Distinct mechanisms of cardiomyocyte apoptosis induced by doxorubicin and hypoxia converge on mitochondria and are inhibited by Bcl-xL

Abstract

Hypoxia and doxorubicin can cause cardiotoxicity and loss of myocardial function. These effects are due, in part, to an induction of apoptosis. Herein we identify the apoptotic pathways activated in H9c2 cells in response to hypoxia (O(2)/N(2)/CO(2), 0.5:94.5:5) and doxorubicin (0.5 muM). Although the apoptosis induced was accompanied by induction of Fas and Fas ligand, the death receptor pathway was not critical for caspase activation by either stimulus. Hypoxia induced the expression of endoplasmic reticulum (ER) stress mediators and processed ER-resident pro-caspase-12 whereas doxorubicin did not induce an ER stress response. Most importantly, both stimuli converged on mitochondria to promote apoptosis. Accumulation of cytochrome c in the cytosol coincided with the processing of pro-caspase-9 and -3. Increasing the expression of the anti-apoptotic protein Bcl-x(L), either by dexamethasone or adenovirus-mediated transduction, protected H9c2 cells from doxorubicin- and hypoxia-induced apoptosis. Bcl-x(L) attenuated mitochondrial cytochrome crelease and reduced downstream pro-caspase processing and apoptosis. These data demonstrate that two distinct cardiomyocyte-damaging stimuli converge on mitochondria thus presenting this organelle as a potentially important therapeutic target for anti-apoptotic strategies for cardiovascular diseases.

1

Hypoxia and doxorubicin induce H9c2 cell apoptosis. H9c2 cells were exposed to hypoxia or treated with 0.5 μM doxorubicin for the times indicated. (A) Gross morphological features were assessed by light microscopy of H&E-stained cells whilst nuclear alterations were examined by fluorescence microscopy of Hoechst-stained nuclei. (B) Caspase-3 processing in response to hypoxia and doxorubicin in whole cell lysates (30 μg), analysed by Western blotting.

2

Expression and activity of death receptor pathway mediators in response to hypoxia and doxorubicin. (A) Expression of Fas, Fas ligand (FasL) and cFLIP mRNA during exposure to hypoxia or doxorubicin. Total RNA was isolated at the time-points indicated and examined for levels by RT-PCR. GAPDH was used as an endogenous loading control as it remained unaltered by treatment. (B) Activity and requirement of the Fas death receptor pathway for hypoxia and doxorubicin-induced cardiomyocyte death. H9c2 cells were exposed to hypoxia or 0.5 μM doxorubicin for 24 hrs, after 3 hrs of treatment, FasL (2 μg/ml) and/or Fas:Fc (2 μg/ml) plus enhancer (1 μg/ml) were added. Cell death was measured by annexin V staining and analysed by flow cytometry. (C) Fas sensitive Jurkat cells were used as positive control to test the functionality of recombinant FasL and neutralizing Fas:Fc. Jurkat cells were treated with 2 μg/ml recombinant FasL with or without 2 μg/ml recombinant Fas:Fc with 1 μg/ml Fas:Fc enhancer for 24 hrs. The values displayed are the means ± S.D. of three independent experiments, *P < 0.05.

3

Hypoxia, but not doxorubicin, initiates the ER stress response in H9c2 cells. (A) mRNA expression levels of the unfolded protein response (UPR) mediators XBP-1, GADD34 and ATF4 assessed by RT-PCR of total RNA samples isolated from cells exposed to hypoxia or 0.5 μM doxorubicin for the times indicated. (B) Alterations in the expression of ER stress associated proteins analysed by Western blotting using protein harvested after exposure to hypoxia or 0.5 μM doxorubicin for up to 48 hrs. Whole cell lysates were analysed for levels of Grp78 (10 μg) and CHOP (30 μg) and processing of pro-caspase-12 (30 μg). ER stress inducer thapsigargin (2 μM; 4 hrs for mRNA, 36 hrs for protein) was used as a positive control (+ve). displayed are the means ± S.D. of three independent experiments, *P < 0.05.

4

Hypoxia and doxorubicin induce mitochondrial cytochrome c release and pro-caspase-9 processing. (A) Cytosolic accumulation of cytochrome c during hypoxia and doxorubicin treatment. After treatment, cells were harvested at the time-points indicated and separated into cytosolic- and mitochondrial-rich fractions by centrifugation. The cytosolic-rich (left-hand panel) and mitochondrial-rich (righthand panel) fractions (10 μg) were analysed by Western blotting using anti-cytochrome c antibody. (B) Pro-caspase-9 processing during hypoxia (left-hand panel) and doxorubicin (righthand panel) treatment. Whole cell lysates (30 μg) were analysed by Western blotting for cleavage of pro-caspase-9 using anti-caspase-9 antibody. Actin levels in samples were analysed to serve as loading control.

5

Dexamethasone increases Bcl-x_L protein expression and reduces hypoxia- and doxorubicin-induced caspase-3-like activity. (A) Bcl-x_L protein expression was examined by Western blotting following 1 μM dexamethasone treatment (0–24 hrs) with and without RU486 pre-treatment. (B) Bcl-xL protein expression was also examined following treatment with dexamethasone (1 μM; 24 hrs) or doxorubicin (0.5 μM; 24 hrs) separately and in combination with RU486. (C) H9c2 cells were exposed to doxorubicin (upper graph) or hypoxia (lower graph) with or without pre-treatment with dexamethasone (1 μM;for 24 hrs) and RU486 (4 μM; 30 min prior to dexamethasone pre-treatment). Caspase-3-like activity was detected by monitoring the cleavage of synthetic substrate DEVD-AMC by active caspases. Results are expressed as nmoles AMC released in 1 min by 1 mg total cellular protein and plotted as mean values ± S.D. of three independent experiments, *P< 0.05.

6

Adenoviral-mediated overexpression of Bcl-x_L reduces hypoxiaand doxorubicin-induced H9c2 cell apoptosis. Cells were transduced with AdBcl-x_L or AdNull at an MOI of 1000 for 1 hr and cultured for 24 hrs before treatment with hypoxia and doxorubicin (0.5 μM) for 48 hrs. (A) Overexpression of Bcl-x_L confirmed by Western blotting in samples 24 hrs after transduction. (B) Hypoxia- and doxorubicin-induced accumulation of cytochrome c in the cytosol is prevented by overexpression of Bcl-x_L. Subcellular fractionation was performed by differential centrifugation of cell lysates exposed to hypoxia or doxorubicin with the cytosolic-rich fractions analysed for presence of cytochrome c by Western blotting. Bcl-x_L prevents pro-caspase-3 processing induced by hypoxia and doxorubicin. Whole cell lysates of nontransduced, AdNull- and AdBcl-xLtransduced cells, untreated and treated, were analysed for pro-caspase-3 cleavage by Western blotting. (C) Non-transduced, AdNull and AdBcl-x_L infected cells were exposed to hypoxia for 48 hrs or doxorubicin for 24 hrs and stained with Hoechst to analyse nuclear fragmentation. Overexpression of Bcl-x_L protects cells from hypoxia- and doxorubicin-induced apoptosis.