu2022discovery,
  author = {Jiawei Liu and Ling Zhang and Jian Gao and Baochen Zhang and Xiaoli Liu and Ninghui Yang and Xiaotong Liu and Xifu Liu and Cheng Yu and Xifu Liu and Xifu Liu and Yu Cheng},
  title = {{Discovery of genistein derivatives as potential SARS-CoV-2 main protease inhibitors by virtual screening, molecular dynamics simulations and ADMET analysis}},
  booktitle = {Frontiers in Pharmacology},
  year = {2022},
  doi = {10.3389/fphar.2022.961154},
}
Figure