Fig. 5: BCSC secretome confers ferroptosis resistance.

a KEGG analysis of the most enriched pathways in MCF-7 cells cultured with CM from BCSCs or parental cells. b Cell viability of MCF-7, MDA-MB-231 or 4T1 cells cultured with respective control CM ± 1 μM Liproxstatin-1 or BCSC CM ± 5 μM Gallocyanine and treated with a graded concentration of Erastin. c MCF-7, MDA-MB-231 or 4T1 cells were co-cultured with respective parental cells ± 1 μM Liproxstatin-1 or BCSCs ± 5 μM Gallocyanine and treated with 10 μM (MCF-7), 5 μM (MDA-MB-231), and 1 μM (4T1) Erastin, respectively. Cell viability was determined by cell count assay. d Lipid ROS in MCF-7, MDA-MB-231, or 4T1 cells cultured with respective control CM or BCSC CM for 48 h in the presence or absence of 5 μM Gallocyanine and treated with 10 μM (MCF-7), 5 μM (MDA-MB-231), and 1 μM (4T1) Erastin, respectively. e Relative GSH levels in MCF-7, MDA-MB-231, or 4T1 cells cultured with respective control CM or BCSC CM ± 5 μM Gallocyanine and treated with Erastin for 48 h. f Cell viability of MCF-7, MDA-MB-231, or 4T1 cells were cultured with 100 ng/ml recombinant DKK1 or 1 μM Liproxstatin-1 and treated with a graded concentration of Erastin for 48 h. g IHC staining of PTGS2 in lung metastases derived from MDA-MB-231-shCONT or -shDKK1 cells. Scale bar: 50 μm. h GSEA analysis of the enrichment of peroxisomal lipid metabolism in TCGA breast cancer patients with different DKK1 levels. i Cell viability of MDA-MB-231-shCONT or -shDKK1 cells ± 1 μM Liproxstatin-1 and treated with a graded concentration of Erastin for 48 h. j BLI of nude mice intravenously injected with 1 × 10⁶ MDA-MB-231-shCONT or -shDKK1 cells and treated with vehicle or Liproxstatin-1 as indicated. Results are shown as mean ± S.D. *P < 0.05; **P < 0.01; ***P < 0.001; ns not significant (Two-way ANOVA test in (b, f, i) unpaired two-tailed Student’s t test in (j), others one-way ANOVA followed by Tukey’s multiple comparison). Source data are provided as a Source Data file.