TL;DR
This study presents a comprehensive approach to identify host-directed drug targets for SARS-CoV-2 by combining large-scale compound screening with network analysis. They identified obatoclax as a promising candidate that significantly reduced viral load in mice, highlighting the potential of host-targeted therapies for viral diseases.
What Problem Does This Paper Address?
The COVID-19 pandemic, caused by SARS-CoV-2, has highlighted the urgent need for effective therapeutics. While vaccines have been developed, the emergence of new variants and the potential for drug resistance necessitate the exploration of alternative therapeutic strategies. This paper addresses the challenge of identifying host factors that are essential for SARS-CoV-2 replication, as targeting these host dependencies offers a more robust approach to antiviral therapy, potentially circumventing viral resistance mechanisms. The authors aim to identify druggable host targets by systematically screening a large library of existing drugs and analyzing their impact on viral infection in vitro and in vivo, coupled with computational network analysis to prioritize targets.
Methods
The study employed a multi-pronged approach:
- High-Throughput Screening (HTS): A library of 6,710 clinical and preclinical compounds (Drug Repurposing Hub) was screened using an immunofluorescence-based assay to detect SARS-CoV-2 N protein expression in Vero E6 cells. This assay was optimized for high-throughput screening and demonstrated a good Z-factor.
- Mechanistic Assays: Promising hits from the HTS were further evaluated in secondary assays to assess their impact on various stages of the viral lifecycle, including cell entry, genome replication, and egress. These assays were conducted in human cell lines (Huh-7, A549) and primary human cell models.
- Computational Analysis:
- Structural Similarity Analysis: Morgan fingerprints were used to encode molecular structures, and Jaccard similarity was employed to identify structurally related compounds.
- Network Analysis: A human protein-protein interaction network was used to analyze the relationships between drug targets. This included calculating the largest connected component (LCC) of drug targets and assessing their network proximity to SARS-CoV-2 binding proteins.
- Gene Ontology (GO) Enrichment: GO enrichment analysis was performed to identify enriched biological processes and molecular functions associated with the identified drug targets.
- Drug-Target Network Construction: Bipartite networks were built to represent the relationships between drugs and their targets.
- In Vivo Studies: The efficacy of promising compounds was evaluated in K18-hACE2 transgenic mice, a model for SARS-CoV-2 infection. Virus load in the lungs was measured after treatment.
- Variant Testing: The efficacy of selected compounds was also tested against Beta and Delta variants of SARS-CoV-2 and MERS-CoV.
Datasets
- Drug Repurposing Hub (DRH) Library: A collection of 6,710 compounds, enriched with FDA-approved drugs, drugs in clinical trials, and extensively pre-clinically characterized molecules.
- SARS-CoV-2 strains: Wild-type SARS-CoV-2 (USA-WA1/2020), SARS-CoV-2 Beta (B.1.351), and SARS-CoV-2 Delta (B.1.617).
- MERS-CoV: EMC/2012 strain.
- Human Protein-Protein Interaction (PPI) Network: Used for computational analysis of drug targets.
Key Insights and Ideas
- Host-Targeted Therapies as a Robust Strategy: The study reinforces the idea that targeting host factors essential for viral replication is a promising strategy for developing broad-spectrum antivirals that are less prone to resistance.
- Obatoclax as a Promising Candidate: Obatoclax emerged as a standout compound, demonstrating potent antiviral activity across multiple in vitro systems and significantly reducing viral load in a mouse model. Its mechanism, potentially involving endosomal acidification, warrants further investigation.
- Network Analysis for Target Prioritization: Integrating compound screening data with protein-protein interaction networks allowed for the identification of drug targets that are not only effective against the virus but also reside in biologically relevant network neighborhoods. This approach helps to prioritize targets and understand their functional context.
- Structural Similarity and Target Overlap: The analysis revealed that structurally similar compounds often target similar or homologous proteins, providing a basis for understanding structure-activity relationships and guiding the development of new drug candidates.
- Distinct Drug Mechanisms: The study identified various mechanisms of action for the active compounds, impacting different stages of the viral lifecycle, from entry to egress.
Key Results
- Identification of Active Compounds: The primary screen identified 172 (2.56%) strongly active, 217 (3.23%) weakly active compounds, and 1.6% as cytotoxic.
- Obatoclax Efficacy: Obatoclax significantly reduced SARS-CoV-2 viral lung load in mice by >88%.
- Broad Spectrum Activity: Several compounds showed activity against different SARS-CoV-2 variants and MERS-CoV.
- Mechanistic Insights: Compounds were found to impact various stages of the viral lifecycle, including entry, RNA replication, and egress. For example, BET inhibitors were identified as potential viral egress inhibitors.
- No Phospholipidosis Induction: None of the tested compounds induced phospholipidosis, a potential confounding factor in drug repurposing screens.
Limitations and Future Work
- Variant Susceptibility Differences: The study noted differences in drug susceptibility for the Beta variant compared to the Washington strain, suggesting that viral evolution can lead to altered dependencies on host processes. This highlights the need to evaluate multiple virus strains.
- Therapeutic Window of Cardiac Glycosides: While potent, cardiac glycosides have narrow therapeutic windows, making them unlikely candidates for direct repurposing, but they offer insights into infection mechanisms.
- Further Characterization of Obatoclax: While obatoclax showed promise, its precise entry-independent modality of infection inhibition requires further detailed characterization.
- Methyltransferase Role: The role of methyltransferases in virus genome production and packaging has not been previously reported and warrants further investigation.
- Drug-Disease Interactions: The authors emphasize the importance of mindful consideration of potential novel drug-disease interactions when repurposing drugs, necessitating prospective clinical trial assessments.
- Data Availability: The authors have made their data publicly available through the DRH portal and GitHub, enabling further research and analysis.
Summary updated