Boston University
Computational identification and mechanistic evaluation of novel Lassa virus entry inhibitors
Abstract
dc:description.abstractLassa virus, a zoonotic Old World mammarenavirus endemic to West Africa, causes significant disease burden and is a serious public health threat. With an estimated 2.7 million infections per year, it is categorized by the World Health Organization as a priority pathogen for research in emergency contexts. It is the causative agent of Lassa fever, a deadly hemorrhagic fever marked by multiorgan failure and deafness in 25% of survivors. Despite demand for effective countermeasures, no therapeutic or vaccine has been approved in the nearly 60 years since its discovery, and current treatment options are limited to supportive care and off-label use of ribavirin, a nucleoside analog with variable efficacy. Lassa virus entry is a coordinated, multistep process driven entirely by the virus glycoprotein, involving interactions with both extra- and intracellular receptors, endosomal trafficking, and pH-dependent membrane fusion. Inhibitors of viral entry are desirable candidates for therapeutic development because they block the earliest stages of infection, effectively reducing viral load and limiting cell-to-cell transmission. To identify novel Lassa virus entry inhibitors, we conducted a high-throughput screen of the National Institutes of Health Molecular Libraries Small Molecule Repository (MLSMR), consisting of approximately 300,000 small molecules. High-throughput screening has become a core component of drug discovery; however, prioritization of actionable candidates from such extensive libraries remains a major bottleneck. Many standard computational strategies rely on cheminformatics methods that initially filter by structural or physicochemical properties and are agnostic to biological activity. In contrast, our work implemented an activity-driven approach by data mining the PubChem database to develop the bioassay activity percentage (BAP) metric, which evaluates the empirical activity profile of a compound across the database. This serves as a quantitative, context-based complement to traditional structure-first filtering to better prioritize highly potent, structurally diverse, and mechanistically selective viral entry inhibitors. Our computational framework was experimentally validated using pseudotyped and wild-type viruses, confirming antiviral activity in 28 of our 30 computationally selected inhibitors. We then prioritized five highly potent, selective, and structurally unique lead candidates for phenotypic and genotypic mechanistic evaluation, elucidating the targeted stages of entry and identifying glycoprotein residues that confer sensitivity to compound inhibition. From these investigations, the five leads were categorized into two mechanistic classes of early-stage attachment and internalization inhibitors and late-stage membrane fusion inhibitors. Ultimately, this work demonstrates the power of integrating activity-driven computational prioritization with in-depth mechanistic evaluation to accelerate the identification of structurally and mechanistically novel small molecule inhibitors of Lassa virus entry.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Microbiology
- Grantor
- Boston University
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Close, Brianna Joy
- Advisor dc:contributor.advisor
-
- Davey, Robert A
- ORCID dc:identifier.orcid
Subjects
dc:subject × 7Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/2144/53461