Fig. 4.

RBM3 prevents HI-induced apoptosis in vivo and in vitro. a Illustration of in vivo HI model and analysis of apoptosis. RCCA right common carotid artery, TUNEL terminal deoxynucleotidyl transferase dUTP nick end labeling. b, c Representative immunofluorescent TUNEL staining in SVZ (b) or DG (c) of RBM3 WT and KO animals treated with HI and recovered for 7 days. Total TUNEL+ cell number and density in the SVZ or DG were estimated. Six animals were counted per group (n = 6). Scale bar: 50 µm. Contra contralateral (uninjured side), Ipsi ipsilateral (injured side), LV lateral ventricle, GCL granular cell layer. Repeated measures two-way ANOVA was used for statistical analysis; n.s. not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. d Illustration of in vitro OGD model and analysis of apoptosis. OGD oxygen-glucose deprivation. Only WT or KO NSPCs underwent hypothermic treatment (32 °C). Plasmid-transfected NSPCs were always cultured at 37 °C. e–h Representative immunofluorescent TUNEL in NSPCs after OGD stress. SVZ-NSPCs (e) and SGZ-NSPCs (f) from RBM3 WT or KO mice were treated with OGD and hypothermia as stated in Fig.1 legend, except for 48 h reoxygenation period instead of 24 h in total. The ratio of TUNEL+/DAPI+ cells was quantified (three independent experiments, n = 3). Three-way ANOVA was used for statistical analysis; n.s. not significant; **p < 0.01; ***p < 0.001; ****p < 0.0001. SVZ-NSPCs (g) and SGZ-NSPCs (h) transfected with empty vector (Vec) or RBM3 overexpression (OE) plasmid were treated with OGD and reoxygenated at 37 °C for 48 h. The ratio of TUNEL+/DAPI+ cells was quantified (three independent experiments, n = 3). Scale bar: 50 µm. Two-way ANOVA was used for statistical analysis; n.s. not significant; **p < 0.01. All data are presented as mean ± SEM