Impaired autophagosome clearance contributes to cardiomyocyte death in ischemia/reperfusion injury

Abstract

Background: In myocardial ischemia, induction of autophagy via the AMP-induced protein kinase pathway is protective, whereas reperfusion stimulates autophagy with BECLIN-1 upregulation and is implicated in causing cell death. We examined flux through the macroautophagy pathway as a determinant of the discrepant outcomes in cardiomyocyte cell death in this setting.

Methods and results: Reversible left anterior descending coronary artery ligation was performed in mice with cardiomyocyte-restricted expression of green fluorescent protein-tagged microtubule-associated protein light chain-3 to induce ischemia (120 minutes) or ischemia/reperfusion (30-90 minutes) with saline or chloroquine pretreatment (n=4 per group). Autophagosome clearance, assessed as the ratio of punctate light chain-3 abundance in saline to chloroquine-treated samples, was markedly impaired with ischemia/reperfusion compared with sham controls. Reoxygenation increased cell death in neonatal rat cardiomyocytes compared with hypoxia alone, markedly increased autophagosomes but not autolysosomes (assessed as punctate dual fluorescent mCherry-green fluorescent protein tandem-tagged light chain-3 expression), and impaired clearance of polyglutamine aggregates, indicating impaired autophagic flux. The resultant autophagosome accumulation was associated with increased reactive oxygen species and mitochondrial permeabilization, leading to cell death, which was attenuated by cyclosporine A pretreatment. Hypoxia-reoxygenation injury was accompanied by reactive oxygen species-mediated BECLIN-1 upregulation and a reduction in lysosome-associated membrane protein-2, a critical determinant of autophagosome-lysosome fusion. Restoration of lysosome-associated membrane protein-2 levels synergizes with partial BECLIN-1 knockdown to restore autophagosome processing and to attenuate cell death after hypoxia-reoxygenation.

Conclusion: Ischemia/reperfusion injury impairs autophagosome clearance mediated in part by reactive oxygen species-induced decline in lysosome-associated membrane protein-2 and upregulation of BECLIN-1, contributing to increased cardiomyocyte death.

Figure 1

Myocardial ischemia-reperfusion injury impairs autophagosome processing, in-vivo. A) Representative electrocardiographic tracings from mice subjected to reversible LAD coronary artery occlusion to induce myocardial ischemia, as evidenced by ST segment elevation (middle panel) followed by reperfusion, with resolution of ST segment elevation (right panel), as compared with baseline (left panel). B) Representative grayscale immunofluorescence images (630X) of left ventricular (LV) myocardial sections from mice with cardiomyocyte-specific expression of GFP-LC3, subjected to sham surgery (left), ischemia (120 min, middle) or ischemia (30 min) followed by reperfusion (90 min, right), in the presence of chloroquine (CQ; 10 mg/kg injected i.p. 1 hour prior to surgery, bottom) or saline-diluent control (DIL, top). C) Quantitative analysis of punctate LC3 dots from (B) (n=4 mice/group). D) Immunoblots; and E–I) Quantitative analysis of autophagosome bound GFP-LC3-II (E), p62 (F), LAMP2 (G), LAMP1 (H) and BECLIN-1 (I) abundance in myocardial extracts from mice modeled as in B; (n=4/group). P values are by post-hoc test after one-way ANOVA (C, E-I).

Figure 2

Scavenging reactive oxygen species restores autophagosome processing in myocardial ischemia-reperfusion injury, in vivo. A–F) Representative immunoblots (A) and quantitative analysis of LC3-II (B), p62 (C), LAMP2 (D), LAMP1 (E), and BECLIN-1 (F) expression in myocardial extracts from mice subjected to sham surgery or ischemia (30 minutes) followed by reperfusion (90 minutes), in the presence of MnTMPyP (6mg/kg i.p.) or saline control injected 1 hour prior to surgery. N=5–6/group. P values are by post-hoc test after one-way ANOVA.

Figure 3

Impaired autophagosome clearance causes cardiomyocyte death. A) Representative immunofluorescence images (630X) of NRCMs expressing mCherry-GFP-LC3 and treated with rapamycin (Rapa, 5μmol/L) (middle), rapamycin (5μmol/L) + chloroquine (CQ, 10mol/L) (bottom) or vehicle control (top) for 24 hours. Nuclei are blue (DAPI). Representative of n=3 experiments. B) Quantitative analysis of autophagosomes (red+green=yellow dots, white bars), autolysosomes (red only dots, black bars) and both (red dots, gray bars) in NRCMs treated as in A. N=30–50 nuclei per group. P values are by paired t-test. C) Representative immunoblot depicting LC3 processing and p62 abundance in NRCMs treated with rapamycin, without and with chloroquine (as in A); a bright line separates two parts of one gel grouped together; representative of n=2 experiments. D) Cell death in NRCMs treated with rapamycin (Rapa, 5μmol/L, 24 hours) or vehicle control, and subjected to nutrient deprivation (ND, 24 hours) in the absence and presence of chloroquine (CQ, 10μmol/L, 24 hours); n=8/group; P values are by post-hoc test. E) Representative immunofluorescence images (630X) of NRCMs loaded with ROS indicator carboxy-H2DCFDA (top), TMRE (middle) and merged images (bottom), treated with CQ (10μmol/L) in the presence of CsA (20μmol/L) or DMSO control. Treatment with H2O2 (500μmol/L, 4 hours) was employed as a positive control. F–G) Representative tracings of flow cytometric analysis of NRCMs treated as in E (left) with quantitation of mean fluorescence (right) (n=3–5/group) for ROS (F) and TMRE (G). * indicates P< 0.05 vs Control; # vs. CQ treatment by post-hoc test. H) Cell death in NRCMs treated with CQ (10μmol/L) with or without treatment with Rapamycin (5 μmol/L) for 24 hours; in the presence of ZVAD-fmk (20μmol/L), CsA (20μmol/L) or DMSO control. * indicates P< 0.05 vs Control; # vs. respective CQ treatment group by post-hoc test (n=10–30/group).

Figure 4

Hypoxia-reoxygenation injury upregulates autophagy with impaired autophagosome processing in NRCMs. A) Representative immunofluorescence images (630X) demonstrating mCherry-GFP-LC3 localization in NRCMs cultured in a normoxic environment (top, as controls), subjected to hypoxia (24 hours, middle) or hypoxia (6 hours) followed by reoxygenation (18 hours, bottom). Nuclei are blue (DAPI). Representative of n=3 experiments. B) Quantitative analysis of autophagosomes (white bars), autolysosomes (black bars) and both (gray bars) in NRCMs treated as in A. P values are by post-hoc test between groups, and paired t-test within a group. N=30–50 nuclei per group. C, D) Representative immunoblots depicting LC3 processing and LAMP2 abundance in response to hypoxia, and hypoxia-reoxygenation injury for 24 and 48 hours; (C) and time course for change in LAMP2, LAMP1, BECLIN-1 and p62 abundance in response to hypoxia-reoxygenation injury (D). Representative of n=2 experiments.

Figure 5

Impaired autophagosome processing is associated with cell death in NRCMs subjected to hypoxia-reoxygenation injury. A, B) Representative immunofluorescence images (630X) depicting CFP-tagged polyglutamine Q80 accumulation (green) (A) and quantitative analysis of Q80 aggregates (B) in NRCMs subjected to hypoxia-reoxygenation (Hyp/Reox), hypoxia (Hyp) and normoxic controls (Nor). Nuclei are red (TOPRO-3). P value is by post-hoc test. Representative of n=2 experiments. N=30–50 nuclei/group. C, D) Representative immunofluorescence images (200X) (C) and quantitative analysis (D) of alive (green) and dead (red) NRCMs treated as in A. * indicates P< 0.05 by post-hoc test; n=24–40/group. E) Representative immunofluorescence images (630X) of NRCMs loaded with ROS indicator carboxy-H2DCFDA (top), TMRE (middle) and merged (bottom) and subjected to hypoxia-reoxygenation as in A, in the presence of CsA (20μmol/L) or DMSO control. F–G) Quantitation of mean fluorescence of flow cytometric analysis of NRCMs treated as in E for ROS (F) and TMRE (G). * indicates P<0.05 vs Nor by post-hoc test (n=3–4/group). H) Cell death in NRCMs subjected to hypoxia-reoxygenation (Hyp/Reox) as in A, in the presence of ZVAD-fmk (20μmol/L), CsA (20μmol/L) or DMSO, and normoxic (Nor) control. * indicates P<0.05 vs Nor by post-hoc test (n=10–20/group).

Figure 6

Hypoxia-reoxygenation-induced LAMP2 decline and BECLIN-1 upregulation are mediated via reactive oxygen species. A) Representative immunoblot (left) and quantitative analysis (right) of LAMP2 abundance in NRCMs subjected to hypoxia (6 hours) followed by reoxygenation (18 hours; Hyp/Reox) or normoxia (Nor) in presence of diluent (control) and MnTMPyP (50μmol/L); n=3/group. P values are by post-hoc test. B) Representative immunoblot (left) and quantitative analysis (right) of BECLIN-1 abundance in NRCMs treated as in A; n=4/group. P values are by post-hoc test. C) Representative immunofluorescence images (630X) depicting mCherry-GFP- LC3 expression in NCRMs treated as in A (Scale bars= 10μm); D) Quantitative analysis of autophagosomes (white bars), autolysosomes (black bars) and both (gray bars) in NRCMs treated as in A. ‘*’ indicates P< 0.05 for autophagosomes, and ‘#’ indicates P< 0.05 for total (autophagosomes+autolysosomes) as compared with respective normoxia control by post-hoc test. P values depicted are for comparisions between autophagosomes and autolysosomes by paired t-test within a group (n=15–20 nuclei/group).

Figure 7

Beclin-1 knockdown and LAMP2 overexpression restore autophagosome processing and rescue cell death in NRCMs subjected to hypoxia-reoxygenation injury. A) Immunoblot depicting BECLIN-1, LC3 and p62 abundance in NRCMs treated with increasing dose of adenovirus transducing Beclin-1 shRNA (for 48 hours) and subjected to hypoxia-reoxygenation (H/R: 6/18 hours) or normoxia (Nor) as control. Representative of n=2 experiments. B) Immunoblot depicting LAMP2, LC3 and p62 expression in NRCMs transduced with adenovirus expressing rat LAMP2A and rat LAMP2B (each 10 MOI) for 24 hours and subjected to hypoxia-reoxygenation as in A. Representative of n=2 experiments. C) Representative immunofluorescent images (630X) depicting mCherry-GFP-LC3 expression in NRCMs transduced with adenovirus expressing LacZ (control), rat LAMP2A, rat LAMP2B (10MOI each); and Beclin-1shRNA (1MOI) + rat LAMP2A (10 MOI) + rat LAMP2B (10 MOI) and subjected to hypoxia-reoxygenation as in A and B. Scale bars= 10μm. D) Quantitative analysis of autophagosomes (white bars), autolysosomes (black bars) and both (gray bars) in NRCMs treated as in C. * indicates P< 0.05 vs normoxia for autophagosomes, # indicates P< 0.05 vs normoxia for autolysosomes, $ indicates P< 0.05 vs normoxia for autolysosomes+autophagosomes within the same treatment condition (n=15–20 nuclei/group) by post-hoc test. P values depicted are by paired t-test within each group. E) Cell death in NRCMs adenovirally transduced with LacZ (control), rat LAMP2A (10 MOI), rat LAMP2B (10 MOI), Beclin-1 shRNA (1 MOI), Beclin-1 shRNA (1MOI) + rat LAMP2A (10 MOI) + rat LAMP2B (10 MOI); and Beclin-1 shRNA (10 MOI) as in C. P values indicated are by post-hoc test (n=8–40/group). * indicates P< 0.05 vs LacZ treated hypoxia-reoxygenation group, # indicates P< 0.05 vs LacZ normoxic control.

Figure 8

BECLIN-1 levels regulate transcription of autophagy genes to affect autophagosome processing. A) Representative immunoblot depicting the time course of changes in protein levels of LAMP1, LC3, p62 and RAB7 in NRCMs adenovirally transduced with shRNA targeting Beclin-1 or a LacZ as non-targeting (NT) shRNA control (both at MOI=1), at 24, 48 and 72 hours after treatment with viral particles. Representative of n=2 experiments. B–E) Real-time PCR-based quantitation of MAP1LC3B (coding for LC3), SQSTM1 (for p62), LAMP1 and RAB7 transcripts in samples treated as in A. N=4/group at each time point. P values are by t-test vs control at each time point. F) Representative immunofluorescence images (630X) demonstrating mCherry-GFP-LC3 localization in NRCMs infected with increasing doses of adenoviruses (MOI=1, 10, 50 and 100) coding for HA tagged-BECLIN-1 or LacZ as control for 48 hours. G) Quantitative analysis of autophagosomes (white bars), autolysosomes (black bars) and both (gray bars) in NRCMs treated as in A; n=10–20 nuclei/group. * and # indicate P< 0.05 vs respective control for autophagosomes and both (autophagosomes and autolysosomes), respectively, by post-hoc test. P value within each group is by paired t-test. H) Immunoblots (representative of n=2 experiments) depicting Beclin-1 (HA), LC3, p62, RAB7 and LAMP1 expression in NRCMs treated as in F. I–L) Real-time PCR-based quantitation of MAP1LC3B, SQSTM1, LAMP1and RAB7 transcripts in NRCMs adenovirally transduced with Beclin-1-HA or LacZ (100MOI) as control for 24 hours. N=4/group at each time point. P values are by t-test vs control. M) Cell death in NRCMs treated as in F; n=8–16/group. P values noted are by post-hoc test.