Abstract
Studying viral–host protein–protein interactions can facilitate the discovery of therapies for viral infection. We use high-throughput yeast two-hybrid experiments and mass spectrometry to generate a comprehensive SARS-CoV-2–human protein–protein interactome network consisting of 739 high-confidence binary and co-complex interactions, validating 218 known SARS-CoV-2 host factors and revealing 361 novel ones. Our results show the highest overlap of interaction partners between published datasets and of genes differentially expressed in samples from COVID-19 patients. We identify an interaction between the viral protein ORF3a and the human transcription factor ZNF579, illustrating a direct viral impact on host transcription. We perform network-based screens of>2,900 FDA-approved or investigational drugs and identify 23 with significant network proximity to SARS-CoV-2 host factors. One of these drugs, carvedilol, shows clinical benefits for COVID-19 patients in an electronic health records analysis and antiviral properties in a human lung cell line infected with SARS-CoV-2. Our study demonstrates the value of network systems biology to understand human–virus interactions and provides hits for further research on COVID-19 therapeutics.
Main
The severity of the global COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) highlights the need to fill in the gaps in our understanding of the interplay between this virus and its hosts. Once inside a cell, viruses interact with intracellular proteins to hijack host mechanisms that facilitate viral replication and evasion of an immune response1. Studying viral–host protein–protein interactions (PPIs) is therefore crucial for understanding the mechanisms of the viral infection and the host response and to develop new strategies for disease treatment and prevention2,3,4,5. Interaction networks are especially important as proteins generally act not in isolation but in concert with their neighborhood of interacting partners. Such interactomes can thus reveal biological pathways and processes impacted by the viral proteome, allowing for the discovery of novel drug targets.
Here we leverage high-throughput yeast two-hybrid (Y2H) and tandem mass tag affinity purification followed by mass spectrometry (TMT-AP–MS) to generate a binary and co-complex SARS-CoV-2–human protein–protein interactome network, which we propose to be a more complete resource for exploration of the viral–host interactome (Fig. 1a). We adopted this approach for several reasons. To date, Y2H and AP–MS are the only two methods available for mapping protein–protein interactome networks on a proteomic scale6,7. Pioneering studies on the earliest SARS-CoV-2–human interactomes utilized label-free AP–MS as their sole method for interaction mapping2,3,4,5. While both Y2H and AP–MS alone produce high-quality interactome datasets, they fundamentally capture different yet complementary aspects of the full network; specifically, Y2H interactions often represent key connections between different protein complexes and pathways8. Thus, Y2H and AP–MS together can provide a more comprehensive view of the topological and biological properties of the interactome8. Moreover, labeled (for example, TMT-based) AP–MS has been shown to provide more precise, accurate and reproducible quantification of proteins compared to label-free AP–MS-based approaches, which is an important criterion when trying to identify true protein interactions and generate high-quality interactome networks9,10,11,12,13,14,15,16,17,18,19.
Fig. 1: SARS-CoV-2–human protein interactome.
a, Pipelines using Y2H and AP–MS for detecting SARS-CoV-2–human protein–protein interactions. b, Edges between viral proteins (diamonds) and human proteins (circles) represent protein–protein interactions. Edge colors indicate the methods used to detect the protein–protein interaction. Several biological processes that are significantly enriched in these human proteins (Supplementary Fig. 2 and Supplementary Table 2) are highlighted with yellow background. Human proteins that interact with only one SARS-CoV-2 protein are shown in the box connected to that specific protein. The interactome can be found in Supplementary Table 1.
Source data
Here we used both Y2H and quantitative TMT-AP–MS to generate a total of 739 high-confidence interactions among 579 human proteins and 28 SARS-CoV-2 proteins. Our interactome had an unprecedented scale and coverage compared with existing ones. Using our interactome, we identified important pathways such as protein translation, mRNA splicing, Golgi transportation, neutrophil-mediated immunity and glucose metabolism. Moreover, we prioritized host-targeting therapies by searching U.S. FDA-approved and investigational drugs for their potential anti-SARS-CoV-2 effect using state-of-the-art network proximity methods. Using two large independent COVID-19 patient databases, we found that usage of one of the top candidates, carvedilol, was associated with a lowered risk (17–20%) of a positive COVID-19 test. Experimental validation shows that carvedilol inhibits SARS-CoV-2 infection with a half-maximal effective concentration (EC50) of 4.1 µM. Altogether, these results suggest that our comprehensive SARS-CoV-2–human protein interactome offers substantial opportunities for understanding the pathobiological process of SARS-CoV-2 in human and identifying host-targeting therapies for COVID-19.
ResultsA comprehensive SARS-CoV-2–human protein–protein interactome
To generate a binary SARS-CoV-2–human protein–protein interactome, we systematically tested all pairwise combinations of 28 SARS-CoV-2 proteins (GenBank accession MN908947) against ~16,000 human proteins (hORFeome V8.1)20 using high-throughput Y2H screens8,21,22,23,24 (Fig. 1a). We treated each protein as both a bait and a prey, yielding over 896,000 (28 × ~16,000 × 2) total tested pair combinations. Before screening, all autoactivating DNA-binding domain (DB) ORF clones were removed from further tests (see Methods). To increase experimental throughput, viral ORF activating domain (AD) and DB clones were mated against pools of 24 human ORF DB or AD clones, respectively. Following auxotrophic selection, AD–DB pairs were identified via PLATE-seq24 to generate a list of candidate interactions (Methods). Interaction candidates were then subsequently re-tested using Y2H to ascertain high reproducibility. In all, we report a total of 299 high-quality binary SARS-CoV-2–human PPIs via our high-throughput Y2H screen, 267 of which were unique to this assay in this study (Supplementary Table 1).
To complement our binary SARS-CoV-2–human protein–protein interactome, we independently expressed each of the 28 SARS-CoV-2 proteins in the human intestinal epithelial cell line Caco-2 (HTB-37; ATCC) to identify viral–host co-complex interactions using TMT-AP–MS proteomics (Fig. 1a). We used Caco-2 as our cell line model owing to its endogenous expression of angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2) required for SARS-CoV-2 cell entry and S protein priming, respectively25, the extensive use of the line in SARS-CoV and SARS-CoV-2 infection studies26,27, supported by known in vivo replication of SARS-CoV-2 in gastrointestinal cells28,29 and desirable cell culture characteristics including robust transfectability and rapid propagation. All Strep-, Myc-, or FLAG-tagged SARS-CoV-2 baits and their corresponding empty vector controls were transfected in biological duplicates, followed by subsequent affinity purification, TMT labeling and synchronous precursor selection (SPS) MS3-based quantification. We filtered for interactions that met stringent fold change (FC) and P value cutoffs (Methods). In all, we report a total of 472 high-confidence co-complex SARS-CoV-2–human PPIs via AP–MS, 440 of which were unique to this assay in this study (Supplementary Table 1). Altogether, our orthogonal approaches generated a network composed of 739 interactions among 28 viral and 579 host proteins (Supplementary Table 1).
We visualized the SARS-CoV-2–human protein–protein interactome through a network shown in Fig. 1b. The colors of the edges between the viral proteins (represented as diamond nodes) and the host proteins (represented as circle nodes) indicate the methods that detected the interaction. Host proteins that interact with a single viral protein are shown in boxes connected to their interacting partner. Several human proteins interact with multiple SARS-CoV-2 proteins, such as ACTN4, ITGB1BP2, TRIM27 and ACTN1, while the majority of human proteins (469, 81%) interact with only one SARS-CoV-2 protein (Supplementary Fig. 1a). Among the viral proteins, N, ORF7b and ORF9b achieved the highest network degrees, whereas E, NSP7 and NSP1 have the lowest network degrees (Supplementary Fig. 1b). In terms of the shared interacting partners, overall, the viral proteins showed low overlap (Supplementary Fig. 1c), consistent with a previously published SARS-CoV-2 interactome network2,3. We examined the overlap of host factors for Y2H and AP–MS separately and found overall low overlap of host factors as well (Supplementary Fig. 1d,e).
For the entire interactome, functional enrichment analysis revealed significantly overrepresented biological processes (Supplementary Fig. 2a and Supplementary Table 2), including protein translation, transcription and neutrophil-mediated immunity (highlighted with yellow background in Fig. 1b). Semantic analysis shows major biological process categories such as ‘ribosome biogenesis,’ ‘rRNA metabolic process,’ and ‘viral gene expression’ (Supplementary Fig. 2b). Pathway enrichment analysis show top enriched pathways such as ‘protein processing in the endoplasmic reticulum,’ ‘tight junction,’ ‘glycolysis,’ ‘ribosome,’ and ‘protein export’ (Supplementary Fig. 2c and Supplementary Table 2). For individual SARS-CoV-2 proteins, many pathways and biological processes are shared in these viral proteins (Supplementary Fig. 3). For example, NSP12, NSP13 and NSP16 share biological processes such as ‘regulation of cellular component movement,’ ‘negative regulation of cell morphogenesis involved in differentiation’ and ‘negative regulation of substrate adhesion-dependent cell spreading’ (Supplementary Fig. 3a); ORF7a, ORF7b, ORF8 and NSP4 share the pathway ‘protein processing in endoplasmic reticulum’ (Supplementary Fig. 3b).
Given the surge of COVID-19-related studies since 2020, we repeated the enrichment analyses using gene set libraries generated before and after the start of the pandemic to evaluate whether bias was introduced to the gene sets. By comparing Gene Ontology (GO) biological process 2018 versus GO biological process 2021 and Kyoto Encyclopedia of Genes and Genomes (KEGG) human pathway 2021 versus KEGG human pathway 2019, we found that our enrichment analyses were not biased by the addition of the COVID-19 research. In both comparisons, we found that terms such as ‘ribosome biogenesis’, ‘rRNA processing’ and ‘rRNA metabolic process’ and pathways such as ‘Ribosome’ were significantly enriched in the gene set libraries both before and after the pandemic, with similar odds ratio and combined score (Supplementary Fig. 3c).
Overall, our interactome is comprised of abundant information that can be utilized for the identification of COVID-19-relevant pathobiology and host-targeting therapies. We also developed an interactive visualization tool for our interactome which can be accessed from https://github.com/ChengF-Lab/COVID-19_PPI.
Coverage and quality of our interactome
To ensure the authenticity when applying our interactome for downstream studies, we first evaluated the quality through several means. We examined three previously published SARS-CoV-2–human protein–protein interactome networks2,4,5. All three of these interactomes were generated using AP–MS-based methods alone. Overall, we found that the host factors of these interactomes significantly overlap (Fisher’s exact test, false discovery rate (FDR)
Information contained on this page is provided by an independent third-party content provider. This website makes no warranties or representations in connection therewith. If you are affiliated with this page and would like it removed please contact editor @saltlakecityutah.business
