Fig 2: Overexpression of PALI1 triggers K562 cell differentiation along the erythroid lineage, with the effect altered by a separation-of-function allosteric-defective PALI1 mutant (K1241A).a Schematic illustration of the competitive cell proliferation assay used to directly compare the proliferation of cells transduced with different constructs. Cells overexpressing different proteins and fluorophores are co-cultured, with the higher-proliferating cells outcompete the lower-proliferating cells. At the end of the experiment, the number of cells carries each vector is quantified using flow cytometry. b Competitive cell proliferation assay of human chronic myeloid leukaemia (K562) cells overexpressing either PALI1 wild-type (WT), the separation-of-function PALI1 mutant (K1241A) or the LacZ negative control, as indicated. The bar plot represents the mean percentage of each cell population after 7 days of competition, as quantified using flow cytometry. Error bars represent standard deviations derived from three independent experiments carried out on different days with the observed values indicated by dots. Evidence for protein expression and nuclear localisation are in Supplementary Fig. 5a. Source data are provided as a Source Data file. c Colour photographs of pelleted K562 cells overexpressing different proteins: LacZ (left), PALI1 WT (middle) and PALI1-K1241A mutant (right). d Representative histograms generated from flow cytometric analysis of the differentiation markers CD235a (top) and CD44 (bottom) exhibited by K562 cells overexpressing PALI1 WT (red), PALI1 K1241A mutant (orange) or LacZ (blue). The other two replicates shown in Supplementary Fig. 9a, b and evidence for protein expression are in Supplementary Fig 8a. e Illustration of the expected expression level of CD235a and CD44 during erythropoiesis (grey) and the actual observed values represented in dashed lines coloured in red and blue for PALI1 and the negative control LacZ, respectively.

Fig 3: Gene signature scores and gene expression levels derived from scRNAseq comparing SCC, CSCs, and FCC groups. Deconvolution score of lipid metabolism signature (A), expression of CD133 (B), SOX2 (C), PTPRZ1 (D), ITGB8 (E), CD44 (F), and cell cycle score (G). All pairwise comparisons comparing groups to the reference population (i.e., SCC-(A), CD133-(B), SOX2-(C), PTPRZ1-(D), ITGB8-(E), CD44-(F), FCC-(G)) were statistically significant (n = 22, Wilcoxon test, all p-values adjusted for multiple comparisons using Bonferroni method were <0.001). Error bars represent the 95% confidence interval, and the box characterizes the IQR. (H) Expression of CSC marker in SCCs.

Fig 4: MSC characterization and in vitro differentiation capacity.a Flow cytometry results show that MSCs from both young and aged donor mice were uniformly positive for MSCs markers CD44 and Sca-1 and negative for hematopoietic stem cells markers CD34 and CD45 and endothelial marker CD31 (n = 6). b MSCs from young and aged donor mice display fibroblast-like morphology on bright field microscopy (left panels; scale bar = 20 μm). Following exposure to differentiation media MSCs from both young and aged mice showed osteogenic (middle panels) and adipogenic (right panels) differentiation as detected by Von Kossa and oil red O staining, respectively. c Quantitative assessment of calcium deposition showed significantly less calcium deposition in aged MSCs compared to young following exposure to differentiation media (p ≤ 0.05; t test; n = 3 each group). d There was no significant difference in lipid content between groups (t test; n = 3 each). MSCs cultured in control medium served as negative control for both conditions. Error bars show s.d.

Fig 5: In vivo detection of two MSP limb bud subpopulations based on MSC-am localization. (A) Transversal section of an E10.5 mouse embryo to show the histological organization at this stage; the immunodetections of CD29, SCA-1, CD44, CD105, CD90, and CD73 markers in E10.5 limb buds are shown. The limb bud for each MSC-am marker corresponds to the region inside the square in panel (A). (B) Dot plots of the multiparametric flow cytometry analysis showing the hierarchy to identify sSca+ and sSca– subpopulations in E10.5, E11.5, and E12.5 mouse limbs. The number of positive cells for each subpopulation is expressed in percentage; ±mean standard deviation. (C) Comparison of the expression percentage of the sSca+ subpopulation between E10.5, E11.5, and E12.5 limbs. Statistical significance was set as follows: E10.5 vs. E11.5 p > 0.0061; 10.5 vs. 12.5 p > 0.0023; E11.5 vs. E12.5 p > 0.4079, **p < 0.005. NT, neural tube; AGM, aorta-gonad-mesonephros; Do, dorsal; V, ventral; Pr, proximal; D, distal. In (A) AER is marked with an asterisk. The scale bar is set at 200 μm and is representative for all images.