Inhibition of Autophagy Prevents Panax Notoginseng Saponins (PNS) Protection on Cardiac Myocytes Against Endoplasmic Reticulum (ER) Stress-Induced Mitochondrial Injury, Ca2+ Homeostasis and Associated Apoptosis

Abstract

Endoplasmic reticulum (ER) stress is often closely linked to autophagy, hypoxia signaling, mitochondrial biogenesis and reactive oxygen species (ROS) responses. Understanding the interaction between ER stress, mitochondrial function and autophagy is of great importance to provide new mechanisms for the pathology, prevention and treatment of cardiovascular diseases. Our previous study has reported that Panax notoginseng saponins (PNS) protection against thapsigargin (TG)-induced ER stress response and associated cell apoptosis in cardiac myocytes is calcium dependent and mediated by ER Ca²⁺ release through RyR₂. However, whether its protection upon ER stress and associated apoptosis is related to mitochondrial function and autophagy remains largely unknown. Here, we investigated the roles of PNS played in TG-induced mitochondrial function, ROS accumulation and autophagy. We also assessed its effects on Ca²⁺ homeostasis, ER stress response and associated cell death in the presence of autophagy inhibition. PNS-pretreated primary cultured neonatal rat cardiomyocytes were stimulated with TG to induce ER stress response. Mitochondrial potential (Δψm) was measured by JC-1. The general and mitochondrial ROS were measured by DCFH-DA and MitoSOX Red, respectively. Autophagy was evaluated by immunofluorescence of LC3, and immunoblots of LC3, p62, ATG7 and PINK1. In addition, mRFP-GFP-LC3 labeling was used to assess the autophagic influx. SiATG7 transfected H9c2 cells were generated to inhibit autophagy. Cytosolic and ER Ca²⁺ dynamics were investigated by calcium imaging. RyR₂ oxidation was tested by oxyblot. Cell viability was examined by TUNEL assay. ER stress response and cell apoptosis were detected by immunoblots of BiP, CHOP, Cleaved Caspase-3 and Caspase-12. The results demonstrated that firstly, PNS protects against TG-induced mitochondrial injury and ROS accumulation. Secondly, PNS enhances autophagy in TG-induced cardiac myocytes. Thirdly, inhibition of autophagy diminishes PNS prevention of TG-induced mitochondrial injury, ROS accumulation and disruption of Ca²⁺ homeostasis. Last but not least, inhibition of autophagy abolishes PNS protection against TG-induced ER stress response and associated apoptosis. In summary, PNS protection against ER stress response and associated apoptosis is related to the regulation of mitochondrial injury and ROS overproduction via modulation of autophagy. These data provide new insights for molecular mechanisms of PNS as a potential preventive approach to the management of cardiovascular diseases.

Keywords: Ca2+ homeostasis; ER stress; PNS; ROS; RyR2 oxidation; apoptosis; autophagy; mitochondrial injury.

FIGURE 1

PNS prevents TG-induced mitochondria injury and ROS accumulation. (A) Primary cultured cardiomyocytes, either untreated (CN group) or pretreated with 40 μg/ml PNS for 12 h (PNS group), before addition of 1 μM thapsigargin (TG group or PNS plus TG group) for 12 h were immunofluorescenced with the primary anti-TOM20 antibody and imaged by a laser scanning confocal microscopy. Representative images in CN and TG group. Scale bar: 30 μm; in box: 10 μm. Bar graph shows percentage of cardiomyocytes with tubular, intermediate or fragmented mitochondria in various groups as indicated (80–100 cells). (B) Primary cultured cardiomyocytes treated as in A were stained with JC-1 and analyzed by a plate reader. Bar graph shows the ratio of aggregated JC-1 (red)/monomeric JC-1 (green) as the mitochondrial membrane potential (Δψm) (Mean ± SEM; 80–100 cells; *p < 0.05, **p < 0.01 relative to CN group, or indicated group). (C) Primary cultured cardiomyocytes treated as in A were immunoblotted with the antibodies to DRP1, MFN2 and β-actin. Bands were quantified relative to β-actin by densitometry (Mean ± SEM; **p < 0.01 relative to CN group, or indicated group). (D) Primary cultured cardiomyocytes treated as in A were stained with DCFH-DA and imaged by a laser scanning confocal microscopy. Representative images in CN and TG group. Scale bar: 80 μm. Bar graph shows the fluorescence intensity of DCFH-DA as the general ROS level (Mean ± SEM; 80–100 cells; **p < 0.01 relative to CN group, or indicated group). (E) Primary cultured cardiomyocytes treated as in A were stained with MitoSOX™ Red and imaged by a laser scanning confocal microscopy. Representative images in CN and TG group. Scale bar: 80 μm. Bar graph shows the fluorescence intensity of MitoSOX Red as the mitochondrial ROS level (Mean ± SEM; 80–100 cells; **p < 0.01 relative to CN group, or indicated group).

FIGURE 2

PNS promotes autophagy and autophagic flux. (A) Primary cultured cardiomyocytes, either untreated (CN group) or pretreated with 40 μg/ml PNS for 12 h (PNS group), before addition of 1 μM thapsigargin (TG group or PNS plus TG group) for 12 h were immunofluorescenced with the primary anti-LC3B antibody and imaged by a laser scanning confocal microscopy. Scale bar: 30 μm; in box: 10 μm. Bar graph shows the number of LC3 puncta per cell in various groups as indicated (Mean ± SEM; 80–100 cells; *p < 0.05, **p < 0.01 relative to CN group, or indicated group). (B) Primary cultured cardiomyocytes treated as in A were transfected with mRFP-GFP-LC3 and imaged by a laser scanning confocal microscopy. Scale bar: 30 μm. Bar graph shows the number of red only and yellow LC3 puncta per cell in various groups as indicated (Mean ± SEM; 60–80 cells; *p < 0.05, **p < 0.01 relative to CN group, or indicated group). (C) Primary cultured cardiomyocytes treated as in A were immunoblotted with the antibodies to LC3B, p62, ATG7, PINK1 and β-actin. Bands were quantified relative to β-actin or proteins as indicated by densitometry (Mean ± SEM; *p < 0.05, **p < 0.01 relative to CN group, or indicated group).

FIGURE 3

Inhibition of autophagy diminishes PNS prevention of TG-induced mitochondria injury and ROS accumulation. (A) SiRNA negative control (siNC group) or siATG7 (siATG7 group) transfected H9c2 cells were identified by immunoblotting of the anti-ATG7 and β-actin antibodies. Bands were quantified relative to β-actin by densitometry (Mean ± SEM; **p < 0.01 relative to siNC group). (B) SiNC or siATG7 transfected H9c2 cells, either untreated (CN group) or pretreated with 40 μg/ml PNS for 12 h before addition of 1 μM TG (TG group or PNS plus TG group) for 12 h were immunoblotted to the antibodies to LC3B and β-actin (C) SiNC or siATG7 transfected H9c2 cells, treated as in B, were immunofluorescenced with the primary anti-LC3B antibody and imaged by a laser scanning confocal microscopy. Scale bar: 30 μm. Bar graph shows the number of LC3 puncta per cell in various groups as indicated (Mean ± SEM; 80–100 cells; *p < 0.05, **p < 0.01 relative to CN group, or indicated group). (D) SiNC or siATG7 transfected H9c2 cells, treated as in B, were immunofluorescenced with the primary anti-TOM20 antibody. Bar graph shows the percentage of cells with tubular, intermediate or fragmented mitochondria in various groups as indicated (80–100 cells) (E) SiNC or siATG7 transfected H9c2 cells, treated as in B, were stained with JC-1 and analyzed by a plate reader. Bar graph shows the ratio of aggregated JC-1 (red)/monomeric JC-1 (green) as the mitochondrial membrane potential (Δψm) (Mean ± SEM; 80–100 cells; *p < 0.05, **p < 0.01 relative to CN group, or indicated group). (F) SiNC or siATG7 transfected H9c2 cells, treated as in B, were stained with DCFH-DA. Bar graph shows the fluorescence intensity of DCFH-DA as the general ROS level. (Mean ± SEM; 80–100 cells; *p < 0.05, **p < 0.01 relative to CN group, or indicated group). (G) SiNC or siATG7 transfected H9c2 cells, treated as in B, were stained with MitoSOX™ Red. Bar graph shows the fluorescence intensity of MitoSOX Red as the mitochondrial ROS level (Mean ± SEM; 80–100 cells; **p < 0.01 relative to CN group, or indicated group).

FIGURE 4

Inhibition of autophagy diminishes PNS suppression on cytosolic Ca²⁺ transits and ER Ca²⁺ releases, as well as RyR₂ oxidation evoked by TG. (A) Representative recordings of TG-evoked cytosolic Ca²⁺ dynamics were recorded by Fura-2 ratios (F ₃₄₀/F ₃₈₀) in siNC (upper panel) or siATG7 (bottom panel) transfected H9c2 cells with (gray, PNS group) or without (dark, CN group) PNS pretreatment (40 μg/ml; 12 h), as indicated. Bar graphs show the cytosolic Ca²⁺ peak amplitude, the area under curve (AUC), as well as the time decay of Ca²⁺ transits in response to TG stimulation (Mean ± SEM; 50–60 responding cells; *p < 0.05 relative to CN group in siNC or siATG7 transfected cells). (B) Representative recordings of TG-induced ER Ca²⁺ dynamics were recorded by the FRET-to-CFP emission ratio (FRET/CFP) in siNC (upper panel) or siATG7 (bottom panel) transfected H9c2 cells with (gray, PNS group) or without (dark, CN group) PNS pretreatment (40 μg/ml; 12 h), as indicated. Bar graphs show the ER Ca²⁺ peak amplitude, the area under curve (AUC), as well as the time decay of Ca²⁺ release in response to TG stimulation (Mean ± SEM; 20–30 responding cells; *p < 0.05 relative to CN group in siNC or siATG7 transfected cells). (C) SiNC or siATG7 transfected H9c2 cells, either untreated (CN group) or pretreated with 40 μg/ml PNS for 12 h before addition of 1 μM TG (TG group or PNS plus TG group) for 12 h were immunoblotted with the antibodies to DNP and RyR₂. Bands were quantified relative to RyR₂ by densitometry (Mean ± SEM; *p < 0.05, **p < 0.01 relative to CN group, or indicated group).

FIGURE 5

Inhibition of autophagy abolishes PNS protection on TG-induced ER stress response and associated apoptosis. (A) SiNC or siATG7 transfected H9c2 cells, either untreated (CN group) or pretreated with 40 μg/ml PNS for 12 h before addition of 1 μM TG (TG group or PNS plus TG group) for 12 h were immunofluorescenced with the primary anti-calnexin antibody and imaged by a laser scanning confocal microscopy. Scale bar: 30 μm; in box: 10 μm. (B) SiNC or siATG7 transfected H9c2 cells, treated as in A, were immunoblotted with the antibodies to BiP, CHOP, Cleaved Caspase-3 and Caspase-12 as well as β-actin. Bands were quantified relative to β-actin by densitometry (Mean ± SEM; **p < 0.01 relative to CN group, or indicated group). (C) SiNC or siATG7 transfected H9c2 cells, treated as in A, were stained with TUNEL and imaged by a laser scanning confocal microscopy. Scale bar: 200 μm. Bar graph shows the percentage of TUNEL positive cells (Mean ± SEM; 200–250 cells; **p < 0.01 relative to CN group, or indicated group).

FIGURE 6

Schematic diagram depicting PNS protection against TG-induced mitochondrial injury, ER stress and associated cell apoptosis. TG inhibits Ca²⁺ reuptake from ER and induces overload of the cytosolic and mitochondrial Ca²⁺ that results in mitochondrial injury, ROS release and activation of the apoptotic pathways subsequently. PNS enhances autophagy to selectively remove damaged mitochondria and results in reduced ROS production and thus the oxidation of RyR₂, which in turn inhibits Ca²⁺ release from ER. PNS thereby attenuates TG-induced Ca²⁺-dependent ER stress and alleviates the associated apoptosis.