To inform drug design for COVID-19: Computational determination of interactions between COVID-19 gene products and structurally diverse natural products and clinical drugs

疫苗是防治包括病毒在内的传染病的首选公共卫生解决方案。然而，仍然迫切需要开发小分子疗法作为治疗病毒感染的正交方法。该提案旨在深入了解COVID-19的潜在药物线索。世界各地正在破解COVID-19冠状病毒基因组产生的蛋白质和酶的x射线晶体结构，并将其存入公开可用的数据库。其中一些三维结构包括与活性位点、变构位点或蛋白质-蛋白质相互作用位点结合的小分子。这一发现为发现用于治疗COVID-19患者的抗病毒药物的铅分子提供了起点。然而，化学多样性比通过x射线晶体学鉴定的少数支架要丰富和多样得多。在这项NSERC联盟提案中，我们将使用计算方法筛选已发表的COVID-19病毒基因产物的x射线结构以及结构多样化的分子（数据库），以确定对接相互作用。因此，我们将根据分子的计算结合亲和力和结构姿态对分子进行排序。这将提供一个多样化的结构数据集，作为药物发现项目的起点，作为药物先导进行评估。本文的初步结果表明，计算对接预测并重现了最近发表的强效抑制剂与SARS-CoV-2主要蛋白酶（负责激活COVID-19多肽的蛋白酶）的x射线结构结合。这些数据表明，计算方法是鉴定铅分子的一个有意义的步骤。及时公布研究结果，确保研究成果迅速传播。此外，结果将与合作伙伴组织（IntelliSyn, Montreal）共享，这是一个合同研究组织和加拿大境内独特的药物发现团队，以这些结果为基础。

Vaccines are the preferred public health solution to combating infectious disease, including viruses. However, there remains a strong need to develop small molecule therapeutics as an orthogonal approach to treat viral infections. This proposal aims to provide insight into potential drug leads for COVID-19. X-ray crystal structures of the proteins and enzymes produced from the COVID-19 coronavirus genome are being solved around the world and deposited into publicly available databases. Some of these three-dimensional structures include small molecules bound to active sites, allosteric sites or sites of protein-protein interactions. This insight offers a starting point for the discovery of lead molecules to be developed as antivirals to treat COVID-19 infected patients. However, chemical diversity is much richer and more varied than the few scaffolds identified through X-ray crystallography. In this NSERC Alliance proposal, we will screen, using computational approaches, published X-ray structures of the gene products from the COVID-19 virus alongside structurally diverse molecules (databases) in order to determine docking interactions. We will thus rank the molecules in terms of their computational binding affinity and structural pose. This will provide a diverse dataset of structures to be evaluated as drug leads as a starting point for drug discovery projects. Preliminary results included herein indicate that computational docking predicts and reproduces the very recently published X-ray structural binding of potent inhibitors to the SARS-CoV-2 main protease, the protease responsible for activating the COVID-19 polypeptide. These data demonstrate that computational approaches are a meaningful step towards identification of lead molecules. The results will be published in a timely manner to ensure rapid dissemination of the research results. Furthermore, the results will be shared with the partner organization (IntelliSyn, Montreal) a contract research organization and drug discovery team uniquely poised within Canada to build upon these results.