Figure 4. CD31^low cells are resistant against doxorubicin-induced apoptosis.

(A) Both sublines were treated with increasing concentrations of doxorubicin for 24 hours and cell survival was determined using MTS assay. At concentrations >500nM, CD31^low cells survived significantly better than CD31^high cells (p<0.001). (B) Immunoblot analysis of CD31^low and CD31^high cells treated with increasing concentrations of doxorubicin showed activation of apoptotic enzymes (cleaved PARP, cleaved caspase-3 and −7) only in CD31^high but not in CD31^low cells. See also Supplementary Figure 2A for densitometric quantification. (C) Parental CD31 heterogeneous ASM cells with a high CD31 subpopulation (CD31^high:92.8% vs CD31^low: 3.7%) were treated with 1 μM doxorubicin (respective isotype controls are presented in Supplementary Figure 2B). After 24 hours treatment, medium was changed to normal. After seven days the surviving ASM cells (termed doxo-survivors) were analyzed for CD31 expression via flow cytometry and (D) Western Blot. See also Supplementary Figure 2B and C for morphology and functional features of parental and doxo-surviving ASM cells. (E) Schematic description of cell trace experiments. CD31^high (or CD31^low cells, respectively) were labeled with the tracer dye CellTrace Violet and then mixed with equal numbers of CD31^low cells (or CD31^high cells, respectively) and then incubated with increasing doses of doxorubicin for 24 hours. (F and G) FACS analysis of these mixed labeled cell populations revealed preferential depletion of CD31^high cells, while the size of the labeled CD31^low population remained stable. All data are mean ± SEM and were analyzed using two-way ANOVA followed by Bonferroni’s multiple comparisons test (*p < 0.05;**p < 0.01; ***p < 0.001).