SARS-CoV-2 Spike Protein Stimulates Macropinocytosis in Murine and Human Macrophages via PKC-NADPH Oxidase Signaling

Abstract

Coronavirus disease 2019 (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). While recent studies have demonstrated that SARS-CoV-2 may enter kidney and colon epithelial cells by inducing receptor-independent macropinocytosis, it remains unknown whether this process also occurs in cell types directly relevant to SARS-CoV-2-associated lung pneumonia, such as alveolar epithelial cells and macrophages. The goal of our study was to investigate the ability of SARS-CoV-2 spike protein subunits to stimulate macropinocytosis in human alveolar epithelial cells and primary human and murine macrophages. Flow cytometry analysis of fluid-phase marker internalization demonstrated that SARS-CoV-2 spike protein subunits S1, the receptor-binding domain (RBD) of S1, and S2 stimulate macropinocytosis in both human and murine macrophages in an angiotensin-converting enzyme 2 (ACE2)-independent manner. Pharmacological and genetic inhibition of macropinocytosis substantially decreased spike-protein-induced fluid-phase marker internalization in macrophages both in vitro and in vivo. High-resolution scanning electron microscopy (SEM) imaging confirmed that spike protein subunits promote the formation of membrane ruffles on the dorsal surface of macrophages. Mechanistic studies demonstrated that SARS-CoV-2 spike protein stimulated macropinocytosis via NADPH oxidase 2 (Nox2)-derived reactive oxygen species (ROS) generation. In addition, inhibition of protein kinase C (PKC) and phosphoinositide 3-kinase (PI3K) in macrophages blocked SARS-CoV-2 spike-protein-induced macropinocytosis. To our knowledge, these results demonstrate for the first time that SARS-CoV-2 spike protein subunits stimulate macropinocytosis in macrophages. These results may contribute to a better understanding of SARS-CoV-2 infection and COVID-19 pathogenesis.

Keywords: SARS-CoV-2; epithelial cell; macrophage; macropinocytosis.

Figure 1

Recombinant SARS-CoV-2 spike proteins stimulate fluid-phase macropinocytosis in murine bone marrow-derived macrophages. (A) Flow cytometry gating strategy. (B) Murine bone marrow-derived macrophages were incubated with FITC-dextran (100 µg/mL) and treated with vehicle (PBS) or with 1 µg/mL of the spike protein subunits S1, RBD, or S2 for 4 h (n = 3). (C) Cells were treated with vehicle or PMA (1 µM, 4 h) in the presence of FITC-dextran (n = 9). Internalization of FITC-dextran was quantified via FACS (ex. 493 nm, em. 518 nm). Cells were pretreated with EIPA (25 µM) prior to incubation with 1 µg/mL of S1 (D), RBD (E), or S2 (F) (n = 4). (G) Murine bone marrow-derived macrophages were pretreated with polymyxin b (10 µg/mL, 30 min preincubation) and stimulated with vehicle (PBS) or PMA (1 µM, 4 h) in the presence of TRITC-dextran (100 µg/mL). (H) Cells were pretreated with polymyxin b (1–10 µg/mL, 30 min preincubation) and stimulated with S1 subunit (1 µg/mL, 4 h) in the presence of TRITC-dextran (100 µg/mL). Internalization of FITC-dextran was quantified via FACS (ex. 550 nm, em. 577 nm). (I) Agarose gel showing representative genotyping experiments. (J–L) Bone marrow macrophages from NHE1^f/f and NHE1^ΔM mice were treated with vehicle, S1, RBD, or S2 spike protein subunits (1 µg/mL, 4 h) in the presence of FITC-dextran (100 µg/mL) (n = 6). Data are presented as means ± SD. ns = not significant. * p < 0.05; ** p < 0.005; *** p < 0.001; **** p < 0.0001. P values were calculated using t-test (C), oneway (B, D–H) or two-way (J–L) ANOVA with Tukey’s test for multiple comparisons.

Figure 2

Recombinant SARS-CoV-2 spike proteins stimulate fluid-phase macropinocytosis in human macrophages. Primary human PBMC-derived macrophages were incubated with FITC-dextran (100 µg/mL) and treated with vehicle (PBS) or spike protein subunits S1 (A), RBD (B), and S2 (C) (1 µg/mL, 4 h) ± EIPA (25 µM, 30 min preincubation). FITC-dextran internalization was analyzed via FACS. (D) Quantification of the number of ruffles per cell in vehicle-, S1-, and S1+EIPA-treated THP1 macrophages normalized to total cell number. (E) THP1 macrophages were treated with spike protein subunits (1 µg/mL, 30 min) ± EIPA (25 µM, 30 min preincubation) and processed for SEM. Analysis of SEM images demonstrated emerging membrane projections (orange arrows), fully formed dorsal ruffles (green arrows), and circularized membrane protrusions forming cups (red arrow). Data are presented as means ± SD. ** p < 0.005; **** p < 0.0001. P values were calculated using one way ANOVA with Tukey’s test for multiple comparisons. Scale bars are 5 µm.

Figure 3

Recombinant SARS-CoV-2 spike proteins do not stimulate fluid-phase macropinocytosis in human alveolar epithelial cells. Cells were incubated with FITC-dextran (100 µg/mL) and treated with vehicle (PBS) or with different concentrations (0.001 µg/mL–10 µg/mL) of the spike protein subunits S1 (A), RBD (B), and S2 (C) for 4 h (n = 4). Epithelial cells were also treated with the chemical macropinocytosis stimulator PMA (1 µM). (D) Cells were treated with EGF (0.5 nM, 4 h) ± EIPA (25 µM, 30 min pretreatment) in the presence of FITC-dextran (100 µg/mL) (n = 10). Internalization of FITC-dextran was quantified via FACS (ex. 493 nm, em. 518 nm). (E) Representative SEM images of epithelial cells treated with spike protein subunit S1, RBD, and S2 (1 and 10 µg/mL, 30 min) (n = 3). Data are presented as means ± SD. * p < 0.05; ** p < 0.005. P values were calculated using one way ANOVA with Tukey’s test for multiple comparisons. Scale bars are 10 µm.

Figure 4

Pharmacological blockade of PKC, PI3K, and NOX2 inhibits SARS-CoV-2 spike protein-induced macropinocytosis in macrophages. (A) Murine bone marrow-derived macrophages were incubated with FITC-dextran (100 µg/mL) and treated with vehicle (PBS) or S1 (1 µg/mL, 4 h) ± LY394004 (10 µm, 30 min pretreatment) or Calphostin c (1 µm, 30 min pretreatment). Internalization of FITC-dextran was quantified via FACS (ex. 493 nm, em. 518 nm) (n = 3). (B,C) Representative histograms showing inhibitory effects of LY294004 (B) and Calphostin c (C). (D) Quantification of intracellular reactive oxygen species in cells treated with S1 (1 µg/mL, 1 h) using H₂DCFDA (5 µM, 30 min) (n = 3). (E) Human primary PBMC-derived macrophages were pretreated with LL37 (5 µg/mL and 10 µg/mL, 30 min) and treated with S1 (1 µg/mL, 4 h). Internalization of FITC-dextran was quantified via FACS. (F) Cells were pretreated with GSK2795039 (20 µg/mL, 30 min), or DPI (5 µM, 30 min) and treated with S1 (1 µg/mL, 4 h) in the presence of TRITC-dextran (100 µg/mL). Internalization of TRITC-dextran was quantified via FACS (ex. 550 nm, em. 577 nm). Data are presented as means ± SD. ns = not significant. * p < 0.05; *** p < 0.001; **** p < 0.0001. P values were calculated using t-test or one-way ANOVA with Tukey’s test for multiple comparisons.

Figure 5

Recombinant SARS-CoV-2 spike protein stimulates macrophage macropinocytosis in vivo. (A) Schematic flow chart of experimental design. Representative FITC-dextran histograms indicating in vivo stimulation of macropinocytosis with S1 (B) and genetic inhibition of S1-induced macropinocytosis (C). (D) Bar graph indicates FITC-dextran MFI fold change in peritoneal macrophages of vehicle- or S1 (20 µg, 4 h)-treated NHE1^f/f and NHE1^ΔM mice (n = 4–7). Data are presented as means ± SD. ns = not significant. * p < 0.05; ** p < 0.005. P values were calculated using two-way ANOVA with Tukey’s test for multiple comparisons.