Article Preview
Top1. Introduction
The COVID-19 pandemic that originated in Wuhan in 2019 is still causing thousands of deaths worldwide and there is not a consensual medicine or therapy that may effectively treat severe cases. At this time, trustworthy data available are scarce despite the strong effort that the scientific community is making. The etiological agent responsible for this disease is a new coronavirus called SARS-CoV-2 that shared 79.5% genetic homology and has a similar receptor-binding domain structure to that of the already known SARS-CoV (Lu et al., 2020). Due to the lack of specifically designed antiviral drugs for SARS-CoV-2, a research strategy is to study antiviral agents currently known to be effective against other RNA viruses such as SARS-CoV (Liu et al., 2020). There is also the possibility of discovering therapeutic agents targeting the highly conserved proteins associated with SARS-CoV and SARS-CoV-2. It is known that a chymotrypsin-like protease (3CLpro) named as the Main protease (Mpro) is crucial for the SARS-CoV replication process, and it has been proposed as a target for the design of anti-SARS drugs (Goetz et al., 2007). The Mpro of SARS-CoV-2 shares over 95% of sequence similarity with that of SARS-CoV (Jin et al., 2020; Morse, Lalonde, Xu, & Liu, 2020). Several recent studies have been focused in the repurposing of existing drugs with known inhibitory activity towards this Mpro of SARS-CoV (Elmezayen, Al-Obaidi, Şahin, & Yelekçi, 2020; Huynh, Wang, & Luan, 2020; Mahanta et al., 2020). Kumar and Roy (2020), for example have studied 69 compounds reported as inhibitors of the 3CLpro of SARS-CoV in the database of ‘The Scripps Research Institute Molecular Screening Center’. They developed a 2D-QSAR model to predict the main molecular characteristics that affect the inhibitory activity of the target compounds and performed in silico predictions of SARS-CoV 3CLpro inhibitory activity of more than 50,000 compounds obtained from different anti-viral drug databases to establish a priority list in the search of existing drugs capable of treat the SARS-CoV-2 virus. The study concludes that presence of sulfur (especially as π-sulfur) in the molecular structure of the candidate compounds is the main molecular characteristic that enhances the inhibitory activity.
In silico methodologies are playing today a crucial role in the discovery of anti -SARS-CoV-2 drugs and they are implemented not only in the search of existing compounds with known activity towards similar targets, but also in the exploration of novel drugs. Searching for new compounds with high antiviral potential, Wang et al. (2017) proposed unsymmetrical aromatic disulfides as non-peptidic inhibitors of SARS CoV Mpro. Their results show that these compounds are reversible and non-competitive inhibitors and constitute a promising new family of biologically active anti-SARS agents. The authors found a compound of this family that yields a remarkable IC50 value of 0.516 mM for the main protease of SARS-CoV.
Recently, Toropov and coworkers (2020) published an in silico study on the 40 aromatic disulfide compounds reported by Wang et al. (2017) to model their inhibitory effect and calculate their binding energies with the purpose of correlate this energy with the inhibitory potential of designed molecules. They proposed a one descriptor model with 
and 
.
Based on the great potential of these compounds, our goal in this work is propose a new QSAR model that accurately predicts the anti-SARS inhibitory activity of aromatic disulfides without the requirement of the knowledge of the conformational characteristics of the molecules. Therefore, only conformation-independent molecular descriptors were considered. To evaluate the predictive applicability of our model, we propose novel compounds based on the findings of this work that may be promising candidates to test as potent SARS CoV Mpro inhibitors. The programs used to calculate the molecular descriptors were selected based on their calculation accuracy, ease of access, free availability and recognition by the scientific community. According to the Organization for Economic Cooperation and Development (OECD) principles (Gramatica, 2007), an acceptable QSAR model must be easily and continuously applicable, and the prediction of the property under investigation can be reproduced by everyone and also applicable to new compounds.