Skewed fate and hematopoiesis of CD34+ HSPCs in umbilical cord blood amid the COVID-19 pandemic

Abstract

Umbilical cord blood (UCB) is an irreplaceable source for hematopoietic stem progenitor cells (HSPCs). However, the effects of SARS-CoV-2 infection and COVID-19 vaccination on UCB phenotype, specifically the HSPCs therein, are currently unknown. We thus evaluated any effects of SARS-CoV-2 infection and/or COVID-19 vaccination from the mother on the fate and functionalities of HSPCs in the UCB. The numbers and frequencies of HSPCs in the UCB decreased significantly in donors with previous SARS-CoV-2 infection and more so with COVID-19 vaccination via the induction of apoptosis, likely mediated by IFN-γ-dependent pathways. Two independent hematopoiesis assays, a colony forming unit assay and a mouse humanization assay, revealed skewed hematopoiesis of HSPCs obtained from donors delivered from mothers with SARS-CoV-2 infection history. These results indicate that SARS-CoV-2 infection and COVID-19 vaccination impair the functionalities and survivability of HSPCs in the UCB, which would make unprecedented concerns on the future of HSPC-based therapies.

Keywords: Developmental biology; Haematology; Immunology; Pregnancy; Stem cells research.

Graphical abstract

Figure 1

Previous SARS-CoV-2 infection and/or vaccination significantly decreases CD34⁺cell frequencies and numbers per mL in UCB (A and B) % Frequencies in MNC fractions (A) and total estimated numbers of CD34⁺ cells per mL of UCB (B) from each donor group. The CD34⁺ frequencies were obtained from flow cytometry by assessing the CD34⁺ population after doublet discrimination as in Figure S4. The total numbers per mL of UCB were calculated by multiplying the frequency of CD34⁺ cells in the MNCs by the MNC count for each donor and then dividing by the donor’s UCB volume. A total 111 donor samples were analyzed for A and B (n = 39 for the negative, n = 21 and 19 for the N-/S+ and N+/S+ non-vaccinated groups, and n = 21 and 11 for the N-/S+ and N+/S+ vaccinated groups for C., respectively). Displayed are the means with standard deviation bars. p values of unpaired two-tailed t-test with Welch’s correction and one-way ANOVA: ns (p>0.05), ∗ (p<0.05), ∗∗ (p<0.005), ∗∗∗ (p<0.0005), ∗∗∗∗ (p<0.0001). (C and D) Linear regression lines of the CD34⁺ cell frequencies in the MNC fraction versus days post-second SARS-CoV-2 vaccination (C), and the estimated CD34⁺ cell numbers per mL of UCB versus days post-second SARS-CoV-2 vaccination (D) The dashed line indicates the average CD34⁺ cell frequency in the MNC fraction (C) and the average CD34⁺ cell numbers per mL of UCB of the negative donor group as references (D), respectively. A total 25 donor samples were analyzed for C and D (n = 16 and 9 for N-/S+ and N+/S+ vaccinated groups). Data points colored black and red indicate donors single positive for anti-SARS-CoV-2 spike (S) protein IgG (N-/S+) and double-positive for both SARS-CoV-2 nucleocapsid (N) and S protein IgGs (N+/S+), respectively. Displayed are the R² values and the p values of non-zero slopes of the linear regression lines. p-values: ns (p>0.05), ∗ (p<0.05), ∗∗ (p<0.005). Any values showing significance are shown.

Figure 2

CD34⁺cells from donors with previous SARS-CoV-2 infection and vaccination are highly susceptible to apoptosis via IFN-γ-related pathways (A) % Annexin-V positivity of CD34⁺ cell fractions. MNCs obtained from donors double-positive for anti-SARS-CoV-2 N and S protein IgGs (N+/S+) were used for CD34⁺ cell purification by autoMACS. 0.25 × 10⁶ CD34⁺ cells (>98% purity) were used for staining with Biolegend PE Annexin-V and 7-AAD and analyzed by flow cytometry for apoptotic cells by Annexin-V+/7-AAD- as in Figure S5. Displayed are the means with standard deviation bars. p values of unpaired two-tailed t-test with Welch’s correction: ns (p>0.05). (B) Bar plot showing enrichment in different gene ontology (GO) terms predicted by GO analysis of top differentially expressed genes (p≤0.05). (C) Heatmap representing differentially expressed genes belonging to the “IFN-γ-mediated signaling pathway” and “Cellular response to IFN-γ” gene ontology terms significantly suppressed in CD34⁺ cells from the N+/S+ non-vaccinated donor group. N1-N4: samples from negative group (n = 4), DP1-DP4: samples from the N+/S+ non-vaccinated group (n = 4).

Figure 3

Hematopoietic differentiation in vitro is skewed by previous SARS-CoV-2 infection (A) Representative images of the in vitro hematopoietic differentiation assay. Circled in the images are characteristic colony morphologies. Green: CFU-GEMM, Blue: CFU-GM, Red: BFU-E, Purple: CFU-E. Scale bar: 5 mm. (B) Total colony numbers and % frequencies of CFU each lineage. 1 × 10³ CD34⁺ cells were used for the methylcellulose assay. After 14 days of incubation at 37°C, 5% CO₂, each 6-well plate was imaged by the EVOS M7000, and numbers of each colony type were manually counted as in A. Three experimental and biological replicates were performed per donor (n = 7 donors for each group). Displayed are the means with standard deviation bars. p values of unpaired two-tailed t-test with Welch’s correction: ns (p>0.05), ∗∗ (p<0.005). CFU: colony formation units.

Figure 4

In vivo hematopoiesis of CD34⁺ cells is skewed by previous SARS-CoV-2 infection (A and B) X-ray irradiated (120 cGy) NSG (A) or NSG-SGM3 (B) neonate mice (days 1-3) were administered 0.1 × 10⁶ CD34⁺ cells via facial vein, which were obtained from 4 donors in the N-/S- (n = 10 NSG and n = 3 NSG-SGM3 mice) and N/-S+ non-vaccinated groups (n = 11 NSG and n = 6 NSG-SGM3 mice). Peripheral blood (100 μL) was obtained from the retro-orbital vein at 10 weeks post-humanization, stained with an antibody cocktail containing anti-human CD4, CD8, CD14, CD19, CD56, CD66b, CD45, and anti-mouse CD45 antibodies, and analyzed on FACSymphony. % Frequencies of each lineage of human blood cells were calculated as a percent of the total hCD45 and analyzed based on the positivity of the marker for each lineage as in Figure S8. % Frequencies of hCD45 was calculated as a percent frequency of the total human and mouse CD45⁺ population. T/B cell ratios were calculated by dividing the frequencies of the T/B cell populations. Displayed are the means with standard deviation bars. p values of unpaired two-tailed t-test with Welch’s correction: ns (p>0.05), ∗ (p<0.05), ∗∗∗ (p<0.0005). Markers used for defining each blood population were listed in Table S2.