Introduction

The acute respiratory infection that we know as the novel coronavirus SARS-CoV-2 was first identified in Wuhan China in the year 2019 (Petersen et al. 2020). This infection is highly contagious and can be transmitted via droplet and contact.Coronavirus (CoV) belongs to the family Coronaviridae, suborder Cornidovirineae, and order Nidovirales. According to the phylogenetic analysis, the Coronaviridae family can be classified into 4 genera, namely alpha, beta, delta, and gamma (Payne 2017). Human and animal cells are both susceptible to infection by the single stranded RNA virus known as the coronavirus, which has a positive virulence. Human coronaviruses are among the rapidly evolving viruses, because of their high recombination rate and nucleotide element dominance (Fehr and Perlman 2015). Several recent reports, both genome wide and at the receptor level provided unique insights on many characteristics of the SARS-CoV-2 virus. One of these characteristics include the ability of different variants of SARS-CoV-2 to bind to Angiotensin-Converting Enzyme 2 (ACE2) cell receptors (Letko et al 2020). It is the binding of the spike protein (S) to the ACE2 receptor that allows SARS-CoV-2 to enter cells, as it facilitates the initiation of viral replication (Shang, J et al, 2020; Wang et al 2020).

SARS-CoV-2 is a virus with a long, single-stranded RNA genome, about 30,000 bases in length. This genome is one of the largest among all known RNA viruses, which boosts the virulence of the virus through reduced dependence on host cells for replication. The virus itself is small, measuring between 50-200 nanometers, and is covered with spike-like proteins that help invade human cells. SARSCoV-2 is responsible for causing Severe Acute Respiratory Syndrome (SARS), which affects the respiratory system and can lead to serious complications (Payne, 2017). The symptoms of infection can vary widely among individuals. While some patients experience mild, cold-like symptoms, others suffer from severe respiratory distress or long-term complications, often referred to as “long COVID.” This variability in how the virus impacts individuals highlights the complexity of SARS-CoV-2 and its interactions with the human body.

Coronaviruses display spherical structure and include the following four main types of structural proteins; membrane (M), envelope (E), nucleocapsid (N) and spike (S) protein (Fehr and Perlman, 2015). Membrane protein lies within the range of 25 to 30 kDa and has three transmembrane N-terminal domains, a glycosylated ectodomain and a C terminal domain (Armstrong et al., 1984). The M protein is diamiamic protein that fuses other outer structural proteins, foreboding the virus to bud out (Neuman et al., 2011). Envelope protein has an approximate mass of about 8 to 12 kDa (Schoeman and Fielding, 2019). The E protein consists of the N terminal ectodomain and C terminal endodomain (Nietto-Torres et al., 2014). The E protein increases the pathogenicity of the virus by acting on its ion channel as transmembrane proteins during the process of budding and the release of the virus (Li et al., 2014). The N protein has N terminal and C terminal domains.

The N protein is rich in phosphate esters and phosphates in general (Stohlman and Lai, 1979). Because of the strong binding of N proteins to viral RNA, the N protein associates with the viral RNA genome and encapsulates it (Kuo and Masters 2013). The N protein assists in the packaging of the viral RNA genome into the replicase-transcriptase complex (RTC) and afterwards the virus genome into the viral virus (Hurst et al. 2013).

Spike protein has a vital role in SARS-CoV-2 virus transmission (Shang et al., 2020). Attachment initiates with the trimetric spike protein’s S1 subunit’s interaction with ACE2 receptors in the heart, kidneys, and lungs (Hamming et al., 2004). S1 and S2 separation in the SARS-CoV-2 spike protein takes place owing to the presence of the proprotein convertase (PPC) section (Walls et al., 2020). The S protein is subsequently cleaved at the host cell membrane by the transmembrane serine protease 2 (TMPRSS2) which results in the release of S1 subunit while triggering the S2 subunit to become rearranged to a post-fusion conformation that assists in the merging of the viral and host cell membranes (Bestle et al., 2020). The peptide domain that is able to modulate this action comes from the S2 subunit and is activated via receptor recognition, allowing it to breach the host cell membrane (Xiu et al., 2020). Delta and Omicron variants have highlighted the importance of targeting conserved viral elements, as these variants exhibit increased transmissibility and resistance to neutralizing antibodies (Shuai et al., 2022).

The primary objective of molecular docking is to characterize and predict molecular recognition, through structural (identifying alternate binding modes) and energy-wise compatibility. Computer-based methods shorten the drug design process, providing low cost and speed drug development environments, while, contributing to the analysis of the interaction between ligands and the target protein structure by calculating the binding affinity (Kitchen et al., 2004). In silico studies, molecular docking in particular, have accelerated the identification of potential inhibitors targeting SARS-CoV-2 proteins. Examples include the identification of drugs that could bind to the spike protein receptor-binding domain through docking studies (Garg et al., 2021; Santos-Martins et al., 2021). Computational techniques have been employed for fast, affordable, and efficient compound screening, paving the way for drug discovery.

In this study, molecular docking simulations are used to identify novel molecules capable of inhibiting the spike protein found in SARS-CoV-2. By evaluating binding affinities and physicochemical properties of ligand candidates, this work aims to contribute to the ongoing efforts of developing therapeutic strategies against SARS-CoV-2 and its evolving variants. FDA-approved molecules identified as inhibitors of the relevant target structure through the ZINC15 database were included in the study as ligand molecules.

Materials and methods

To obtain the crystal structures of the proteins important for SARS-CoV-2, the ACE2-linked structure with PDB ID 6M0J (Figure 1) at 2.45 Šresolution was selected from the Protein Data Bank (https://www.rcsb.org/) database (Lan et al., 2020). In order to perform multiple docking operations in a concise manner, it was essential to define a specific region of interest on the protein surface where ligand binding is most likely to occur. The gridbox center coordinates for 6M0J were determined as x=-31.454, y=29.553, z=21.871 and a gridbox box with a size of 40x40x40 ų was determined. In this study, 1,615 different FDA-approved molecules were retrieved from the ZINC15 database (https://zinc15.docking.org). These molecules were then docked to the target protein. Through Autodock Vina, the binding affinities of the candidate molecules were scored and compared. The obtained physicochemical parameters of the ligand candidate moleculesfor effective treatment of SARS-CoV-2 disease were taken into consideration. All the docking studies were implemented using Autodock Vina (Eberhardt et al., 2021). After docking studies were performed, the interaction maps of the ligands with the target protein structures were investigated in detail. The topmost 10 ligands were selected ranked by their binding affinities. These ligands were simulated for 2 nanoseconds using NAMD. Subsequently, the overall energy change was observed between the ligands and the target protein.

Figure 1. Crystal structure of SARS-CoV-2 spike receptor-binding domain bound with ACE2.

Results and discussion

Table 1 shows molecular weights and lipophilicities of candidate molecules that are were docked against S spike protein of SARS-CoV-2. These physicochemical parameters, including molecular weight (Mw) and lipophilicity (logP), play a crucial role in determining the pharmacokinetic behavior of the ligands, such as absorption, distribution, and membrane permeability.

As demonstrated in Table 1, it is evident that the binding affinities of the candidate ligands to the S spike protein of SARS-CoV-2 vary significantly. -7.0 kcal/mol is acknowledged as a threshold for high protein-ligand binding affinity. As clearly shown on Table 1, multiple candidate ligands surpass this threshold with a relatively high margin. This underlines the fact that alternative options for highly affinitive ligands exist ubiquitously among FDA approved ligands.

Moreover, the 10 most suitable ligands compared based on Binding Affinities (kcal/mol) are put into molecular dynamics simulation for further analysis. The protein–ligand complex was subjected to a 2 nanosecond molecular dynamics simulation using Nanoscale Molecular Dynamics (NAMD) (Philips et al., 2020). After the simulation was complete, visualization and trajectory analyses were performed using Visual Molecular Dynamics (VMD) (Humphrey, 1996). The system was solvated in a TIP3P water box, neutralized with counterions, and simulated under periodic boundary conditions using the CHARMM36 force field. Insights from these simulations will help identify ligands with both strong and stable interactions, potentially guiding future in vitro validation and drug repurposing efforts against SARS-CoV-2.

Figure 2 shows the complex formed by ZINC000052955754 and spike protein S of SARS-CoV-2. Taking Table 1 into consideration, this ligand shows high binding affinity towards the spike protein of SARS-CoV-2. After this initial docking analysis, the complex was subjected to molecular dynamics simulation to evaluate the stability and behavior of the ligand within the binding pocket over time. Non-bonded interactions between the ligand and the protein were calculated through NAMD to provide a better viewpoint on the inhibition feature of the ligand. Calculated frame-by-frame, the mean of interaction energy was calculated as -62.811 kcal/mol. The negative value of the interaction energy suggests that the binding process is thermodynamically favorable and likely spontaneous under physiological conditions. This energy profile reinforces the notion that ZINC000052955754 forms a stable and energetically favorable complex with the spike protein.

Figure 2. ZINC000052955754 Docked to spike protein S of SARS-CoV-2.

Conclusions

This study identified potential candidate molecules for inhibiting the SARS-CoV-2 spike protein by molecular docking studies. Among the 1,615 compounds screened, the top ten candidates were further investigated through molecular docking and molecular dynamics simulations. The frequent and significant interactions between these ligands and the amino acids suggest their potential as inhibitors against SARS-CoV-2 spike protein. Furthermore, ligand with the ZINC ID of ZINC000052955754 demonstrated a strong and spontaneous interaction with the spike protein, supported by a mean interaction energy of -62.811 kcal/mol. These findings emphasize the potential of repurposing existing drugs as effective inhibitors of SARS-CoV-2. In silico methodologies applied in this study provide a cost-effective and efficient pathway for drug discovery. These results align with the understanding that structural compatibility between the ligand and active site is crucial for achieving high binding affinity. Moreover, the results reinforce the importance of structural complementarity and physicochemical optimization in achieving high-affinity protein–ligand interactions. Future work should focus on experimental validation through in vitro and in vivo assays to confirm the therapeutic potential of these candidates. Expanding the compound library and enhancing computational accuracy will further strengthen the applicability of such approaches in addressing current and emerging viral threats.