Nutrient deprivation regulates DNA damage repair in cardiomyocytes via loss of the base-excision repair enzyme OGG1

Abstract

Oxidative stress contributes to the pathogenesis of many diseases, including heart failure, but the role and regulation of oxidative DNA damage in many cases have not been studied. Here, we set out to examine how oxidative DNA damage is regulated in cardiomyocytes. Compared to normal healthy controls, human hearts in end-stage cardiomyopathy (EsCM) showed a high degree of DNA damage by histological evidence of damage markers, including 8-oxoG and γH2AX (8-oxoG: 4.7±0.88 vs. 99.9±0.11%; γH2AX: 2.1±0.33 vs. 85.0±13.8%; P<0.01) This raised the possibility that defective DNA repair may be partly responsible. Indeed, nutrient deprivation led to impaired base-excision repair (BER) in cardiomyocytes in vitro, accompanied by loss of the BER enzyme OGG1, while BER activity was rescued by recombinant OGG1 (control vs. nutrient deprived vs. nutrient deprived+OGG1; 100±2.96 vs. 68.2±7.53 vs. 94.0±0.72%; ANOVA, P<0.01). Hearts from humans with EsCM and two murine models of myocardial stress also showed a loss of OGG1 protein. OGG1 loss was inhibited by the autophagy inhibitor bafilomycin and in autophagy-deficient Atg5(-/-) mouse embryonic fibroblasts. However, pharmacological activation of autophagy, itself, did not induce OGG1 loss, suggesting that autophagy is necessary but not sufficient for OGG1 turnover, and OGG1 loss requires concurrent nutrient deprivation. Finally, we found that the role of autophagy in nutrient starvation is complex, since it balanced the positive effects of ROS inhibition against the negative effect of OGG1 loss. Therefore, we have identified a central role for OGG1 in regulating DNA repair in cardiomyopathy. The manipulation of OGG1 may be used in future studies to examine the direct contribution of oxidative DNA damage to the progression of heart failure.

Figure 1

DNA damage and activation of the DNA damage response (DDR) in human end-stage cardiomyopathy. A) Representative images from immunohistochemical detection of γH2AX, p-ATM, NBS1, and 8-oxoG in healthy control (CTRL) and EsCM human LV cardiac tissue (brown-stained nuclei). H3K36me3 antibody was used as a positive control. B) Quantification was performed by counting positively stained nuclei from ≥100 nuclei in each cardiac section. **P < 0.01, ***P < 0.001; Mann-Whitney U test. C) Western blot analysis of human cardiac lysates showed DDR and ATM activation by ATM phosphorylation (p-ATM) and significant levels of ATM/ATR substrate phorsphorylation in EsCM but not CTRL. Equal protein loading was demonstrated by blotting for sarcomeric actin.

Figure 2

Nutrient deprivation regulates DNA repair. A) DNA DSB damage and repair was tracked by Western blot for γH2AX following the recovery from a stimulus with 100 μM tert-butyl hydroperoxide (BHP). Lysates were taken from HL1 myocytes cultured either in normal or SGF conditions before and during the recovery from the stimulus with BHP. B) An ex vivo assay of BER activity was set up using a fluorescent 8-oxoG-containing molecular beacon. BER activity was determined using lysates from HL1 myocytes cultured in SGF over the time course indicated. Decreased BER activity in SGF-treated cells was inhibited when these were cultured with ethyl pyruvate, a cell-permeable supplemental source of ATP. Rescue of BER activity reached similar levels as found in cells in control serum-containing medium. **P < 0.01, *** P < 0.001; Student’s t test.

Figure 3

Reduced OGG1 protein abundance regulates DNA damage accumulation. A) Rat neonatal ventricular myocytes were cultured in 5% myocyte medium (control; lane 1) or in control medium plus 1 μM AngII (lane 2), 100 μM PE (lane3), 2 μg/ml LPS (lane 4), or SGF medium (lane 5) for 72 h. ROS production was assessed using DCFDA probe, and 8-oxoG was assessed by immunofluorescence using DAPI as a nuclear counterstain. *P < 0.05; ANOVA. B) Lysates from myocytes treated with the same conditions (lanes 1–5) were assessed at 48 h by Western blot for OGG1, PARP1, and APE1 protein abundance. RhoGDI was used to show equal protein loading. C) As in Fig. 2B, 8-oxoG BER ex vivo excision was assayed using lysates from myocytes treated for 48 h in control or SGF conditions, with and without 10 U of recombinant human OGG1 supplemented in the ex vivo reaction. **P < 0.01; ANOVA. D, E) OGG1 protein abundance in human control (CTRL) and EsCM cardiac tissue was assessed by immunohistochemistry (D) and Western blot analysis (E). Other repair enzymes (PARP1, XRCC1, and APE1) were also assessed by Western blot to highlight the significance of OGG1 in this context. Sarcomeric actin was used to show equal protein loading. **P < 0.01; Student’s t test. F) OGG1 protein abundance was assessed by Western blot using cardiac lysates from mice that had undergone transverse aortic constriction (pressure-overload cardiomyopathy) and 48 h in vivo starvation, compared to sham operation and normal fed conditions, respectively.

Figure 4

Autophagy regulates OGG1 protein abundance under conditions of nutrient deprivation. A) Western blot analysis of OGG1 and APE1 protein abundance in myocytes cultured in SGF conditions for 48 h, with or without 50 μM calpeptin or 100 nM bafilomycin, or control medium with or without 100 nM rapamycin. B) Culture in SGF conditions for 48 h induced OGG1 protein loss in Atg5^+/+ MEFs, which was inhibited by 100 nM bafilomycin, and SGF-induced OGG1 loss was not observed in autophagy-deficient Atg5^−/− MEFs. RhoGDI was used to show equal protein loading. C) OGG1 protein half-life was prolonged in Atg5^−/− MEFs compared to Atg5^+/+ MEFs, as determined by treatment with the translation inhibitor cycloheximide (100 μg/ml).

Figure 5

Autophagy is adaptive and essential under conditions of nutrient deprivation. A) Autophagy inhibition with bafilomycin decreased cell viability under SGF conditions. *P < 0.05. B) Autophagy inhibition with bafilomycin up-regulates ROS production under SGF conditions. **P < 0.01. C) Western blot for γH2AX showed that genotoxic stress under SGF conditions was augmented by the presence of autophagy inhibition with bafilomycin.

Figure 6

Model of 8-oxoG accumulation in cardiac stress. 8-oxoG accumulation is induced by nutrient deprivation through OGG1 protein loss that is dependent on autophagy, but autophagy also promotes nutrient regeneration and inhibits ROS production.