Fig 2: SNAI1 directly regulates CD73 expression in human TNBC MDA-MB-468-iSNAI1 cells. (A) Schematic representation of the different E-boxes identified in silico in the human CD73 promoter (CD73 mRNA, NCBI Reference Sequence: NM_002526). The transcription Start Site (TSS) and the ATG start codon are reported at positions +1 and +58, respectively. The control motif (+800 to +995) corresponds to a region containing no E-box and is used as a negative control. (B) ChIP was performed on MDA-MB-468-iSNAI1 cells cultured under EPI or EMT conditions using anti-SNAI1 antibodies followed by five pairs of primers flanking the identified E-boxes (E-box-1-5) or primers flanking control region. E-cadherin (E-CAD) primers were used as a positive control. For each gene, the RT-qPCR signals were normalized to control IgG. SNAI1 ChIP signal was reported as fold enrichment over IgG control. Two individual experiments (done in triplicate) were performed (*P < 0.05, **P <0.01, ***P < 0.001 calculated by unpaired t-test). (C) Correlation between NT5E/CD73 and SNAI1 in tumor mRNA from 12 TNBC patients. Pearson correlation coefficient (r) and P-value are shown. (D) Analysis process of METABRIC dataset. (E) Correlation between NT5E, SNAI1, VIM, and CDH1 gene expression in TNBC patients from METABRIC dataset (n = 258). For each association: Upper panel: correlation between NT5E/CD73 and the indicated gene. Pearson coefficients and P values are shown. Lower panel: TNBC samples were separated according to NT5E/CD73 expression level to form high (Z score = +0.5) and low (Z score =-0.5) NT5E/CD73 groups. In each group. SNAI1, VIM, and CDH1 gene expression were evaluated. Dots represent mRNA level ± SEM (**P <0.01, ****P < 0.0001 calculated by unpaired t-test). ns, not significant.

Fig 3: Characterization of vascular progenitor cells (VPCs). A, Representative images of c-kit staining in the adventitia of aorta from control and Marfan syndrome (MFS) donors. B, The c-kit positive cells in the adventitia of the aorta from control and MFS donors were calculated and are presented as a percentage of total cells. C, Surface marker profiling determined by fluorescence-activated cell sorting (FACS) in VPCs. that is, positive for c-kit, CD29, CD73, CD90, CD105; negative for CD31, CD45. D, Representative images of a-SMA and Calponin staining showing VPCs differentiated into smooth muscle cells. F, The differentiation capacities of adipogenesis and osteogenesis of VPCs in vitro were determined by Oil red staining and Alizarin red staining respectively. E, Quantitative analysis of fluorescence intensity of a-SMA and Calponin in control-VPCs and MFS-VPCs after differentiation. G, Quantification of the area occupied by the oil red O staining and Alizarin red staining in control-VPCs and MFS-VPCs after differentiation. Data are expressed as mean ± SEM. **P < 0.01. Scale bar = 100 µm

Fig 4: GMP stem cell manufacturing and characterization. A Conceptual figure showing COVID-19 ARDS patient lung microenvironment and methodological approach utilized in the study: All three cell products (BM-MSCs, HDCs, & UC-MSCs) were prepared to the clinical-grade standards in the cell manufacturing facility. Using COVID-19 clinical literature survey and flow cytometry cytokine receptor expression profile, five COVID-19 ARDS related cytokines (IL-1ß, IL-2, IL-8, IL-10, and TNF-a) were selected for experimentation in the current study. Followed by an assessment of individual cytokine dose response on cell viability, the impact of cytokine cocktail containing all five cytokines at clinically relevant doses on paracrine output (extracellular vesicle production) and cell viability of three cell types was evaluated. All experiments were conducted at hypoxia (1% oxygen) for 48-h to simulate transplanted cell residence in hypoxic COVID-19 ARDS lung microenvironment. B Flow cytometry analysis showed BM-MSCs & UC-MSCs were positive for markers of mesenchymal stromal cell origin (CD73, CD90, CD105, CD166, and CD44) and negative for markers of hematopoietic cell origin (CD14, CD19, CD34, CD45 and HLA-DR). C qPCR showed IFN-? treatment for 24-h induces IDO-gene expression in all three cell types (GAPDH was used as a reference gene). D Colorimetric assay showed Tissue factor activity of all three cell types was found to be similar. E qPCR showed none of the cell types express ACE2 (?Cq values < -20 when normalized to reference gene, GAPDH). All data are presented as individual and mean values ± SEM, n = 3 biological replicates; each circle represents one data point from one unique biological replicate

Fig 5: SNAI1-mediated CD73 upregulation results in increased extracellular ADO production. (A) Concentration of extracellular ADO released in the culture medium of MDA-MB-468-iSNAI1 cells cultured under control conditions (EPI), after 5 days of Dox (EMT), and after Dox removal (MET). Bars represent means from three independent experiments ± SEM (*P < 0.05, **P < 0.01 calculated by unpaired t-test). (B) Concentration of extracellular ADO released in the culture medium of MDA-MB-468-iSNAI1 cells cultured under control conditions (EPI) and after 5 days of Dox (EMT) in the absence or presence (+APCP) of APCP (100 µM). Bars represent means from three independent experiments ± SEM (*P < 0.05, **P < 0.01 calculated by unpaired t-test). (C) NK cells from four healthy donors (NKD-1 to 4) were pre-treated with CADO at 5 µM or 10 µM during 4 days (NKD-1 and NKD-2) or 6 days (NKD-3 and NKD-4). NK cytotoxic activity against K562 cells was measured at the indicated E:T ratios. Bars represent the mean percentage of lysis ± SD. (D) Cytotoxic activity of NK92-MI cells pre-treated for 6 days with CADO at 5 µM or 10 µM against K562 cells at the indicated E: T ratios. Bars represent mean percentage of lysis from three independent experiments ± SD (*P < 0.05, **P < 0.01, ***P < 0.001 calculated by unpaired t-test). Only one experiment was performed for each NK healthy donor reported in panels (C, D). Experiments were performed either in duplicates or triplicates.