Nucleolin mediates SARS-CoV-2 replication and viral-induced apoptosis of host cells

Abstract

Host-oriented antiviral therapeutics are promising treatment options to combat COVID-19 and its emerging variants. However, relatively little is known about the cellular proteins hijacked by SARS-CoV-2 for its replication. Here we show that SARS-CoV-2 induces expression and cytoplasmic translocation of the nucleolar protein, nucleolin (NCL). NCL interacts with SARS-CoV-2 viral proteins and co-localizes with N-protein in the nucleolus and in stress granules. Knockdown of NCL decreases the stress granule component G3BP1, viral replication and improved survival of infected host cells. NCL mediates viral-induced apoptosis and stress response via p53. SARS-CoV-2 increases NCL expression and nucleolar size and number in lungs of infected hamsters. Inhibition of NCL with the aptamer AS-1411 decreases viral replication and apoptosis of infected cells. These results suggest nucleolin as a suitable target for anti-COVID therapies.

Keywords: Aptamer; COVID; Nucleolin; Nucleoprotein; SARS-CoV-2.

Mechanisms of nucleolin-mediated SARS-CoV-2 infection: 1. SARS-CoV-2 induces nucleolar to cytoplasmic translocation of nucleolin; In the cytoplasm, nucleolin interacts with the viral nucleoprotein and host G3BP1 in stress granules and increases viral proliferation via a direct increase in translation of replication proteins such as G3BP1 (2), nucleoprotein and RdRp (3); Nucleolin mediates viral-induction of apoptosis (4) and cell cycle arrest (5) via stimulation of p53 and Rb.

Fig. 1

SARS-CoV-2 viral proteins interact with nucleolin and induce its cytoplasmic translocation. A. Immunofluorescence staining and confocal images of Vero E6 cells transfected with spike-RBD and N-protein for 48h, with anti-nucleolin (NCL, green or red), anti-spike-RBD (red), N-protein (green), TIAR (red), and Hoechst (blue). The merge of the fluorescent channels, including higher magnification, are shown (right). White arrows indicate co-localization. B. Western blot determination of NCL and SARS-CoV-2 proteins in the nuclear and cytoplasmic fractions of Vero E6 cells transfected with control (ctrl) empty plasmid (pcDNA3.1⁺), spike-RBD, N-protein and Nsp1 for 48h. Rb and GAPDH were used as nuclear and cytoplasmic markers, respectively. Exposure time for NCL development were 1 s (1″) and 2 min (2′). Lysates from Vero E6 cells transfected with spike-RBD, total N-protein, and control (pcDNA3.1⁺) (C) and C- and N-terminal domains (conjugated to a histidine-tag), Nsp1 and Nsp12/RdRp (D) were co-immunoprecipitated with NCL. Western blot determination of NCL and SARS-CoV-2 proteins, using total protein lysate (input) or immunoprecipitated complexes (IP). β-actin: loading control. Fluorescence correlation spectroscopy (FCS) auto-correlation plot for Cy5 labeled NCL alone or in complex with SARS-CoV-2 heat-inactivated virion (E) and percentage of NCL-Cy5 binding to virion (F). Average values from triplicate measurements are shown. Error bars indicate standard deviations. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 2

SARS-CoV-2 increases nucleolin levels and translocation to cytoplasm. A. Western blot determination, in triplicate, and ImageJ quantification of NCL levels in Vero E6-TMPRSS and Caco-2 cells, transduced with empty Lenticrispr V2 containing CAS9 vector control and 2 different NCL-knockdown lines (KD1 and KD2), and infected with SARS-CoV-2 for 48h (MOI 0.01). N-protein was used to detect the presence of virus. β-actin: loading control. B. Confocal immunofluorescence of NCL (red) and N-protein (green) in Vero E6-TMPRSS vector and NCL-KD following viral infection (as above). Hoechst (blue) and the fluorescence merge are shown. Yellow, blue and white arrows indicate the SG-like structures, NCL-proficient and -deficient cells, respectively. Western blot determination of NCL, N-protein and spike-RBD in nuclear and cytoplasmic fractions (C) and G3BP1 in total lysate (D) of Vero E6-TMPRSS control and NCL-KD cells infected for 48 h with mock and SARS-CoV-2 (MOI 0.01). Rb and GAPDH were used as nuclear and cytoplasmic markers, respectively. The student t-test was performed. E. Confocal immunofluorescence of G3BP1 (red) and NCL (green) in Vero E6-TMPRSS cells following viral infection (as above). Hoechst stain (blue) and the fluorescence merge are shown. Yellow arrow indicates G3BP1 and NCL co-localization. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 3

Nucleolin increases viral protein and replication. A. Western blot determination, in triplicate, of NCL and viral proteins in Caco-2 cells control and NCL-KD infected with mock and SARS-CoV-2 (MOI 0.01) for 48 h. β-actin: loading control. B. Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR), using primers against N-protein, to determine the viral titer in the supernatant of Vero E6 and Caco-2 control and NCL-KD cells at the indicated time points, following SARS-CoV-2 infection (MOI 0.001). Four independent infection experiments and qRT-PCR were performed. C. Cell titer glow assay, in triplicate, to determine cell viability at 48h of Vero E6 control and NCL-KD cells (1 × 10⁴ cells/well in 96 well plate) following viral infection (MOI 0.001 and 0.01). The viral inoculums were removed after 1 h of incubation in both replication and survival assays. Cell viability is represented as percentage relative to non-infected cells. The student t-test was performed. NS: not significant. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 4

Nucleolin mediates SARS-CoV-2 induction of the stress response and apoptosis. Western blot determination of NCL, apoptotic, stress response and DNA damage proteins in total cell lysate (A) and nuclear and cytoplasmic fractions (B) of Vero E6-TMPRSS cells, control and NCL-KD infected with mock and SARS-CoV-2 (MOI 0.01) for 48 h. β-actin: loading control. Rb and GAPDH were used as nuclear and cytoplasmic markers, respectively. C. Flow cytometric determination, in triplicate, and ImageJ quantification of apoptosis by staining of Vero E6-TMPRSS control and NCL-KD cells with annexin V-Alexa 488, propidium iodide (PI) and mitotracker conjugated with phycoerythrin (PE). The number of positive cells is shown (%). D. Western blot determination in triplicate and ImageJ quantification of NCL, N-protein and cytochrome C (Cyt c) in cytoplasmic and mitochondrial fractions of Vero E6-TMPRSS cells, control and NCL-KD, infected with mock and SARS-CoV-2 (as above). The ratio of Cyt c expression in cytoplasm (C) by mitochondria (M) is shown. E. Merged confocal immunofluorescence image of cytochrome c (Cyt c, green), mitochondrial marker Cox IV (red) and Hoechst (blue) in Vero E6-TMPRSS cells infected with SARS-CoV-2 (MOI 0.01) for 48 h. The student t-test was performed. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 5

Nucleolin inhibition decreases SARS-CoV-2 replication and apoptosis of host cells. A. Western blot determination of NCL, apoptosis, stress response, DNA damage and viral proteins in total cell lysate from Caco-2 cells and nuclear and cytoplasmic fraction from Vero E6-TMPRSS cells treated with CRO and AS-1411 (10 μM) for 24 h prior to SARS-CoV-2 (MOI 0.01) infection for 48 h. β-actin: loading control. Confocal immunofluorescence of mock- and viral-infected Vero E6-TMPRSS cells stained with anti-cleaved caspase 3 (CC3) (white) and Hoechst (blue) (B) and merged images of NCL (red), p53 (green) and Hoechst (blue) (C). White arrows indicate cytoplasmic p53. D. String interaction network of NCL, G3BP1 and p53 apoptotic pathway. E. Flow cytometry determination, in triplicate, of apoptosis by staining Vero E6-TMPRSS cells treated with vehicle, CRO and AS-1411 (10 μM) for 24 prior to SARS-CoV-2 infection (MOI 0.01) for 48 h. The percentage of annexin V, PI- and mitotracker-positive cells is shown. Cell viability assay of Vero E6 cells (F) and qRT-PCR detection of SARS-CoV-2 E-protein in the supernatant of Vero E6-TMPRSS and Caco-2 cells (G) treated with vehicle, remdesivir (5 μM), CRO and AS1411 (10 μM) 24 h prior to viral infection (MOI 0.001 for replication). Cell viability is represented as percentage relative to non-infected treated cells. Four independent infection experiments and qRT-PCR were performed. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 6

SARS-CoV-2 interacts with nucleolin and regulates its expression in lungs of infected hamsters. Representative consecutive images (A) and higher magnification (B) of immunohistochemistry (IHC) detection of NCL and Spike in lungs of hamsters inoculated with mock and SARS-CoV-2 (n = 5 each). Hematoxylin and eosin (HE) stain of tissue sections is shown. Blue and red arrows indicate NCL cytoplasmic and nuclear stain, respectively. C. Quantification of total NCL stain, nucleolar size and number using Halo software. The student t-test was performed. *P < 0.05, **P < 0.01 and ***P < 0.001.