Human Hematopoietic Stem, Progenitor, and Immune Cells Respond Ex Vivo to SARS-CoV-2 Spike Protein

Abstract

Despite evidence that SARS-CoV-2 infection is systemic in nature, there is little known about the effects that SARS-CoV-2 infection or exposure has on many host cell types, including primitive and mature hematopoietic cells. The hematopoietic system is responsible for giving rise to the very immune cells that defend against viral infection and is a source of hematopoietic stem cells (HSCs) and progenitor cells (HPCs) which are used for hematopoietic cell transplantation (HCT) to treat hematologic disorders, thus there is a strong need to understand how exposure to the virus may affect hematopoietic cell functions. We examined the expression of ACE2, to which SARS-CoV-2 Spike (S) protein binds to facilitate viral entry, in cord blood derived HSCs/HPCs and in peripheral blood derived immune cell subtypes. ACE2 is expressed in low numbers of immune cells, higher numbers of HPCs, and up to 65% of rigorously defined HSCs. We also examined effects of exposing HSCs/HPCs and immune cells to SARS-CoV-2 S protein ex vivo. HSCs and HPCs expand less effectively and have less functional colony forming capacity when grown with S protein, while peripheral blood monocytes upregulate CD14 expression and show distinct changes in size and granularity. That these effects are induced by recombinant S protein alone and not the infectious viral particle suggests that simple exposure to SARS-CoV-2 may impact HSCs/HPCs and immune cells via S protein interactions with the cells, regardless of whether they can be infected. These data have implications for immune response to SARS-CoV-2 and for HCT. Graphical Abstract • Human HSCs, HPCs, and immune cells express ACE2 on the cell surface, making them potentially susceptible to SARS-CoV-2 infection. • SARS-CoV-2 S protein, which binds to ACE2, induces defects in the colony forming capacity of human HPC and inhibits the expansion of HSC/HPC subpopulations ex vivo. These effects can be at least partially neutralized by treatment with SARS-CoV-2 targeting antibody, recombinant human ACE2, or Angiotensin1-7. • S protein also induces aberrant morphological changes in peripheral blood derived monocytes ex vivo. • Thus, there are many different manners in which SARS-CoV-2 virus may impact the functional hematopoietic system, which has important implications for hematological manifestations of COVID-19 (i.e. thrombocytopenia and lymphopenia), immune response, and hematopoietic stem cell transplant in the era of COVID-19.

Keywords: ACE2; COVID-19; Cord blood; Hematopoiesis; Hematopoietic cell expansion; Hematopoietic stem and progenitor cells; Immune cells; SARS-CoV-2; Spike protein.

Graphical Abstract

• Human HSCs, HPCs, and immune cells express ACE2 on the cell surface, making them potentially susceptible to SARS-CoV-2 infection. • SARS-CoV-2 S protein, which binds to ACE2, induces defects in the colony forming capacity of human HPC and inhibits the expansion of HSC/HPC subpopulations ex vivo. These effects can be at least partially neutralized by treatment with SARS-CoV-2 targeting antibody, recombinant human ACE2, or Angiotensin1–7. • S protein also induces aberrant morphological changes in peripheral blood derived monocytes ex vivo. • Thus, there are many different manners in which SARS-CoV-2 virus may impact the functional hematopoietic system, which has important implications for hematological manifestations of COVID-19 (i.e. thrombocytopenia and lymphopenia), immune response, and hematopoietic stem cell transplant in the era of COVID-19.

Fig. 1

Subpopulations of cord blood derived HSCs/HPCs express cell surface ACE2. (a/b) RT-qPCR to test for ACE2 mRNA expression (a) and SDS-PAGE followed by western blot with indicated antibodies to test for ACE2 protein expression (b) in CB lineage enriched (L = Lin+) cells; low density CB lineage depleted and CD34+ enriched cells (C = Lin-CD34+), and CB high density polymorphonuclear cells (H=PMN). ACE2 expression is shown relative to GAPDH expression. Matching numbers in labels indicate samples that came from the same cord blood unit. (c) Low density cord blood CD34+ enriched cells were stained with fluorochrome conjugated antibodies and analyzed with flow cytometry to define the indicated immunophenotypes and determine ACE2 expression on these subpopulations. ACE2+ gate was defined using rabbit IgG isotype control. Matched colors of points indicate the same cord blood unit. ACE2 staining n = 8, IgG control n = 3; stats: t-tests comparing FITC+ % in ACE2 staining versus IgG staining, corrected for multiple testing. *P < 0.05 (d) Histogram shows representative ACE2 staining for HSCs from two cord blood units with the highest and lowest number of ACE2+ cells. Also shown is rabbit IgG isotype control

Fig. 2

Cord blood HSCs/HPCs exhibit reduced colony forming capacity in the presence of SARS-CoV-2 S protein. (a) CD34+ enriched cells were plated at 100,000 cells/mL in media with stimulating growth factors (rhuTPO/rhuSCF/rhuFLT3L) and with 1 μg/mL recombinant S protein or PBS control and grown for 4 days in 5% O₂ and 5% CO₂ at 37 °C. Viable cells were counted using a hemocytometer and Trypan Blue viability stain. n = 5, stats: paired t-test, different symbols indicate the same cord blood unit in different treatment conditions. (b-c) CD34+ enriched cells were either taken for direct plating (unstimulated) or were grown for 24 h with stimulating growth factors (stimulated). 300 CD34+ enriched cells were plated in triplicate with the indicated doses of recombinant S protein or PBS control in 1% v/v methylcellulose with growth factors and serum and grown for 12 days in 5% O₂ and 5% CO₂ at 37 °C. CFU/1x10⁶cells were calculated. Stats: general linearized modeling including stimulated and unstimulated cells at varying doses in the same model followed by ANOVA with TukeyHSD post hoc tests. Significance codes indicate comparison of sample to 0 ng/mL (PBS) control in unstimulated cells. (d-e) 350 freshly harvested CD34+ cells were plated in triplicate in 1% v/v methylcellulose with serum and growth factors and with PBS control, 250 ng/mL recombinant S protein alone, 250 ng/mL S protein pre-incubated with 250 ng/mL SARS-CoV-2 neutralizing antibody (Antibody), or 250 ng/mL S protein with 250 ng/mL Angiotensin1–7 (Ang1–7) and grown for 12 days in 5% O₂ and 5% CO₂ at 37 °C. Total CFU/1x10⁶cells were calculated. Stats: general linearized modeling followed by ANOVA with TukeyHSD post hoc tests, matched symbols indicate the same cord blood unit in different treatment conditions. Shown are significance codes comparing all treatment levels to PBS control. *P < 0.05, ***P < 0.0005

Fig. 3

Cord blood HSCs/HPCs exhibit reduced expansion in the presence of SARS-CoV-2 S protein. (a-h) CD34+ enriched cells were plated at 100,000–200,000 cells/mL in media with stimulating growth factors and with PBS control, 1 μg/mL recombinant S protein alone, 1 μg/mL S protein pre-incubated with 1 μg/mL SARS-CoV-2 neutralizing antibody (Antibody), 1 μg/mL S protein pre-incubated with 1 μg/mL rhu ACE2, or 1 μg/mL S protein with 1 μg/mL Angiotensin1–7 (Ang1–7) and grown for 7 days in 5% O₂ and 5% CO₂ at 37 °C. (a-d). Cells were then analyzed by flow cytometry for the indicated cell populations and total cell numbers were calculated or (e-h) 350–500 CD34+ cells were plated in triplicate in 1% v/v methylcellulose with serum and growth factors and grown for 12 days in 5% O₂ and 5% CO₂ at 37 °C. Total CFU were calculated. n = 4/2 for (a-e)/(f-h), stats: generalized linear modelling followed by ANOVA with TukeyHSD post hoc tests, matched colors of points for 3a-e and matched symbols for points for 3f-h indicate the same cord blood unit in different treatment conditions. For (f-h) significance codes shown are for the comparison of the indicated treatment PBS control. *P < 0.05, **P < 0.005,***P < 0.0005

Fig. 4

Low numbers of peripheral blood derived immune cells express cell surface ACE2. (a) RT-qPCR to test for ACE2 mRNA expression in low density PB cells. ACE2 expression is shown relative to GAPDH expression. (b) Protein was harvested from low density PB cells and subjected to SDS-PAGE in non-reducing conditions followed by western blot with the indicated antibodies (IB = immunoblot). (c) Low density PB cells were stained with fluorochrome conjugated antibodies and analyzed with flow cytometry to define the indicated immunophenotypes and determine ACE2 expression on these subpopulations. ACE2+ gate was defined using rabbit IgG isotype control. Matched colors of points indicate the same peripheral blood donor. ACE2 staining n = 5, IgG control n = 5; stats: t-tests comparing FITC+ % in ACE2 staining versus IgG staining, corrected for multiple testing. *P < 0.05

Fig. 5

Peripheral blood cells respond ex vivo to exposure to SARS-CoV-2 S protein. (a-d) Low density PB was incubated for 2 h with 1 μg/mL SARS-CoV-2 S recombinant S protein or PBS control. (a) Representative histogram from 1 PB sample treated with S protein or PBS showing CD14 expression of forward and side scatter gated monocytes and (b) mean fluorescence intensities for CD14 staining. n = 4, stats: paired t-test, different colors of points indicate PBs from the same donor. (c) Representative contour plot from 1 PB sample treated with S protein or PBS showing CD14 expression of forward and side scatter gated monocytes, the gating strategy for CD14hi cells and (d) difference in CD14hi monocytes represented as a fold-change of S protein treated cells relative to PBS treated cells. n = 4, stats: paired t-test performed on total numbers of CD14hi monocytes. (e-g) Low density PB was incubated for 18 h with serum in the presence of 1 μg/mL SARS-CoV-2 S recombinant S protein or PBS control. (e) Representative contour plot from 1 PB sample showing all cells excluding debris and the monocyte gate used. (f) Mean values for forward scatter (FSC) and (g) side scatter (SSC) of monocytes from PB samples after incubation with S protein or PBS. n = 5, stats = paired t-test, different colors of points indicate PBs from the same donor. FACS = fluorescence activated flow cytometry. *P < 0.05, ***P < 0.0005