Use of Molecular Dynamics Simulations to Characterize Structural Interactions of Protein and Drug in Antiviral Drug Discovery

Authors

• Caitlin Joan Smith, University of Alabama at Birmingham

Advisor(s)

Kevin Harrod

Committee Member(s)

Elliot Lefkowitz
Todd Green

School

Joint Health Sciences (Interdisciplinary)

Document Type

Thesis

Department (new version)

Joint Health Sciences

Date of Award

9-9-2024

Abstract

Since 2019, the novel coronavirus SARS-CoV-2 the etiological cause of the COVID-19 pandemic has led researchers and physicians to look for effective antiviral treatments as interventions in mitigating COVID-19 disease. The severity of the disease and the limitations of current treatments have led to antiviral drug research beyond currently approved antiviral drugs. Many antivirals exist and are currently approved by the Food and Drug Administration (FDA) for other viral illnesses, yet currently only remdesivir and nirmatrelvir/ritonivir are approved for COVID-19 (October 2020, May 2023, respectively), while molnupiravir is still under emergency FDA use authorization [1, 2, 3]. Here, we have used a computational approach to investigate the interaction of novel antivirals against the RNA-dependent RNA-polymerase (RdRp). The RdRp of coronaviruses consists of the non-structural protein 12 (nsp12) RNA synthesis protein, two nsp8 proteins, and a single nsp7 protein [4]. The nsp8 and nsp7 proteins serve as accessory proteins supporting the function of nsp12. Our lab has identified d-α-Tocopherol-polyethelene-glycol-succinate (TPGS) through an AI drug discovery approach as a candidate in antiviral discovery. In this study we use an in-silico binding and molecular dynamics approach to model the interaction of TPGS with the replication transcription complex (RTC) for the possibility of using these treatments to inhibit replication of SARS-CoV-2. Each protein was reconstructed using AlphaFold2 and the drug interactions were modeled using Vina AutoDock. Comparison of the molecular dynamics were simulated using GROMACS on the UAB supercomputer CHEAHA, and analysis performed using the root mean squared deviation (RMSD) and root mean squared fluctuation (RMSF). We found high confidence binding sites in several areas of the individual proteins and the RdRp complex. Statistical significance was determined by an ANOVA of the RMSD measured in nanometers. Here, the effects of introducing TPGS or its derivatives into the replication-transcription complex to reduce viral transcription are assessed. The findings from this work inform about the molecular interactions of a novel class of putative antivirals against SARS-CoV-2.

ProQuest ID

https://proquest.com/docview/3101935082