ABSTRACT
Background: Five variants of concern (VOCs) have dominated COVID-19 disease etiology since 2020-Alpha, Beta, Gamma, Delta, and Omicron-possessing over 150 defining genomic alterations. Here, we used global proteomic and genomic approaches to study the host responses and selective forces driving VOC evolution. Method(s): We infected Calu-3 human lung epithelial cells with 5 VOCs and 2 wave 1 (W1) controls and performed mass spectrometry abundance proteomics, phosphoproteomics, and mRNA sequencing at 10 and 24 hours post infection. We additionally performed affinity purification mass spectrometry (APMS) by individually expressing all VOC mutant viral proteins (52) and corresponding W1 forms in human cells to quantify differential virus-host protein-protein interactions. Data was integrated using network modeling and bioinformatics to pinpoint VOC-specific differences. Four novel mutant viruses were developed using reverse genetics technology to validate the impact of specific genomic alterations. Result(s): We discovered VOCs evolved convergent molecular strategies to remodel the host response by modulating viral RNA and protein levels (most notably of N, Orf9b, and Orf6), altering nucleocapsid phosphorylation, and rewiring virus-host protein complexes. Integrative systems analyses revealed that Alpha, Beta, Gamma, and Delta ultimately converged in the suppression of interferon stimulated genes (ISGs) relative to W1 viruses, but Omicron BA.1 did not, and Delta induced more pro-inflammatory genes compared to other VOCs. Altered regulation of ISGs correlated with the expression of viral innate immune antagonist proteins, including Orf6, N, and Orf9b;for example, Omicron BA.1 depicted a 2-fold decrease in Orf6 expression. We identified mutations that alter expression of Orf9b (N D3L and N -3A del) and the novel VOC protein N* (N R203K/G204R), and confirmed Orf6 innate immune antagonism using recombinant virus technology. Remarkably, Omicron BA.4 and BA.5 regained strengthened innate immune antagonism compared to BA.1, which again correlated with enhanced Orf6 expression, though dampened in BA.4 by a mutation (D61L) that we discovered disrupts the Orf6-nuclear pore interaction. Conclusion(s): Collectively, our findings suggest SARS-CoV-2 convergent evolution overcomes human innate immune barriers, laying the groundwork to understand future coronavirus evolution associated with immune escape and enhanced human-to-human transmission.
ABSTRACT
Background: The novel coronavirus SARS-CoV-2, the causative agent of COVID-19, has caused worldwide social and economic disruption. Initial efforts to treat SARS-CoV-2 were hampered by limited knowledge of the molecular details of SARS-CoV-2 infection. To identify molecular targets for SARS-CoV-2 therapeutics, we mapped the host-pathogen protein interactions of SARS-CoV- 2, and investigated host dependency pathways that are required for SARS-CoV-2 infection using drug, knockdown and knockout screens. Concerns regarding the mutagenic potential of SARS-CoV-2 also led us to inquire whether a conserved set of human host factors may be required for infection by all highly pathogenic coronaviruses, thus representing pan-coronavirus drug targets. Therefore, we also mapped the host protein interactions of SARS-CoV-1 and MERS-CoV. Methods: We cloned, tagged and expressed proteins encoded by SARS-CoV-2, SARS-CoV-1, and MERS-CoV in HEK-293T cells, which are permissive to infection by all three viruses. Cells expressing individual proteins were harvested, affinity purifications performed in 96-well format, and protein mass spectrometry was utilized to identify physical interaction partners of each viral protein. Drug treatments, RNAi knockdowns and CRISPR/Cas9 knockouts were tested for SARS-CoV-2 viral phenotypes in Vero, Caco2 or A549-ACE2 cells. Results: We report 389 high-confidence interactors of SARS-CoV-2, 366 interactions for SARS-CoV-1, and 296 interactions for MERS-CoV. Among the SARS-CoV-2 interactors, we identified at least 66 druggable human proteins or host factors, and screening small molecules targeting these pathways using multiple viral assays have identified at least four sets of pharmacological agents that demonstrate antiviral activity against SARS-CoV-2. Comparison of the host-pathogen interactomes of SARS-CoV-2 with the other highly pathogenic coronaviruses SARS-CoV-1 and MERS highlights shared host interactions which may represent pan-coronavirus drug targets. Conclusion: We successfully utilized systematic protein interaction mapping to identify drug targets for SARS-CoV-2, leading to several Covid-19 clinical studies investigating the efficacy of drugs perturbing these pathways. Furthermore, comparative proteomics of the related coronaviruses SARS-CoV-1 and MERS-CoV identified shared host interactions which may represent pan-coronavirus drug targets. For a full list of contributing authors see: Gordon, D. E. et al. Nature 583, 459-468 (2020);Gordon, D. E. et al. Science 370 (2020).
ABSTRACT
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of the coronavirus disease2019 (COVID-19) pandemic, has infected millions and killed hundreds of thousands of people worldwide. Cliniciansare fighting the battle against the virus with a limited arsenal of drugs that have been shown to be safe andefficacious in treating SARS-CoV-2 infection. We thus rely on gaining a deeper understanding of the molecularmechanisms of this virus, and desperately need new strategies to get drugs to patients in need quickly. To combineboth of these aims, my lab has focused much of our recent work on applying systems biology approaches to identifythe cellular pathways hijacked by SARS-CoV-2, with the aim of pinpointing promising clinically available drugs forrapid repurposing to treat COVID-19. Identifying which cellular pathways and mechanisms are affected acrossdifferent diseases will open the existing toolbox of drug treatment for COVID-19. Cancer drugs are a particularlypromising set of drugs in this regard: During the process of virus replication, viruses rely on a multitude ofinteractions with their host cell, and hijack similar pathways that are affected in cancer cells. For example, virusesmanipulate the cell cycle for their own benefit, recruit host DNA-damage machinery to viral replication sites, rely onthe host translation machinery for viral protein production, and interfere with several signaling pathways, forexample, to suppress cellular antiviral defenses. This list, while not comprehensive, makes apparent thecommonalities between the exploitation of molecular mechanisms by cancer and virus infection. To get a full pictureof cellular pathways targeted by SARS-CoV-2, we recently generated two virus-host interaction networks usingsystems biology approaches. First, we used affinity-purification mass spectrometry to create a virus-host protein-protein interaction (PPI) map. We cloned, tagged, and expressed 26 of the 29 SARS-CoV-2 proteins in human cellsand identified the human proteins and complexes that physically associate with the individual viral proteins. Our maprevealed 332 high-confidence PPIs with human proteins involved in a wide spectrum of cell biology, highlightingseveral oncogenic pathways. We identified 66 druggable human proteins at the virus-host interface, targeted by 69compounds (of which 29 drugs are approved by the US Food and Drug Administration, 12 are in clinical trials, and28 are preclinical compounds). We screened a subset of these in multiple viral assays and found that, for example, inhibitors of mRNA translation displayed antiviral activity. In a separate study building on these results, we created aquantitative mass spectrometry-based phosphoproteomics survey of SARS-CoV-2 infection in cell culture. Focusingon phosphorylation events, which are also highly misregulated in cancer, showed stark changes for both host andviral proteins and revealed dramatic rewiring of a number of signaling pathways. For example, SARS-CoV 2infection promoted casein kinase II (CK2) and p38 MAP kinase activation and shutdown of mitotic kinases. Weidentified 87 drugs and compounds by mapping global phosphorylation profiles to dysregulated kinases andpathways. Pharmacologic inhibition of p38, CK2, CDKs, AXL, and PIKFYVE kinases possessed antiviral efficacy, representing potential COVID-19 therapies. Clinical trials with some drugs and compounds implicated by our studiesare currently being discussed or are already under way. To prepare for future outbreaks, we need to further increaseour knowledge of cellular pathways targeted during virus infection. To this end, in addition to a number of infectiousagents my lab has studied previously, we are currently performing similar studies on the closely relatedcoronaviruses responsible for outbreaks of SARS and Middle East respiratory syndrome (MERS) in 2003 and 2012, respectively. Combined with the results discussed above, we hope to position the health care community tosuccessfully fight this and potential future outbreaks of novel infectious diseas s. Importantly, the input of physiciansand experts from other fields will be crucial to leverage the knowledge gained from our studies about the pathwaystargeted across diseases.