Binding repurposed drugs and aminothiourea derivatives to SARS-CoV-2 enzymes—a docking perspective CURRENT

Binding of over 20 approved drugs proposed for the repurposing for COVID-19 treatment and over 160 aminothioureas derivatives to SARS-CoV-2 enzymes which structures became available very recently have been evaluated using a few docking algorithms. These studies support potential effectiveness of homobarringtonine, chloroquine, rimcazole, and benserazine. From among studied aminothioureas thiadiazoles with pyrrol-derived substituents at the carbon atom, and ortho hydroxyphenyl at the nitrogen atom are potential lead compounds for future drugs development. Highlights Based on docking results to SARS-CoV-2 enzymes homobarringtonine, chloroquine, rimcazole, and benserazine are approved drugs with potential use in treatment of COVID-19 infection Pyrrol substituted thiadiazoles with ortho or meta substituted phenyl group attached to the nitrogen atom show strongest binding from among the studied aminothioureas


Introduction
With the outbreak of the SARS-CoV-21 devastating pandemic that claimed already lives of over 50000 people worldwide2 the need for therapeutics is needed urgently. For the future prevention a few vaccine strategies are being tested. At the same time medicines that can be used in fighting the infection are being sought.
The family of coronaviruses is relatively well known due to the previous severe for humans epidemies of SARS3 and MERS. 4 In haste search for medicines instantly available, repurposing of already approved drugs is carried out as the first line of defenses. This approach has been attempted during the MERS epidemy,5 and is also being tried currently. 6 In the first of these studies an NIH library of over 700 compounds has been screened experimentally, identifying around 80 with the anticoronavirus effect. The recent study concentrated on the interaction of the spike protein (S-protein) of the virus with the human ACE2 receptor, which is thought to be responsible for the viral recognition of host cells. By combining virtual high-throughput screening with ensemble docking of over 9000 compounds from the SWEETLAND library,7 and using COVID-19 S-protein (NCBI: YP_009724390.1) and human ACE2 receptor (PDB: 2AJF) as a template to generate the SARS-CoV-2 S-protein and ACE2 receptor complex model, around 80 compounds already in clinical use that should exhibit anticoronavirus activity have identified.
While repurposing of compounds currently in medical use is the necessity for quick response to public health threat, in the long run drugs specific to this particular virus will be needed. With this aim in mind we have used the SARS-CoV-2 S-protein -ACE2 receptor complex model and its components to evaluate binding properties of thiosemicarbazides, thiadiazoles, and triazoles that we have studied for their anti-Toxoplasma gondii, 8,9,10,11 antiviral,12 antibacterial,13,14,15,16,17,18 anticancer,17,19,20,21

Studied compounds
Approved drugs, selected as the best candidates in studies previously reported in literature,5,6 and compounds containing N-N-C(S)-N skeleton have been studied. The considered drugs are listed in Table 1 and referred to as series L1 (best from reference 6) or L2 (best from reference 5).
Additionally, we have studied three drugs (labeled A-1 -A-3). Chloroquine (A-1) was originally used as antiparasitic drug for treatment of malaria, but it has been shown to have also antiviral activity32 and has been currently recommended recently for use in SARS CoV-2 infection.33,34 Thus we have considered the other two which also are potent antiparasitic.
Among the compounds comprising the N-N-C(S)-N skeleton we have evaluated representatives of three classes of linear or cyclic molecules that have been studied in our laboratory in recent years for their inhibitory activity against a few enzymes. Their structures contain three main cores; linear carbonylthiosemicarbazide and its two cyclic derivatives: 1,3,4-thiadiazole and 1,2,4-triazole each decorated with two substituents. As the C-substituent one of the four five-member rings was used while the N-substituent was a benzene ring or its ortho, meta, or para substituted derivatives. All these components are collected in Table 2 where also partial codes for the fragments are provided (in bold). Thus, for example, compound code ICpNO2 corresponds to 1-(4-methylimidazol-5-oyl)-4-(4-nitrophenyl)thiosemicarbazide, illustrated in Fig. 1. This compound (together with CCpNO2, CCmI, other IC) has been shown to have no cytotoxicity in our recent studies on anti-Toxoplasma gondii activity (unpublished results).
In total, 191 have been studied; 167 of the above compounds and 24 drugs.

In silico evaluations
We have used the published model6 of the interface between the viral S-protein and the human ACE2 program has been used. However, the scores obtained for strongest inhibitors were very close and not discriminating. We have, therefore switch to FlexX algorithm,37 as implemented in the LeadIT platform. 38 The scoring function of this program did better job in differentiation of binding affinities of studied compounds but showed some constrains, the most serious being the limit on the size of the acceptor, which precluded in some cases searches of the whole enzyme. SwissDock 39 at our hands also turned out not practical as only a single ligand per submission to the server was possible. We have therefore finally decided to continue with the ChemPLP algorithm40 as implemented in the Gold program.41 It is also noteworthy that this algorithm has been considered as one of the best in most recent benchmark studies.42

Results And Discussion
The focus of the present studies was on the binding to virus S-protein, ACE2 human receptor, and their interface. The best 10 scores for each target, obtained using ChemPLP docking algorithm, are collected in the first 23 rows of Table 3  In the bottom part of Table 3 Table 4. As can be seen again the currently used drugs, L2-1 and L1-5 exhibit the largest scores. Also, A1 recently recommended for inclusion in the anticoronavirus treatment shows very high affinity to the phosphatase although this class of enzymes is generally considered undruggable. 48

Conclusions
Presented computational studies support recent observations that repurposing some drugs seems a promising way for COVID-19 treatment. Out of the studied drugs the most promising is homobarringtonine (L2-6) in agreement with recently suggested49 its potential effectiveness of this drug in treating SARS-CoV-2 infection. Similar arguments can be claimed for chloroquine (A1).
Furthermore, our studies point to rimcazole (L2-1), and benserazine (L1-5) as equally potential antivirus drugs especially because they seem to be nonspecific and strongly binding to a few virus enzymes.
From among studied compounds that are not approved for clinical use, it seems that thiadiazoles with pyrrol or its derivatives substituents at the carbon atom, and ortho-hydroxyphenyl at the nitrogen atom might provide reasonable lead compounds for future drugs. Figure 2 Chloroquine (A-1) bound to virus S-protein -human ACE2 receptor interface