The SARS-CoV-2 RNA interactome

Abstract

SARS-CoV-2 is an RNA virus whose success as a pathogen relies on its abilities to repurpose host RNA-binding proteins (RBPs) and to evade antiviral RBPs. To uncover the SARS-CoV-2 RNA interactome, we here develop a robust ribonucleoprotein (RNP) capture protocol and identify 109 host factors that directly bind to SARS-CoV-2 RNAs. Applying RNP capture on another coronavirus, HCoV-OC43, revealed evolutionarily conserved interactions between coronaviral RNAs and host proteins. Transcriptome analyses and knockdown experiments delineated 17 antiviral RBPs, including ZC3HAV1, TRIM25, PARP12, and SHFL, and 8 proviral RBPs, such as EIF3D and CSDE1, which are responsible for co-opting multiple steps of the mRNA life cycle. This also led to the identification of LARP1, a downstream target of the mTOR signaling pathway, as an antiviral host factor that interacts with the SARS-CoV-2 RNAs. Overall, this study provides a comprehensive list of RBPs regulating coronaviral replication and opens new avenues for therapeutic interventions.

Keywords: COVID-19; HCoV-OC43; LARP1; RNA; RNA interactome capture; RNA-binding proteins; SARS-CoV-2; coronavirus; mass spectrometry; virus.

Graphical abstract

Figure 1

Comprehensive identification host and viral proteins that directly interact with the SARS-CoV-2 RNAs (A) Schematic of the modified RAP-MS protocol in SARS-CoV-2-infected Vero cells. (B) Schematic of two separate pools of 90 nt antisense oligonucleotides and their SARS-CoV-2 RNA coverage. The first probe set, “probe I,” consists of 707 oligonucleotides that cover the unique region of gRNA, and the second probe set, “probe II,” consists of 275 oligonucleotides that cover the common region of gRNA and sgRNAs. (C) Spectral count ratio of probe I (x axis) and probe II (y axis) experiments over no-probe control in SARS-CoV-2-infected Vero cells (n = 3 technical replicates). Host proteins are marked by gray circles, and viral proteins (n = 9) are marked and labeled in black. The mean spectral count ratio of probe I and of probe II experiments are marked by vertical and horizontal dashed lines, respectively. (D) Statistical analysis of the quantity of viral proteins over no-probe control (i.e., probe binding). Adjusted (adj.) p values of probe I experiments and of probe II experiments are shown in yellow and green, respectively. Threshold for statistical significance (adj. p value < 0.01) is indicated by horizontal dashed lines. (E) Spectral count ratio of probe I (x axis) and probe II (y axis) experiments in SARS-CoV-2-infected Vero cells compared with RNP capture experiments in uninfected cells (n = 3 technical replicates). Statistically significant host proteins (n = 37, adj. p value < 0.05) in both probe I and probe II experiments are marked by black circles. Of those, representative host proteins are labeled. The mean spectral count ratio of probe I and of probe II experiments are marked by vertical and horizontal dashed lines, respectively. (F) Statistical analysis of host proteins enriched in both probe I and probe II experiment (i.e., probe enriched). Adjusted (adj.) p values of probe I experiments and of probe II experiments are shown in yellow and green, respectively. The Venn diagrams represent the number of identified proteins common and distinct between probe I and probe II experiments. (G and H) Statistical analysis of host proteins enriched (G) in only probe I experiment and (H) in only probe II experiment.

Figure 2

Comparison of SARS-CoV-2 and HCoV-OC43 RNA interactome (A) Schematic of time course RNP capture experiment in HCoV-OC43-infected HCT-8 cells. (B) LFQ intensity of abundant viral proteins identified in HCoV-OC43 RNP capture experiment at 12, 24, 36, and 48 h post-infection (hpi). (C) Heatmap of normalized LFQ intensity (nLFQ) of HCoV-OC43 experiment of host proteins enriched in (left) both SARS-CoV-2 probe I and probe II experiments, (middle) only SARS-CoV-2 probe I experiment, and (right) only SARS-CoV-2 probe II experiment. nLFQ is the log₁₀ fold change over the median LFQ intensity across each probe set (i.e., probe I or probe II). A pseudo-value of 1 × 10⁶ was added to handle missing values. (D) Spectral count ratio of probe I (x axis) and probe II (y axis) experiments over no-probe control in HCoV-OC43-infected HCT-8 cells of 36 hpi. Host proteins are marked by gray circles, and viral proteins (n = 14) are marked and labeled in black. The mean spectral count ratio of probe I and of probe II experiments are marked by vertical and horizontal dashed lines, respectively. (E) Spectral count ratio of probe I (x axis) and probe II (y axis) experiments in HCoV-OC43-infected HCT-8 cells compared with uninfected cells of 36 hpi. Statistically significant host proteins (n = 38, adj. p value < 0.05) in both probe I and probe II experiments are marked by black circles. Of those, representative host proteins are labeled. The mean spectral count ratio of probe I and of probe II experiments are marked by vertical and horizontal dashed lines, respectively. (F) Upset plot of HCoV-OC43 probe I-enriched (n = 67), HCoV-OC43 probe II-enriched (n = 70), SARS-CoV-2 probe I-enriched (n = 73), and SARS-CoV-2 probe II-enriched (n = 72) proteins. Fourteen host proteins were enriched in all four experiments. The Venn diagrams represent the number of identified proteins common and distinct between the SARS-CoV-2 and OC43 RNA interactome.

Figure 3

Molecular network of the SARS-CoV-2 RNA interactome (A) Physical interaction network of SARS-CoV-2 RNA interactome but excluding ribosomal proteins and EIF3 proteins. Host factors are marked in red. Host and viral proteins are marked by circles and diamonds, respectively. The size of the network node is proportional to the number of network edges (i.e., degree) of the node. Largest connected component of the network was chosen for visualization and interpretation. (B) Gene Ontology biological process term enrichment analysis of proteins retrieved in (A). Threshold for statistical significance (adj. p value < 0.01) is indicated by a vertical dashed line. (C) Subcellular localization of the SARS-CoV-2 RNA interactome and cellular mRNA interactomes. The number of proteins with localization information is shown in parentheses.

Figure 4

Regulation of the SARS-CoV-2 RNA interactome (A) Differential expression analysis of ACE2-expressing A549 cells (A549-ACE2) after SARS-CoV-2 infection (MOI = 2.0). mRNA fold changes (FCs) of −1 and 1 are marked by vertical dashed lines. FDR of 1% is marked by a horizontal dashed line. Host genes of the SARS-CoV-2 RNA interactome are marked in red if differentially expressed (FDR < 1% and |FC| > 1) and otherwise in black. (B) Comparison of mRNA fold change of SARS-CoV-2-infected (MOI = 2.0) A549-ACE2 cells after treatment of ruxolitinib. Host genes of the SARS-CoV-2 RNA interactome are marked in black. (C) Differential expression analysis of normal human bronchial epithelial (NHBE) cells after IFN-β treatment. Graph notations as in (D). (D) MA plot of published mRNA-seq data of post-mortem lung samples from COVID-19 patients and healthy lung tissue from uninfected individuals. Mean fold change is marked by a horizontal dashed line. Host genes of the SARS-CoV-2 RNA interactome are marked in red if differentially expressed (FDR < 5%) and otherwise in black. Other differentially expressed genes are in dark gray.

Figure 5

Effects of RBP depletion on viral RNA abundance (A) Schematic of small interfering RNA (siRNA) knockdown in SARS-CoV-2-infected Calu-3 cells. (B) Change in gRNA (amplicon nsp12, dark gray) and N sgRNA (amplicon N, light gray) levels after siRNA depletion were measured using qRT-PCR). Data are represented as mean ± SEM (n = 3 or 6 independent experiments). RNA levels in siNC-treated cells were used as negative control and marked by a horizontal dashed line. 18S rRNA was used for normalization. ^∗p < 0.05 and ^∗∗p < 0.01, two-sided Student’s t test. (C) Relative mRNA levels of β-interferon and ISGs after depletion of antiviral RBPs in SARS-CoV-2-infected Calu-3 cells measured using qRT-PCR. Data are represented as mean ± SEM (n = 3 independent experiments). RNA levels in siNC-treated cells were used as negative control. 18S rRNA was used for normalization. ^∗p < 0.05 and ^∗∗p < 0.01, two-sided Student’s t test. (D) Relative viral and host RNA levels after ectopic expression of respective FLAG-tagged plasmids (x axis) in SARS-CoV-2-infected HEK293T cells were measured using qRT-PCR. Proteins were expressed 24 h followed by virus infection for another 24 h. Data are represented as mean ± SEM (n = 3 independent experiments). RNA levels are shown relative to that in HEK293T cells transfected with FLAG-tagged GFP plasmids. 28S rRNA was used for total RNA normalization. ^∗p < 0.05, one-sided Student’s t test.