Fig 2: Inhibition of perforin-mediated neurotoxicity facilitates neurogenesis. (A) Flow cytometry was used to detect astrocytes and neuron stem/progenitor cells by according to Glast and 7AAD expression, and these cells were divided into Glast+ and Glast– subtypes 14 days after ischemia surgery and sham operation group in the focal ischemic brain area. Then, these three subtypes were divided into stem/progenitor cells according to CD133 and CD24 expression. (B,C) The box graph shows the numbers of mature astrocytes and glial progenitors. The data are shown as the median and the value of min to max; n = (6–10) animals in each group. Prf1–/– mice vs. WT mice at each time point. **P < 0.01.

Fig 3: Expansion of spheres derived from single dysplastic stem cells in Meta4.a Phase contrast images of spheres derived from isolated Triple-negative (TN), Double-positive (DP) or Triple-positive cells (TP) at passage 0 (P0), 2 (P2) and 4 (P4). Images were captured at about 1 week after plating (P0), at day 5 of P2, at day 5 of P4. TN (CD44neg/CD133neg/CD166neg) cells were isolated as a control population. Scale bars indicate 1000 µm. b Quantitation of the number of spheres formed from the Triple-negative (TN), Double-positive (DP) or Triple-positive (TP) cells at P0, P2, and P4. Data are presented as mean values with standard deviation (n = 3). P-values were calculated using paired two-tailed t-test. *P = 0.048 (DP vs. TP at P0); n.s., not significant. c Representative FACS profiles of the Triple-negative (TN), Double-positive (DP) and Triple-positive (TP) cells analyzed in unsorted Meta4 organoids and spheres derived from isolated DP or TP cells at P2 and P4. CD44 positive gate was used to distinguish DP and TP cells (right graphs) among CD133+/CD166+cells (top right boxes, middle graphs) or to distinguish TN cells (left graphs) from CD133neg/CD166neg cells (bottom left boxes, middle graphs). The percentage of each TN, DP, and TP cells was not significantly different between unsorted Meta4, DP and TP at P2, and DP and TP at P4. Source data are provided as a Source Data file.

Fig 4: Identification of dysplastic stem cells in Meta4.a–e Clonal sphere forming efficiency from Meta4 subpopulations. a Meta4 organoids were dissociated to single cells, stained for CD44, CD133, CD166, and FACS isolated. Following doublet-discrimination and dead cell exclusion, CD133/CD166 positive cells (gray gate) were applied (red arrow) to a CD44 positive gate to distinguish between double-positive CD44neg/CD133+/CD166+ (orange gate) and triple-positive CD44+/CD133+/CD166+cells (green gate). Double-positive and Triple-positive cells were collected by FACS and applied to CRAs. b Immediately after plating, single cells were quantified and their physical address on the CRA was determined. c, d After 8 days of culture, spheres were quantified and their physical address on the CRA was determined. Upper-panels are low magnification of each CRA containing 2025 wells. Red-boxes indicate the regions in the higher magnification images depicted in the lower panels. e Spheres-Forming Efficiency (SFE) of clonal events was determined by quantifying the number of spheres that developed from all single cell events at t = 0. There were 619 Double-positive clonal events identified in the CRAs, and 28.9% of those generated spheres (orange chart). There were 509 Triple-positive clonal events identified in the CRAs, and 6.3% of those generated spheres (green chart). The data represent binary outcomes. f Quantitation of immunofluorescence staining of Ki67-positive cells in either Double-positive (DP) or Triple-positive (TP) cell zone in Mist1-Kras mouse stomach corpus at 4 months after tamoxifen injection. Data are presented as mean values with standard deviation (n = 12). P-values were calculated using paired two-tailed t-test. g Immunostaining for CSC markers, CD44 (basolateral membrane), CD133 (apical membrane) and CD166 (basolateral membrane), metaplasia and gastric cancer markers, Sox9 and CD44v9, and a proliferation marker, Ki67 in Mist1-Kras mice at 4 months after tamoxifen injection. Yellow arrow indicates Double-positive (DP) cells and white arrow indicates Triple-positive (TP) cells. Dotted boxes indicate enlarged area. Scale bars indicate 100 µm. Source data are provided as a Source Data file.

Fig 5: SLC25A1-dependent mitochondrial respiration drives therapy resistance. a, b T1 and T2 cells were treated with cisplatin (0.5–1 µM) or AZD9291 (1 µM), as indicated in each panel, passaged in the presence of the drugs several times, then subjected to OCR analysis with the Seahorse analyzer. c SLC25A1 protein levels in T1 and T2 cells cultured in the presence or absence of cisplatin. d FACS analysis of CD166 and CD133 markers in the cisplatin or AZD9291-resistant T1 or T2 cells. e–g T1 e, T2 f and T4 g cells were untreated or treated with the indicated drugs. The concentration of drugs employed is indicated at the bottom of each panel. Viability was assessed with Crystal Violet after 5 days of treatment. Relative (R) index calculations were used to assess the type of drug interactions and are indicated in each panel. The R index is calculated as the expected cell survival (Sexp; the product of relative survival in cisplatin and relative survival in CTPI-2) divided by the observed relative survival in the presence of both drugs (Sobs). Sexp/Sobs = 1.0 denotes an additive interaction, whereas >1.0 denotes a synergistic interaction, R index values approaching 2.0 are indicative of strong synergy. h Kaplan–Meier survival curves relative to SLC25A1 expression in lung AdenoCa patients that received chemotherapy. Red and black lines indicate high and low SLC25A1 expression, respectively. Analysis was performed by using the Km plotter database (http://kmplot.com/analysis/). Bars represent the standard deviation, asterisks refer to **P = 0.01; ***P = 0.001; ****P = 0.0001 by unpaired T-test