The dual role of poly(ADP-ribose) polymerase-1 in modulating parthanatos and autophagy under oxidative stress in rat cochlear marginal cells of the stria vascularis

Abstract

Oxidative stress is reported to regulate several apoptotic and necrotic cell death pathways in auditory tissues. Poly(ADP-ribose) polymerase-1 (PARP-1) can be activated under oxidative stress, which is the hallmark of parthanatos. Autophagy, which serves either a pro-survival or pro-death function, can also be stimulated by oxidative stress, but the role of autophagy and its relationship with parthanatos underlying this activation in the inner ear remains unknown. In this study, we established an oxidative stress model in vitro by glucose oxidase/glucose (GO/G), which could continuously generate low concentrations of H₂O₂ to mimic continuous exposure to H₂O₂ in physiological conditions, for investigation of oxidative stress-induced cell death mechanisms and the regulatory role of PARP-1 in this process. We observed that GO/G induced stria marginal cells (MCs) death via upregulation of PARP-1 expression, accumulation of polyADP-ribose (PAR) polymers, decline of mitochondrial membrane potential (MMP) and nuclear translocation of apoptosis-inducing factor (AIF), which all are biochemical features of parthanatos. PARP-1 knockdown rescued GO/G-induced MCs death, as well as abrogated downstream molecular events of PARP-1 activation. In addition, we demonstrated that GO/G stimulated autophagy and PARP-1 knockdown suppressed GO/G-induced autophagy in MCs. Interestingly, autophagy suppression by 3-Methyladenine (3-MA) accelerated GO/G-induced parthanatos, indicating a pro-survival function of autophagy in GO/G-induced MCs death. Taken together, these data suggested that PARP-1 played dual roles by modulating parthanatos and autophagy in oxidative stress-induced MCs death, which may be considered as a promising therapeutic target for ameliorating oxidative stress-related hearing disorders.

Keywords: Autophagy; Glucose oxidase; Marginal cells; Oxidative stress; PARP-1; Parthanatos.

Graphical abstract

Fig. 1

GO/G inhibited MCs viability and increased intracellular ROS production. (A) MCs were treated with GO/G (0–100 U/L GO/5 mM G) for 4 h. Cell viability was assessed by CCK-8 assay. The GO/G (0 U/L GO/5 mM G) group was considered as the control. The MCs' viability was strongly suppressed by higher concentrations than GO/G (60 U/L GO/5 mM G). (B) MCs were treated with GO/G (60 U/L GO/5 mM G) for different times and then stained with DCFH-DA for flow cytometry analysis. Statistical analysis showed the up-regulation of intracellular ROS production in a time-dependent manner following GO/G (60 U/L GO/5 mM G) treatment in MCs. Data are expressed as mean ± S.E.M. Representative results of at least three experiments are shown (** P < 0.01 and *** P < 0.0001).

Fig. 2

GO/G up-regulated the expression of parthanatos-related proteins. (A) Western blot analysis of PAR polymer levels in MCs following treatment with GO/G (60 U/L GO/5 mM G) for various times. β-actin was used as a loading control. (B) Immunofluorescent analysis of the accumulation of PAR polymer (red) in MCs upon treatment with GO/G (60 U/L GO/5 mM G) for 4 h by confocal microscopy. Nuclei were stained with DAPI (blue). Scale bars: 20 µm. (C) After GO/G (60 U/L GO/5 mM G) treatment for 4 h, mitochondrial, cytoplasmic and nuclear fractions were isolated from the treated cells and subjected to western blot for detection of the protein levels of PARP-1 and AIF. COX Ⅳ was used as a mitochondrial fraction loading control. β-actin and H3 were used as cytoplasmic and nuclear loading control respectively. (D) MCs were treated as above to detect the translocation of AIF from mitochondria to nucleus by confocal microscopy. AIF was shown in green and nuclei were stained with DAPI (blue). Scale bars: 10 µm. Data are expressed as mean ± S.E.M. Representative results of three experiments are shown (* P < 0.05 and ** P < 0.01).

Fig. 3

GO/G induced mitochondrial depolarization. (A) The cells were treated with GO/G (60 U/L GO/5 mM G) for various times, stained with JC-1 and analyzed by flow cytometry with quantification showing decreased ratios of red/green fluorescence intensity after GO/G (60 U/L GO/5 mM G) treatment. CCCP was used as a positive control. JC-1 monomer: green; J-aggregate: red. (B) MCs treated as above were labeled with fluorescent probe JC-1 to evaluate MMP changes in situ by confocal microscopy. Representative micrographs showed that the control cells displayed higher level of red fluorescence, whereas GO/G-treated cells exhibited higher level of green fluorescence. J-aggregate: red; JC-1 monomer: green. Scale bars: 20 µm. Data are expressed as mean ± S.E.M. Representative results of at least three experiments are shown (* P < 0.05 and ** P < 0.01).

Fig. 4

PARP-1 knockdown ameliorated GO/G-induced parthanatos. MCs were transfected with Ad-PARP1-RNAi or Ad-control-RNAi (Ad-Con) recombinant virus. 48 h after transfection, MCs were treated with GO/G (60 U/L GO/5 mM G) for 4 h. (A) Annexin V/PI staining and flow cytometry analysis of cell death. Percentages of PI positive cells are shown. (B) JC-1 staining and flow cytometry analysis of MMP. The ratios of red/green fluorescence intensity are shown. JC-1 monomer: green; J-aggregate: red. (C) The accumulation of PAR polymer was measured by western blot. The level of β-actin was used as a loading control. (D) The mitochondrial, cytoplasmic and nuclear levels of PARP-1 and AIF were analyzed by western blot. COX Ⅳ was used as a mitochondrial fraction loading control. β-actin and H3 were used as cytoplasmic and nuclear loading control respectively. (E) The detection of translocation of AIF from mitochondria to nucleus by confocal microscopy. AIF is shown in green and nuclei were stained with DAPI (blue). Scale bars: 10 µm. Data are expressed as mean ± S.E.M. Representative results of three experiments are shown (* P < 0.05 and ** P < 0.01).

Fig. 5

PARP-1 knockdown suppressed GO/G-induced autophagy. (A) Autophagy measurement by detection of the expression of p62 protein and the conversion of LC3-I to LC3-II by western blot. The MCs were treated with GO/G (60 U/L GO/5 mM G) for various times. The level of β-actin was used as a loading control. (B) Measurement of p62 protein levels and the conversion of LC3-I to LC3-II following GO/G (60 U/L GO/5 mM G) treatment with or without 3-MA pretreatment (5 mM) by western blot. β-actin was used as a loading control. (C) Detection of autophagosomes using immunostaining with LC3 antibody by confocal microscope. Cells were treated with GO/G (60 U/L GO/5 mM G) for 4 h with or without 3-MA pretreatment (5 mM). LC3 is shown in green and nuclei were stained with DAPI (blue). Scale bars: 50 µm. (D) Western blot analysis of protein levels of p62 and the conversion of LC3-I to LC3-II following GO/G treatment with or without PARP-1 knockdown. β-actin was used as a loading control. Data are expressed as mean ± S.E.M. Representative results of three experiments are shown (* P < 0.05 and ** P < 0.01).

Fig. 6

Autophagy inhibition sensitized MCs to GO/G-induced parthanatos. (A) Annexin V/PI staining and flow cytometry analysis of cell death. MCs were treated with GO/G (60 U/L GO/5 mM G) for 4 h with or without 3-MA pretreatment (5 mM). Percentages of PI positive cells are shown. (B) JC-1 staining and flow cytometry analysis of MMP. Cells were treated as above. The ratios of red/green fluorescence intensity are shown. JC-1 monomer: green; J-aggregate: red. (C) The accumulation of PAR polymer was measured by western blot. Cells were treated as above and subjected to western blot analysis. β-actin was used as a loading control. (D) The protein levels of PARP-1 and AIF in the mitochondrial, cytoplasmic and nuclear fractions of MCs were measured by western blot. Cells were treated as above and subjected to western blot analysis. COX Ⅳ was used as a mitochondrial fraction loading control. β-actin and H3 were used as cytoplasmic fraction loading control and nuclear fraction loading control respectively. Data are expressed as mean ± S.E.M. Representative results of three experiments are shown (* P < 0.05 and ** P < 0.01).