p53 Regulates oxidative stress-mediated retrograde signaling: a novel mechanism for chemotherapy-induced cardiac injury

Abstract

The side effects of cancer therapy on normal tissues limit the success of therapy. Generation of reactive oxygen species (ROS) has been implicated for numerous chemotherapeutic agents including doxorubicin (DOX), a potent cancer chemotherapeutic drug. The production of ROS by DOX has been linked to DNA damage, nuclear translocation of p53, and mitochondrial injury; however, the causal relationship and molecular mechanisms underlying these events are unknown. The present study used wild-type (WT) and p53 homozygous knock-out (p53(-/-)) mice to investigate the role of p53 in the crosstalk between mitochondria and nucleus. Injecting mice with DOX (20 mg/kg) causes oxidative stress in cardiac tissue as demonstrated by immunogold analysis of the levels of 4-hydroxy-2'-nonenal (4HNE)-adducted protein, a lipid peroxidation product bound to proteins. 4HNE levels increased in both nuclei and mitochondria of WT DOX-treated mice but only in nuclei of DOX-treated p53((-/-)) mice, implicating a critical role for p53 in causing DOX-induced oxidative stress in mitochondria. The stress-activated protein c-Jun amino-terminal kinase (JNKs) was activated in response to increased 4HNE in WT mice but not p53((-/-)) mice receiving DOX treatment, as determined by co-immunoprecipitation of HNE and pJNK. The activation of JNK in DOX treated WT mice was accompanied by Bcl-2 dissociation from Beclin in mitochondria and induction of type II cell death (autophagic cell death), as evidenced by an increase in LC3-I/LC-3-II ratio and γ-H2AX, a biomarker for DNA damage. The absence of p53 significantly reduces mitochondrial injury, assessed by quantitative morphology, and decline in cardiac function, assessed by left ventricular ejection fraction and fraction shortening. These results demonstrate that p53 plays a critical role in DOX-induced cardiac toxicity, in part, by the induction of oxidative stress mediated retrograde signaling.

Figure 1. Absence of p53 selectively reduces DOX-induced oxidative damage in the mitochondria.

A. Representative immunogold electron micrographs using antibody against 4HNE protein adducts in cardiomyocytes. Electron dense beads indicate positive staining for 4HNE protein adducts (arrow). Left ventricular tissues from WT (a) and p53^(−/−) (b) mice treated with saline demonstrate low labeling of 4HNE in mitochondria (M) and myofilaments (Myo) (a and b). Significant increase in labeling of 4HNE protein adducts was observed in both mitochondria and myofilaments of the WT mice hearts (c) but not in the p53^(−/−) mice (d) treated with DOX. Scale bar, 1 µm. B, C, and D. 4HNE-immunoreactive protein adducts were quantified in cardiomyocyte cytoplasm (B), mitochondria (C) and nuclei (D) for both WT and p53^(−/−) mice treated with saline and DOX. Labeling density is expressed in gold beads/µm². All graphs represent the Mean ± SEM for each group. *p<0.05 when compared with saline treated mice of the same genotype, ^# p<0.05 when compared with WT mice similarly treated.

Figure 2. DOX induces HNE adduction with pJNK and Bcl-2 phosphorylation, and triggers autophagic response.

A. PhosphoJNK was detected by immunoprecipitation using a polyclonal goat anti-HNE antibody followed by Western blot analysis with a rabbit polyclonal Thr183/Tyr185 phosphorylation specific JNK antibody. Quantitative analysis represents the Mean ± SEM (n = 5) in each group; *p<0.001 as compared to other groups. B. Western blot analysis of p-Bcl2 in cardiac mitochondria from WT and p53^(−/−) mice treated with saline or DOX and quantitative analysis represents the Mean ± SEM n = 5 in each group; *p<0.001 as compared to other groups. C. Western blot analysis of Beclin1 in cardiac mitochondria from WT and p53^(−/−) mice treated with saline or DOX and quantitative analysis represents the Mean ± SEM n = 5 in each group; *p<0.001 as compared to other groups. D. Co-immunoprecipitation of Beclin1 with Bcl-2 in cardiac mitochondria from WT and p53^(−/−) mice treated with saline or DOX and quantitative analysis represents the Mean ± SEM n = 5 in each group; *p<0.01 as compared to other groups.

Figure 3. p53 triggers autophagy and cell death.

A. Biochemical marker for cardiac injury by autophagy. Western blot analysis of autophagy marker (LC3) expression in heart tissue homogenates from WT and p53^(−/−) mice treated with saline or DOX. Quantitative analysis represents the Mean ± SEM n = 5 in each group; ^#p<0.001 as compared to other groups. B. Western blot analysis of gamma H2AX marker for cell death in heart tissue nuclear extracts from WT and p53^(−/−) mice treated with saline or DOX. Quantitative analysis represents the Mean ± SEM n = 5 in each group. ^#p<0.001 as compared to other groups.

Figure 4. p53 contributes to DOX-induced cardiac injury.

A. Representative electron micrographs demonstrating cardiomyocyte injury and morphometric quantification of subcellular injury in cardiomyocytes in WT and p53^(−/−) mice. WT (a) and p53^(−/−) (b) mice treated with saline demonstrated normal ultrastructure of heart muscle with cardiomyocytes showing numerous mitochondria (M), prominent myofilaments (Myo) and lipids (Lip). WT (c) and p53^(−/−) (d) mice treated with DOX demonstrated the following pathologic changes: lysosomal degradation of mitochondria (asterisk), mitochondria with loss of cristae (arrow), peri-mitochondrial swelling (double arrow), disruption of mitochondrial membranes (arrowhead), and mitochondrial degeneration (d); scale bar, 1 µm. B. Pathologic changes were quantified for saline and DOX treated mice in the mitochondria and total cellular area (excluding nuclei). Quantification of damage for each specific compartment is expressed in the ratio of damaged area versus the total area. All graphs represent the Mean ± SEM for each group. *p<0.05 when compared with saline treated mice of the same genotype, and ^#p<0.05 when compared with WT mice treated with DOX. n = 6 or 7. C & D. Left ventricular function, assessed by percentage of ejection fraction (LV%EF) (C) and fractional shortening (LV%FS) (D), is expressed in percentage of change from basal levels caused by saline and DOX three days after injections for WT and p53^(−/−) mice. All bar graphs represent the Mean ± SEM for each group. ^#p<0.05 when compared to all three other groups. n = 6–11.

Figure 5. p53 enhances DOX-induced cardiac injury, in part, via enhancement of oxidative stress in mitochondria.

The absence of p53 selectively prevents DOX-mediated increase in oxidative stress indicators, including 4HNE-adducted proteins in mitochondria but not in the nucleus. DOX-induced p53-dependent increased oxidative stress in mitochondria is associated with sustained activation of JNK1 and subsequent phosphorylation of Bcl-2 and release of beclin from the Bcl-2-beclin complex resulting in detrimental effect of autophagy (cellular injury).