Coronaviruses belong to Coronaviridae family which causes severe respiratory diseases in human beings. These viruses were initially isolated from a human in the year 1965 and are positive-sense, single-stranded RNA viruses. Until now, there are three major outbreaks caused due to coronavirus, namely, severe acute respiratory syndrome coronavirus (SARS-CoV) in the year 2002, Middle East respiratory syndrome coronavirus (MERS-CoV) in the year 2012, and recently, SARS-CoV-2 in the year 2019 [1]. The recent pandemic of coronavirus disease 2019 (COVID-19) that emerged in December 2019 has affected the whole nation and has put restrictions on the movement of people. It has been cautioned to every individual globally to remain at home and not to assemble, wash hands regularly, use face mask and hand sanitizer and keep physical distance [2]. SARS-CoV-2 which triggers the coronavirus disease is considered to have originated in the laboratories of Wuhan, China. It is the seventh human coronavirus that can cause severe health issues, majorly infecting the lungs that can eventually cause death. The frequently reported symptoms of COVID include shivering with high body temperature, cough, and difficulty breathing [3].

Currently, there are no relevant medications available for effective treatment of coronavirus. Drug repurposing has shown positive effects but they are incapable of preventing the disease. To treat coronavirus, it is very important to identify the therapeutic target or a drug that efficiently arrests the replication of SARS-CoV-2 [4]. One of the methods can be finding an antiviral drug or a molecule that can interfere with the spike protein or CD147 expression [5].

There are mainly two receptors that could be pathways for SARS-CoV-2 to get inside the host cell, namely angiotensin-converting enzyme 2 (ACE2) and CD147. The spike protein present externally on coronavirus binds with either ACE2 or CD147 receptor, thus intervening capture and spreading of the virus to another cell [5]. Recently, Wang et al. [6] illustrated that the expression level of ACE2 is low and there may be other routes for viral entry which later they found to be CD147. They came across the activity of the spike protein of SARS-CoV-2 and CD147 receptor-mediated by meplazumab which is an anti-CD147 humanized antibody. The use of this antibody blocks the host cell receptor CD147 which in turn inhibits the amplification of SARS-CoV-2. This particular work revealed an important target that can be utilized in developing a therapeutic drug against COVID-19 [7]. Furthermore, the study conducted by Geng and a co-worker revealed CD147 as a universal receptor for SARS-CoV-2 as well as its other variants [8].

CD147 or basigin or extracellular matrix metalloproteinase inducer is an integral membrane glycoprotein that comprises mainly two immunoglobulin units present in the extracellular region. The two domains of immunoglobulin are a single transmembrane domain and a cytoplasmic domain [9]. Apart from facilitating SARS-CoV infection, it also facilitates tumor growth, Plasmodium capture, and infection of viruses and bacteria [7]. CD147 is thought to regulate cytokine production and leucocyte chemotaxis when it binds to cyclophilin A (CyPA) [10]. Cyclophilins belong to the PPlase family of intracellular proteins which is highly conserved. It is the most commonly expressed protein of this family consisting of seven types of which CyPA is the most abundant and dominant protein [4].

This work aims to design and identify inhibitors of human basigin to stop viral entry into the cell by developing an e-Pharmaco­phore model of receptor-ligand interaction between CD147 and various U.S. Food and Drug Administration (FDA)–approved drugs for COVID-19. Pharmacophore modeling is a part of computer-aided drug design (CADD) that plays a key role when there is no structure of receptor presents [11]. Pharmacophores may be described as the arrangement of a molecule that stores the essential characteristics necessary for the biological or pharmacological interaction of drugs. It is used for screening the molecules virtually that triggers a biological response. A pharmacophore model usually has two procedures, ligand-related and structure-related [12]. In ligand-related pharmacophores modeling, a set of active molecules are superposed in order to get the chemical activities that are necessary for biological activities [13]. However, in structure-related method a pharmacophore model is developed using the structural features of the protein. In this model, the interaction between ligands and macromolecular targets is taken into account [14].

The technique of molecular docking has been widely applied for many decades in the field of drug discovery and many other aspects of bioinformatics. The method takes into account the orientation of a molecule and how the molecule changes its conformation after binding to a particular molecular target [15]. It is a structure-based drug design where binding occurs between a receptor and a ligand. There are various software developed to date, among which are some well-known, AutoDock, AutoDock Vina, ZDOCK, RDOCK, Glide, and Gold [16]. The first step in molecular docking involves generating a three-dimensional structure of a ligand and receptor from various databases available like Protein Data Bank (PDB) or can be drawn using tools e.g., ChemSketch, and preparing it. The next step involves defining and calculating of grid where the ligand can be docked. And finally, the docking process can be carried out and the results can be visualized using a visualization software [17].

In this work, with the help of the Biovia Discovery Studio platform, an e-pharmacophore model was developed using the protein Human Basigin (PDB ID 7DCE) in a structure-based manner. The developed model was used for screening the drug targets. A total of eleven drugs (Fig. 1), drawn online using Chem-Space were used for screening. These drugs were selected based on the literature studies [18,19] that have shown activity in COVID-19 therapy. All the drugs act differently and have several mechanisms against coronavirus. Various in vitro studies showed the drugs to be helpful in blocking the viral production of this deadly virus. Therefore, these drugs have been utilized for repurposing studies and to understand their ability to block the CD147 receptor to treat COVID-19. Further, the reasons for selecting these eleven drugs are described below.

The cryo-EM structure of human XKR8-Basigin complex bound to Fab fragment (PDB ID 7DCE) was acquired from Protein Data Bank at a resolution of 3.80 Å. An e-pharmacophore model was developed using this protein (7DCE) to generate pockets where drug targets can fit accordingly. Eleven drugs, namely, azithromycin, hydroxychloroquine, chloroquine, lopinavir, ritonavir, nafamostat, camostat, famotidine, umifenovir, nitazoxanide, and fluvoxamine were considered for this work as these drugs are currently being utilized for the treating the symptoms of coronavirus. These drugs after preparation were screened in the pharmacophore model amongst which seven of them, viz, hydroxychloroquine, lopinavir, ritonavir, camostat, famotidine, umifenovir, and fluvoxamine fitted to the pharmacophore pockets. The rest of the drugs i.e., azithromycin, chloroquine, nafamostat, and nitazoxanide could not screen as these drugs do not interfere with the CD147 receptor. Later, the protein was prepared and energy minimization was performed to determine a proper molecular arrangement. After defining the site sphere molecular docking was performed between the receptor (human basigin) and the screened drug targets using CDOCKER. According to the CDOCKER energy generated, ritonavir emerged as the best drug demonstrating a higher CDOCKER energy of –57.30 and its corresponding CDOCKER interaction energy of –53.38, respectively. In summary, our results contribute to the findings of relevant drugs for COVID-19 and can be of great importance for other scientific works in the future.