Discovery of natural compounds that inhibit SARS-CoV-2 main protease through virtual screening

The fight against the COVID-19 pandemic is an international scientific challenge regarding the search for strategies to fight the viral infection and prevent its damaging consequences. Besides the vaccines and approved drugs by the health authorities, it is necessary to find new treatments to fight against the viral infection once this appears.

A study published in the Journal of Chemical Information and Modeling has identified a series of natural products able to inhibit M^pro, the main protease of SARS-CoV-2. M^pro is a non-structural protein with an essential role in the replication and transcription of this virus, and it is regarded as a potential therapeutic target, for its inhibition could prevent the virus from progressing.

The study is led by Professor Jaime Rubio Martínez, from the Faculty of Chemistry and the Institute of Theoretical and Computational Chemistry (IQTC) of the University of Barcelona, and it can shed light on the design of new therapeutic strategies to fight COVID-19. This new scientific advance is based on the use of in silico virtual screening, a computational analytical technique of molecules placed in databases to identify and select compounds that are candidates of potential pharmacological interest.

Virtual screening: searching for future drugs

Now, the clinical approach of COVID-19 is based on the symptomatic treatments through anti-inflammatory agents (for instance, dexamethasone or cytokine inhibitors) combined with antibiotics to treat secondary infections. Recently, the antiviral ritonavir has been approved, but there is still a need for new medicines. Within the context of global health, it is urgent to develop SARS-CoV-2-specific antiviral therapeutics and this led to the application of several strategies to identify bioactive compounds that can be used as therapeutic agents (including the available drugs and natural products). In this context, the virtual screening is a reliable, fast and efficient procedure for finding bioactive compounds of large chemical collections against a specific molecular target.

As part of the study, the team conducted a virtual screening of the natural product database Selleck —a chemical library with nearly 2,000 compounds— for a series of conformations of the SARS-CoV-2 M^pro characterized with a molecular dynamics study.

As a result, the study presents the characterization of the dynamic profile of M^pro protease in its apo shape through conventional simulations (cMD) and Gaussian accelerated molecular dynamics (GaMD) simulations, and it defines a series of representative structures. 

These structures were later used to carry out the ensemble docking, a bioinformatic method that enables the prediction and calculation using computational techniques of the most favourable position of interaction between a ligand and its target protein. Then, they carried out an iterative procedure which increased the length of cMD of protein-ligand complexes and calculated the free binding energy to select the most promising candidates.

Having obtained the results, they selected eleven compounds and these were tested in vitro for their capability to inhibit M^pro protease. Among these, five SARS-CoV-2 antiviral candidates were identified as inhibitors of Mp^ro protease, using the natural products database.