Mechanism of Snhg8/miR-384/Hoxa13/FAM3A axis regulating neuronal apoptosis in ischemic mice model

Abstract

Long noncoding RNAs, a subgroup of noncoding RNAs, are implicated in ischemic brain injury. The expression levels of Snhg8, miR-384, Hoxa13, and FAM3A were measured in chronic cerebral ischemia-induced HT22 cells and hippocampal tissues. The role of the Snhg8/miR-384/Hoxa13/FAM3A axis was evaluated in chronic cerebral ischemia models in vivo and in vitro. In this study, we found that Snhg8 and Hoxa13 were downregulated, while miR-384 was upregulated in chronic cerebral ischemia-induced HT22 cells and hippocampal tissues. Overexpression of Snhg8 and Hoxa13, and silencing of miR-384, all inhibited chronic cerebral ischemia-induced apoptosis of HT22 cells. Moreover, Snhg8 bound to miR-384 in a sequence-dependent manner and there was a reciprocal repression between Snhg8 and miR-384. Besides, overexpression of miR-384 impaired Hoxa13 expression by targeting its 3'UTR and regulated chronic cerebral ischemia-induced neuronal apoptosis. Hoxa13 bound to the promoter of FAM3A and enhanced its promotor activity, which regulated chronic cerebral ischemia-induced neuronal apoptosis. Remarkably, the in vivo experiments demonstrated that Snhg8 overexpression combined with miR-384 knockdown led to an anti-apoptosis effect. These results reveal that the Snhg8/miR-384/Hoxa13/FAM3A axis plays a critical role in the regulation of chronic cerebral ischemia-induced neuronal apoptosis.

Fig. 1. Effect of chronic cerebral ischemia (CCI) to neuronal apoptosis and relative expression levels of Snhg8, miR-384, and Hoxa13.

a H&E-stained (upper part, ×100 or ×400, scale bar = 100 or 25 μm) and terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-biotin nick end labeling (TUNEL) assay (lower part, ×200, scale bar = 50 μm) sections of the hippocampus CA1 region of Sham and CCI group (n = 4, each group). Data are represented as mean ± SD, *P < 0.05 vs. Sham group. b Quantitative real-time PCR and western blot (WB) were used to detect the relative expression levels of Snhg8 (upper left part, n = 5, each group), miR-384 (upper right part, n = 5, each group), and Hoxa13 (lower right part, n = 5, each group). Data are represented as mean ± SD, *P < 0.05 vs. Sham group. c Flow cytometry analysis of CCI-induced HT22 cells (n = 5, each group). Data are represented as mean ± SD, **P < 0.01 vs. Control group. d The expression levels of Snhg8 RNA (upper left part, n = 5, each group), miR-384 RNA (upper right part, n = 6, each group), and Hoxa13 protein (lower part, n = 5, each group) in CCI-induced HT22 cells. Data are represented as mean ± SD, *P < 0.05 vs. Control group, **P < 0.01 vs. Control group

Fig. 2. Snhg8 protected chronic cerebral ischemia (CCI)-induced neuronal apoptosis, while miR-384 exacerbated CCI-induced neuronal apoptosis.

a Flow cytometry analysis of CCI-induced HT22 cells with overexpression of Snhg8 (n = 5, each group). Data are represented as mean ± SD, *P < 0.05 vs. Pre-NC group. b Expression analysis of miR-384 RNA (left part, n = 5, each group), Hoxa13 protein (middle part, n = 5, each group), and FAM3A protein (right part, n = 5, each group) in CCI-induced HT22 cells. Data are represented as mean ± SD, *P < 0.05 vs. Pre-NC group, **P < 0.01 vs. Pre-NC group. c Flow cytometry analysis of CCI-induced HT22 cells with transfection of miR-384 (n = 4, each group). Data are represented as mean ± SD, *P < 0.05 vs. Pre-NC group, ^##P < 0.01 vs. anti-NC group. d Expression analysis of Snhg8 RNA (left part, n = 5, each group), Hoxa13 protein (middle part, n = 4, each group) and FAM3A protein (right part, n = 6, each group) in CCI-induced HT22 cells with transfection of miR-384. Data are represented as mean ± SD, *P < 0.05 vs. Pre-NC group, ^#P < 0.05 vs. anti-NC group, ^##P < 0.01 vs. anti-NC group

Fig. 3. miR-384-targeted Snhg8 and reversed the Snhg8-mediated attenuation of chronic cerebral ischemia (CCI)-induced neuronal apoptosis.

a The predicted miR-384-binding site in Snhg8 (Snhg8-Wt) and the designed mutant sequence (Snhg8-Mut) were indicated. Luciferase reporter assay of HT22 cells cotransfected with Snhg8-Wt or Snhg8-Mut and miR-384 or the miR-384-NC (data are presented as the mean ± SD (n = 4, each group). Data are displayed as mean ± SD, *P < 0.05 vs. Snhg8-Wt+miR-384-NC group. b miR-384 was identified in Snhg8-RISC complex. Relative expression levels of Snhg8 (n = 4, each group) and miR-384 were determined by qRT PCR (n = 4, each group). Data are displayed as mean ± SD, *P < 0.05 vs. anti-normal IgG. c Flow cytometry analysis of CCI-induced HT22 cells cotransfected with Snhg8 and miR-384 (n = 4, each group). Data are displayed as mean ± SD, **P < 0.01 vs. pre-NC+anti-NC group. d Hoxa13 protein levels in CCI-induced HT22 cells cotransfected with Snhg8 and miR-384 (n = 4, each group). Data are displayed as mean ± SD, **P < 0.01 vs. pre-NC+anti-NC group. e FAM3A protein levels in CCI-induced HT22 cells cotransfected with Snhg8 and miR-384 (n = 4, each group). Data are displayed as mean ± SD, **P < 0.01 vs. pre-NC+anti-NC group

Fig. 4. Overexpression of Hoxa13 and overexpression of FAM3A inhibited chronic cerebral ischemia (CCI) induced neuronal apoptosis.

a Flow cytometry analysis of CCI-induced HT22 cells with overexpression of Hoxa13 (upper part, n = 4, each group). Expression analysis of FAM3A RNA (lower left part, n = 5, each group) and FAM3A protein (lower right part, n = 5, each group) in CCI-induced HT22 cells transfected with Hoxa13. Data are represented as mean ± SD, *P < 0.05 vs. Hoxa13 (+)-NC group, **P < 0.01 vs. Hoxa13 (+)-NC group. b Flow cytometry analysis of chronic cerebral ischemia CCI-induced HT22 cells with overexpression of FAM3A (upper part, n = 3, each group). Expression analysis of FAM3A RNA (lower left part, n = 4, each group) and FAM3A protein (lower right part, n = 6, each group) in CCI-induced HT22 cells. Data are represented as mean ± SD, *P < 0.05 vs. FAM3A (+)-NC group, *P < 0.05 vs. Control group, **P < 0.01 vs. Control group. c Putative FAM3A-binding sites are indicated. Immunoprecipitated DNA was amplified by PCR. Normal rabbit IgG was used as a negative control

Fig. 5. Overexpression of miR-384 impaired Hoxa13-induced attenuation of chronic cerebral ischemia (CCI)-induced neuronal apoptosis by targeting 3´-UTR of Hoxa13.

a The predicted miR-384-binding site in Hoxa13 and the designed mutant sequences were indicated. Dual-luciferase report assay of HT22 cells cotransfected with Hoxa13-Wt and miR-384-NC or Hoxa13-Wt and miR-384, and Hoxa13-Mut and miR-384-NC or Hoxa13-Mut and miR-384 (n = 4, each group). Data are presented as mean ± SD, *P < 0.05 vs. Hoxa13-Wt+miR-384-NC group. b Flow cytometry analysis of CCI-induced HT22 cells cotransfected with miR-384 and Hoxa13/Hoxa13- (non-3´UTR) (n = 4, each group). Data are presented as mean ± SD, **P < 0.01 vs. miR-384+Hoxa13-NC group, ^#P < 0.05 vs. miR-384+Hoxa13 group. c FAM3A protein levels in CCI-induced HT22 cells cotransfected with miR-384 and Hoxa13/Hoxa13-(non-3´UTR) (n = 4, each group). Data are presented as mean ± SD, *P < 0.05 vs. miR-384+Hoxa13-NC group, ^#P < 0.05 vs. miR-384+Hoxa13 group

Fig. 6. Overexpression of Snhg8 combined with knockdown of miR-384 protected against chronic cerebral ischemia (CCI) induced neuronal apoptosis in vivo.

a H&E-stained (×100, ×400, scale bar = 100 and 25 μm) and b terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-biotin nick end labeling (TUNEL) assay (×200, scale bar = 50 μm) 0f CA1 area of hippocampus tansfected with Snhg8 and miR-384 in vivo (n = 5, each group). Data are presented as mean ± SD. *P < 0.05 vs. CCI group, ***P < 0.001 vs. CCI group, ^#P < 0.05 vs. CCI+pre-Snhg8, △P < 0.05 vs. CCI+miR-384 group. c Experimental schedule to explore the effects of Snhg8 and miR-384 on CCI-induced neuronal damage in vivo. d schematic illustration of the mechanism of Snhg8/miR-384/ Hoxa13/FAM3A axis regulating neuronal apoptosis