Highly accurate protein structure prediction and drug screen of monkeypox virus proteome

We recently came accross an interesting article published in this journal by Usman Ayub Awan et al.,1Awan U.A. et al.Monkeypox: a new threat at our doorstep!.J Infect. 2022; 85: e47-e48https://doi.org/10.1016/j.jinf.2022.05.027Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar which highlighted that the monkeypox virus had become another global emergency threatening human health under the COVID-19 pandemic. An outbreak of the Monkeypox virus, which was first reported in the United Kingdom in May of 2022 and has since spread to more than 72 territories causing up to 14,533 cases according to the World Health Organization (WHO), as of the 20th of July 2022 (www.who.int/). In response to rising concerns over monkeypox, researchers are focused on acquiring structures of essential monkeypox proteins and have succeeded in producing crystal structures of the A42R Profilin-like protein (PDB: 4QWO), as well as simulating the structure of envelope protein F13, also known as C19L.2Li D. Liu Y. Li K. Zhang L. Targeting F13 from monkeypox virus and variola virus by tecovirimat: molecular simulation analysis.J Infect. 2022; https://doi.org/10.1016/j.jinf.2022.07.001Abstract Full Text Full Text PDF Scopus (17) Google Scholar The structures of most of the monkeypox virus proteins remain unknown, knowledge of which would significantly enhance our understanding of the molecular mechanisms underlying critical viral processes such as viral entry and replication. AlphaFold2 is a powerful open-source computational approach developed to help predict protein structures,3Jumper J. et al.Highly accurate protein structure prediction with AlphaFold.Nature. 2021; 596: 583-589https://doi.org/10.1038/s41586-021-03819-2Crossref PubMed Scopus (8095) Google Scholar which has been used successfully in acquiring accurate protein structures of the human and SARS- CoV-2 proteomes.4Tunyasuvunakool K. et al.Highly accurate protein structure prediction for the human proteome.Nature. 2021; 596: 590-596https://doi.org/10.1038/s41586-021-03828-1Crossref PubMed Scopus (982) Google Scholar, 5Yang Q. et al.Structural comparison and drug screening of spike proteins of ten SARS-CoV-2 variants.Research. 2022; 20229781758https://doi.org/10.34133/2022/9781758Crossref Scopus (8) Google Scholar, 6Yang Q. Syed A.A.S. Fahira A. Shi Y. Structural analysis of the SARS-CoV-2 omicron variant proteins.Research. 2021; 20219769586https://doi.org/10.34133/2021/9769586Crossref Scopus (13) Google Scholar Here, we used AlphaFold2 to predict the protein structures of the reference monkeypox virus proteome (Uniprot ID: UP1012697Shchelkunov S.N. et al.Human monkeypox and smallpox viruses: genomic comparison.FEBS Lett. 2001; 509: 66-70https://doi.org/10.1016/s0014-5793(01)03144-1Crossref PubMed Scopus (0) Google Scholar), yielding a total of 186 highly accurate protein structures (Supplemental Table S1). The mean predicted Local Distance Difference Test (pLDDT) values of 156 of the 186 proteins are above the threshold of 70, which suggests that the predicted structures for these proteins are highly accurate (Fig. 1A). Furthermore, among these proteins, 77 protein structures had mean pLDDT values of between 80 and 90, while 57 protein structures showed mean pLDDT values above 90. To further evaluate the predicted protein structures, we compared the AlphaFold-predicted structure with the experimental crystal structure (PDB: 4QWO) of A42R Profilin-like protein. The resulting Template Modelling (TM), Global Distance Test (GDT)-TS, and the maximal subset (MaxSub) scores between two protein structures are above 0.95 (Fig. 1B). The aligned structure between two proteins has a root means square deviation (RMSD) value of 0.39 (Figs. 1B and 1C), suggesting that the two protein structures are highly similar (Figs. 1B and 1C). Taken together the above suggests that the predicted structures can be considered to be highly accurate. Previously approved drugs for smallpox, tecovirimat and brincidofovir have been shown to be effective against the monkeypox virus, both in vitro and in animals8Delaune D. Iseni F. Drug development against smallpox: present and future.Antimicrob Agents Chemother. 2020; 64https://doi.org/10.1128/aac.01683-19Crossref PubMed Google Scholar; however, given the global emergency, it is necessary to identify alternative drugs that could be used to fight the outbreak. Thus, we endeavored to uncover potential drugs to treat monkeypox based on accurate prediction of the monkeypox protein structures and 5903 approved drugs from the Zinc database9Sterling T. Irwin J.J. ZINC 15–ligand discovery for everyone.J Chem Inf Model. 2015; 55: 2324-2337https://doi.org/10.1021/acs.jcim.5b00559Crossref PubMed Scopus (1751) Google Scholar (Supplemental Table S2). We first selected ten target proteins based on their high pLDDT values, essential functions, and potential pharmacophores. The selected proteins were B4R, A42R, PRO132, VITF3L, E8L, I1L, D10L, P28, and G9R. We subsequently calculated the binding energies of all ten proteins with the 5903 drugs (Supplemental Table S3). The top best docking drugs targeting the ten proteins showed low binding energies (Table 1, S3, and FigureS1-S10), suggesting they might have strong interactions.Table 1Virtual screening of potential drugs targeting the ten proteins. Related to Table S3.ProteinsDrugsValuesProteinsDrugsValuesB4RZINC164528615−11.3G9RTrypan Blue−9.4Trypan Blue−11.2ZINC3934128−8.7ZINC208774715−11.1ZINC40164432−8.7Ixabepilone−10.9ZINC14879959−8.6ZINC40165257−10.8Irinotecan−8.5Differin−10.7ZINC8220175−8.5ZINC43195321−10.6ZINC14879961−8.4Vumon−10.5Indacaterol-8-O-Glucuronide−8.4P28Yaz−9.7D10LLumacaftor−9.3Ergotamine−9.2ZINC3934128−9.1Xaliproden−9Dihydroergotamine−9Indocyanine Green−9Trypan Blue−9Trypan Blue−9Dihydroergotoxine−8.9Bisoctrizole−8.9Gliquidone−8.8Orobronze−8.9Lestaurtinib−8.8ZINC1530886−8.8ZINC253632968−8.7I1LCepharanthine−11.1E4RGlycyrrhizinate Dipotassium−9.7Trypan Blue−11.1Trypan Blue−9.7ZINC3934128−10.2Nilotinib−9.6ZINC14880001−10.2Naldemedine−9.6Avodart−10.1Lifitegrast−9.5ZINC3917540−9.9ZINC936069565−9.5Lumacaftor−9.9ZINC14880001−9.4Nilotinib−9.7ZINC43195321−9.4PRO132Antrafenine−11.4VITF3LTrypan Blue−9.6Dihydroergotamine−11ZINC164528615−9.1ZINC43195321−11Midostaurin−9Cabozantinib−11Nilotinib−8.9Trypan Blue−11Cepharanthine−8.9Lorazepam Glucuronide−10.9ZINC253633751−8.9ZINC8234383−10.8Avodart−8.8Lomitapide−10.8ZINC14880001−8.8E8LTrypan Blue−11.4A42RTipranavir−7.9Dihydroergotoxine−10.9Trypan Blue−7.9Naldemedine−10.9ZINC936069565−7.9ZINC14880001−10.6Cepharanthine−7.8Irinotecan−10.3Daclatasvir−7.8Cepharanthine−10.2Dihydroergotamine−7.7Fluspirilene−10.1Penfluridol−7.7Dihydroergotamine−10.1Dihydroergotoxine−7.7 Open table in a new tab Trypan Blue and Cepharanthine display significant binding affinities to all ten target proteins (Supplemental Table S3). Trypan Blue interacts with the proteins by forming hydrogen bonds, salt bridges and hydrophobic and pi-cation interactions (supplemental materials). Cepharanthine shows high binding affinities to I1L (Fig. 1D1), VITF3L (Fig. 1 D2), A42R (Fig. 1 D3), and E8L (Fig. 1D4) through hydrophobic interactions and hydrogen bond formation. We have generated an extensive drug dataset consisting of 43,800 protein-drug paired data for the monkeypox virus, which could contribute to discovering the potential drugs for fighting monkeypox. Combatting the outbreak would require rapid research in the following: 1. Development of suitable pseudovirus and animal models of the current subtype of monkeypox virus; 2. Clarification of the mechanism of infection, reproduction, and transmission of monkeypox virus; 3. Identify and validate potential drugs that can be used to treat the viral infection. In conclusion, here we acquired 186 highly accurate protein structures of the monkeypox virus reference proteome using AlphaFold2, which provided the most comprehensive database of monkeypox virus protein structures worldwide. Moreover, we have screened the potential drugs for binding the ten crucial proteins of the monkeypox virus, including B4R, A42R, PRO132, VITF3L, E8L, I1L, D10L, P28, and G9R, and generated a drug dataset containing total 43,800 protein-drug paired data, which could be helpful for drug discovery to the monkeypox virus. Q.Y. and S.Y. conceived and designed the study. Q.Y. and X.S. analyzed and interpreted the data. Q.Y., A.A.S.S, and S.Y. wrote the first draft. Z.W. provided the softwares. The detailed methods are described in supplemental Methods. The protein structural files predicted by AlphaFold2 and the interaction analysis results between drugs and proteins were available on Github (https://github.com/lg10is1/monkeypox-virus-protein-structures). The RStudio code used in this study to perform statistical analysis and visualize data are available upon request. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. This work was supported by Shanghai Municipal Science and Technology Major Project (2017SHZDZX01), Natural Science Foundation of China (32070679, U1804284, 81871055), the National Key R&D Program of China (2019YFA0905400, 2017YFC0908105, 2021YFC2702103), Taishan Scholar Program of Shandong Province (tsqn201812153) and Natural Science Foundation of Shandong Province (ZR2019YQ14). The computations in this paper were run on the Siyuan cluster supported by the Center for High-Performance Computing at Shanghai Jiao Tong University. Download .pdf (3.46 MB) Help with pdf files Download .docx (.04 MB) Help with docx files Download .csv (.0 MB) Help with csv files Download .csv (.65 MB) Help with csv files Download .xlsx (1.58 MB) Help with xlsx files