Epigenetic repression of antiviral genes by SARS-CoV-2 NSP1
Dimitrios G. Anastasakis, Daniel Benhalevy, Nicolas Çuburu, Nihal Altan-Bonnet, Markus Hafner
Abstract
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) evades the innate immune machinery through multiple viral proteins, including nonstructural protein 1 (NSP1). While NSP1 is known to suppress translation of host mRNAs, the mechanisms underlying its immune evasion properties remain elusive. By integrating RNA-seq, ribosome footprinting, and ChIP-seq in A549 cells we found that NSP1 predominantly represses transcription of immune-related genes by favoring Histone 3 Lysine 9 dimethylation (H3K9me2). G9a/GLP H3K9 methyltransferase inhibitor UNC0638 restored expression of antiviral genes and restricted SARS-CoV-2 replication. Our multi-omics study unravels an epigenetic mechanism underlying host immune evasion by SARS-CoV-2 NSP1. Elucidating the factors involved in this phenomenon, may have implications for understanding and treating viral infections and other immunomodulatory diseases.
Introduction
The recent COVID-19 pandemic caused by the severe acute respiratory syndrome coronavirus (SARS-CoV-2) further intensified the study of host-pathogen interactions and the mechanisms through which viruses suppress cellular antiviral responses [1–4]. SARS-CoV-2 evolved elaborate ways to evade the innate immune machinery through the independent action of several of the 29 proteins it encodes. Here, we focused on the multifunctional SARS-CoV-2 nonstructural protein 1 (NSP1), which may contribute to the exceptional pathogenesis of SARS-CoV in humans [5, 6]. Multiple molecular mechanisms for NSP1 function were proposed. Early studies showed that NSP1 selectively suppresses transcription of genes driven by various promotors simian virus 40 (SV40), cytomegalovirus (CMV), interferon (IFN)-b), without affecting actin and rRNA levels [7]. Other reports demonstrate direct suppression of cellular mRNAs by NSP1, either through endonucleolytic cleavage or binding to the 40S scanning ribosomal subunit causing a stall in the mRNA 5’ untranslated regions (UTR) and subsequent mRNA cleavage [4, 8–10].
Materials and methods
Cell culture
A549 cells (ATCC CCL-185) were cultured in DMEM medium (Gibco) supplemented with 10% (v/v) fetal bovine serum (FBS), 100 μg/ml zeocin, and 10 μg/ml blasticidin (Gibco). pEXP(FLAG/HA-NSP-WT) and pEXP(FLAG/HA-NSP K164A/H165A), (Addgene IDs 188781, and 188782, respectively) were generated using the Gateway system (Invitrogen) as described before [17]. For transient transfection plasmids were transfected into cells using Lipofectamine 3000 (Invitrogen) according to manufacturer’s instructions, expression quantification by immunofluorescence was performed as previously described [18]. Poly(I:C) was transfected with the plasmid at 1/10 of the plasmid (500 ng plasmid per 300,000 cells). RNA from transfected cells was isolated using the Direct-zol RNA Miniprep Kit (Zymo Research, Cat# R2050) according to the manufacturer’s instructions.
Results
Previous studies indicated that NSP1 globally suppresses host cell mRNA translation by stoichiometrically forming complexes with the mammalian translation machinery [2, 4, 7, 8, 10–12, 14, 31]. Nevertheless, we and others observed that transgenic expression of some NSP1 constructs resulted in cellular toxicity already at low expression levels [31–33], far from matching ribosome stoichiometry. It proved impossible to generate HEK293 and A549 stable cell lines expressing untagged NSP1 or NSP1 fused to a short, unfolded FLAG-HA tag (FH, 1.3 kDa). Even in transiently transfected cells using pFRT-TO-DEST plasmid as backbone the expressed FH-NSP1 transgene remained undetectable by standard Western blotting (Fig 1A). We were only able to detect NSP1 N-terminally fused to the large, globular GFP tag (~17 kDa), in the cytoplasm of A549 cells (consistent with the NSP1-ribosome interaction, Fig 1A and 1B). To confirm that our expression construct produced some levels of FH-NSP1, we therefore concentrated FH-NSP1 by FLAG immunoprecipitation (Fig 1A). We used the same approach to concentrate any FH-NSP1 from the growth media, which ruled out the possibility NSP1 is not detected in cell extracts due to secretion (Fig 1A).
Discussion
Our results indicate that like many other viral proteins, SARS-CoV-2 NSP1 is multifunctional. In addition to previously reported functions, including its well-documented role in translational repression [3–5, 7–16, 31, 45, 46], SARS-CoV-2 NSP1 may suppress host innate immune genes by epigenetic reprogramming. We propose that early in infection, while still at low copy number, NSP1 induces G9a-mediated H3K9 methylation of specific host gene loci, resulting in downregulated transcription of immune-related genes and reduced antiviral surveillance. Epigenetic silencing of antiviral genes early upon infection may cause the discrepancy between the levels of type I/III interferons and proinflammatory cytokines observed in COVID-19 patients. Compared to influenza patients, induction of both IFN-λ and type I IFNs is both impaired and delayed in patients with COVID-19 while pro-inflammatory cytokines are detected at similar levels [47]. This imbalance is the main reason SARS-CoV-2 can delay antiviral response and persist for a long period of time. Because severe symptoms are caused by high levels of pro-inflammatory cytokines and tissue damage, it is highly unlikely that inhibition of histone methylation can be used for therapy. However, aerosol-delivered H3K9 methyltransferase inhibitors could potentially have a beneficial effect prior or early after viral exposure and act as preventive drugs for frontline health-care workers combating an outbreak. Although UNC0638 has poor pharmacokinetic properties, the related UNC0642 has improved in vivo characteristics [48].
Conclusions
The fact that expression of a single viral protein alone induced a viral infection specific phenotype in the absence of viral infection provides confidence in our conclusions. It is unlikely that with the low copy numbers we have in our expression system, NSP1 is able to stoichiometrically interact and inhibit the ribosome. The observed changes in translational output measured by Ribo-seq and RNA-seq can be completely attributed to changes in mRNA levels. We do not contradict that global translational shutdown occurs at a later point of infection. Nevertheless, we propose that at the global scale immune-related genes are downregulated at the transcriptional level by NSP1. H3K9me2 marks are associated with heterochromatin and are involved in chromatin organization unlike transcription factors that can be directly involved in gene promoter activity [49, 50]. In this study, we show that NSP1 favors H3K9me2 marks either by inducing dimethylation or by preventing dynamic demethylation during antiviral response.
Acknowledgments
The authors thank Faiza Naz, Shamima Islam, and Dr. Stefania dell’Orso (NIAMS/NIH) for sequencing support and Parthena Konstantinidou (Haase Lab, NIDDK/NIH) for technical support and useful discussion, as well as Dr. Vittorio Sartorelli (NIAMS/NIH) for critical reading of the manuscript. The authors also thank David Eric Anderson (NIDDK NIH) for mass-spectrometry support, Jonathan Yewdell (NIAID/NIH) for providing stable ACE2 transfected A549 cells and Nikki Ostrenga for administrative support during the COVID-19 pandemic. The authors are grateful to Yolanda L. Jones, NIH Library Editing Services, for editing assistance.
Citation: Anastasakis DG, Benhalevy D, Çuburu N, Altan-Bonnet N, Hafner M (2024) Epigenetic repression of antiviral genes by SARS-CoV-2 NSP1. PLoS ONE 19(1): e0297262. https://doi.org/10.1371/journal.pone.0297262
Editor: Helene Minyi Liu, National Taiwan University, TAIWAN
Received: August 31, 2023; Accepted: January 2, 2024; Published: January 26, 2024
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All high-throughput data used in this study can be accessed via the Gene Expression Omnibus at https://www.ncbi.nlm.nih.gov/geo/ via accession number GSE208116.
Funding: National Institutes of Health, Intramural Research Program. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: he authors have declared that no competing interests exist.
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0297262#abstract0
