Integrin αvβ1 facilitates ACE2-mediated entry of SARS-CoV-2

Abstract

Integrins have been suggested to be involved in SARS-CoV-2 infection, but the underlying mechanisms remain largely unclear. This study aimed to investigate how integrins facilitate the ACE2-mediated cellular entry of SARS-CoV-2. We first tested the susceptibility of a panel of human cell lines to SARS-CoV-2 infection using the spike protein pseudotyped virus assay and examined the expression levels of integrins in these cell lines by qPCR, western blot and flow cytometry. We found that integrin αvβ1 was highly enriched in the SARS-CoV-2 susceptible cell lines. Additional studies demonstrated that RGD (403-405)→AAA mutant was defective in binding to integrin αvβ1 compared to its wild type counterpart, and anti-αvβ1 integrin antibodies significantly inhibited the entry of SARS-CoV-2 into the cells. Further studies using mouse NIH3T3 cells expressing human ACE2, integrin αv, integrin β1, and/or integrin αvβ1 suggest that integrin αvβ1 was unable to function as an independent receptor but could significantly facilitate the cellular entry of SASR-CoV-2. Finally, we observed that the Omicron exhibited a significant increase in the ACE2-mediated viral entry. Our findings may enhance our understanding of the pathogenesis of SARS-CoV-2 infection and offer potential therapeutic target for COVID-19.

Keywords: ACE2; COVID-19; Integrin αvβ1; SARS-CoV-2; Viral entry.

Fig. 1

The high expression of integrin αvβ1 in the SARS-CoV-2 susceptible cell lines. A. Entry of SARS-CoV-2 S pseudotyped virions into human cell lines SW620, IOSE-80, MDA-MB-231, HepG2, PANC-1, HGC-27, BEAS-2B, Calu-3, 293T, and 293T/hACE2). Cells were infected with SARS-CoV-2 S-pseudotyped (red) and VSV-pseudotyped viruses, respectively, or mock infection (white). At 72 h post infection, viral entry efficiency was measured by luciferase activity assay. Significant difference from mock were determined by two-tailed unpaired t-test. ***P < 0.001. Error bars indicate SD (n = 3). B. A schematic diagram showing the RGD- and non-RGD binding integrins. C. Heatmap of the mRNA expression profile of RGD-binding integrins in different human organs obtained from the GEPIA2 database. The mRNA expression was exhibited by Log2(TPM+1) scale, and TPM denotes Transcripts Per Million. D. Heatmap of the mRNA expression profile of integrin ITGA3, A5, AV, B1, B5, and ACE2 in indicated cell lines detected by RT-qPCR. The mRNA expression levels were normalized to the level of GAPDH. E. The protein expression level of integrin ITGAV, B1, and ACE2 in indicated cell lines was analyzed by western blot with anti-integrin av, β1, and ACE2 antibodies, respectively. F. The expression levels of ITGAV and ITGB1 on the cell membrane were analyzed by flow cytometry. Histograms indicate the expression of integrins. Red: anti-ACE2 Ab, pink: anti-integrin av Ab, blue: anti-integrin β1 Ab, gray: isotype-control Ab.

Fig. 2

The interaction between RGD motif on SARS-CoV-2 and integrin αvβ1. A. Three susceptible cell lines (Calu-3, 293T and 293T/hACE2) were infected with SARS-CoV-2 wild type (WT) and the RGD (403–405)→AAA mutant S pseudotyped viruses, respectively. The luciferase activity (LUC) was measured at 72 h post infection. The VSV-pseudotyped viruses were used as a positive control whereas the mock infection served as a negative control. Significant difference from wild-type were determined by two-tailed unpaired t-test. ***P < 0.001. Error bars indicate SD (n = 3). B. Detection of interaction between the RGD motif on RBD of SARS-CoV-2 spike and integrin αvβ1 by immunoprecipitation (IP). 293T cells were transfected with integrin av, integrin β1, Flag-tagged WT RBD, and Flag-tagged RGD (403–405)→AAA mutant RBD, respectively. The cell lysates for IP were prepared at 48 h post transfection. RGD (403–405)→AAA mutant is defective in binding to integrin αvβ1 compared to its wild type counterpart. C. Detection of WT S or the RGD(403–405)→AAA mutant S proteins in the pseudotyped virions by western blot with antibodies against SARS-CoV-2 spike protein. D. Structural alignment of WT SARS-CoV-2 S and the RGD (403–405)→AAA variant. (I) Mutant region on spike protein. (II) Enlarged view of structural alignment of mutant region (blue: wild type; red: RGD (403–405)→AAA variant). E. MM/PBSA binding free energy of SARS-CoV-2 RBD-hACE2 complex and RGD variant RBD-hACE2 complex for 50 ns.

Fig. 3

Anti-αvβ1 integrin antibodies inhibited the SARS-CoV-2 entry into the susceptible cell lines. A-C. Inhibition of SARS-CoV-2 S pseudotyped virus entry by anti-αvβ1 integrin antibodies in three susceptible cell lines, Calu-3 (A), 293T (B), and 293T/hACE2 (C). Cells were infected with SARS-CoV-2 S-pseudotyped (red) and VSV-pseudotyped viruses in the presence or absence of respective antibodies at various concentrations (0, 1, 10 μg/mL), respectively. Anti-ACE2 antibody was used as a positive control whereas the isotype-matched IgG served as a negative control. The viral entry efficiency was measured by luciferase activity assay at 72 h post infection. Significant difference between the untreated groups and the 10 μg/mL antibody treated groups were determined by two-tailed unpaired t-test. ns: not significant, p > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001. Error bars indicate SD (n = 3).

Fig. 4

Integrin αvβ1 facilitates SARS-CoV-2 virus entry into cells A. The 293T cells stably expressing human ACE2, integrin αv, integrin β1, or integrin αvβ1 were infected with mock (white), SARS-CoV-2 S-pseudotyped (red), and VSV-pseudotyped viruses (blue), respectively. B. The 293T cells stably expressing human ACE2 (293T/hACE2) were transiently transfected with integrin αv, integrin β1, or integrin αvβ1, followed by infection of mock (white), SARS-CoV-2 S-pseudotyped viruses (red), and VSV-pseudotyped viruses (blue), respectively. C. The NIH3T3 cells stably expressing human ACE2, integrin αv, integrin β1, or integrin αvβ1 were infected with mock (white), SARS-CoV-2 S-pseudotyped (red), and VSV-pseudotyped viruses (blue), respectively. D. The NIH3T3 cells stably expressing human ACE2 (NIH3T3/hACE2) were transiently transfected with integrin αv, integrin β1, or integrin αvβ1, followed by infection of mock (white), SARS-CoV-2 S-pseudotyped viruses (red), and VSV-pseudotyped viruses (blue), respectively. The viral entry efficiency was measured by luciferase activity assay at 72 h post infection. Significant difference between the groups were determined by two-tailed unpaired t-test. ns: not significant, p > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001. Error bars indicate SD (n = 3).

Fig. 5

The effect of integrin αvβ1 on the viral entry of different SARS-CoV-2 variants. NIH3T3, NIH3T3/hACE2 and NIH3T3/hACE2-ITGAVB1 cells were infected with different pseudotyped SARS-CoV-2 variants (wild-type, Gamma, Delta and Omicron), respectively. The viral entry efficiency was measured by luciferase activity assay at 72 h post infection. Significant difference between the groups were determined by two-tailed unpaired t-test. *P < 0.05; Error bars indicate SD (n = 3).

Fig. 6

A proposed model illustrating the action and mechanism of integrin αvβ1 that facilitates ACE2-mediated viral entry of SARS-CoV-2 A. ACE2 functions as an independent receptor for SARS-CoV-2 entry into the host cells; B. Integrin αvβ1 is unable to serve as an independent receptor for SARS-CoV-2 entry into the host cells; C. Integrin αvβ1 may significantly enhance the ACE2-mediated viral entry of SARS-CoV-2.