The circadian gene Rev-erbα improves cellular bioenergetics and provides preconditioning for protection against oxidative stress

Abstract

Diurnal oscillations in the expression of antioxidant genes imply that protection against oxidative stress is circadian-gated. We hypothesized that stabilization of the core circadian gene Rev-erbα (Nr1d1) improves cellular bioenergetics and protects against nutrient deprivation and oxidative stress. Compared to WT, mouse lung fibroblasts (MLG) stably transfected with a degradation resistant Rev-erbα (Ser(55/59) to Asp; hence referred to as SD) had 40% higher protein content, 1.5-fold higher mitochondrial area (confocal microscopy), doubled oxidative phosphorylation by high-resolution respirometry (Oroboros) and were resistant to glucose deprivation for 24h. This resulted from a 4-fold reduction in mitophagy (L3CB co-localized with MitoTracker Red) versus WT. Although PGC1α protein expression was comparable between SD and WT MLG cells, the role of mitochondrial biogenesis in explaining increased mitochondrial mass in SD cells was less clear. Embryonic fibroblasts (MEF) from C57Bl/6-SD transgenic mice, had a 9-fold induction of FoxO1 mRNA and increased mRNA of downstream antioxidant targets heme oxygenase-1 (HO-1), Mn superoxide dismutase and catalase (1.5, 2 fold and 2 fold respectively) versus WT. This allowed the SD cells to survive 1h incubation with 500 µM H2O2 as well as 24h of exposure to 95% O2 and remain attached whereas most WT cells did not. These observations establish a mechanistic link between the metabolic functions of Rev-erbα with mitochondrial homeostasis and protection against oxidative stress.

Keywords: Bioenergetics; Circadian; Energy metabolism; Mitochondria; NR1D1; Oxidative stress; Peroxisome proliferator‐activated receptor gamma coactivator 1‐alpha (PGC‐1α)(PPARGC1A); Preconditioning.

Figure 1. Rev-erbα stabilized (SD) cells overexpress Rev-erbα protein and downregulate the key target gene Bmal1

(A) Densitometric analysis of Rev-erbα protein expression of SD or WT mouse lung fibroblasts (MLG) [normalized to GAPDH as loading control] (n=4 independent experiments) (B) mRNA levels (n=3 independent experiments) of core circadian genes Bmal1 (primary target of Rev-erbα), Rev-erbβ and Rorα in WT and SD MLG cells. *p<0.05 by t-test. Data expressed as mean ± SEM.

Figure 2. Size and protein content of Rev-erbα stabilized (SD) MLG cells

(A) Representative images depicting the fluorescence intensity of Mito Tracker Deep Red, actin (in green) and nuclear staining with DAPI (blue) of WT and SD MLG cells. Scale bar at 10μm. (B) Quantification of cell area. (n=5 independent experiments; 3 images/experiment for each genotype) (C) Relative amount of total protein content per cell determined by Bradford assay (n=3 independent experiments); Equal number of cells were processed for each cell type and normalized to WT cells. Data are mean ± SEM. **p<0.001 by t-test.

Figure 3. Survival of Rev-erbα stabilized (SD) and WT MLG cells under nutrient deprivation

(A) (Upper panel) Representative phase microscopy images depicting cell growth and confluence of SD and WT MLG. Cells were grown in standard media overnight to 50–70% confluence, media removed, washed with PBS and then incubated in glucose free media (middle panel), or glucose, pyruvate and glutamine free media (Right panel) for 8–24 hrs. Control cells (Left panel; n=4). Images were taken at 200× magnification. Lower Panel: Bar graph represents number of healthy/live cells under nutrient deprivation normalized to cells that were not exposed to nutrient deprivation. The cell densities were comparable at the beginning of the nutrient deprivation conditions between the two genotypes. (B) Relative mRNA levels of hexokinase-2(HK-2) and phosphofructokinase (PF3Kb), the rate limiting enzymes of glycolysis, in WT and SD MLG cells. All data are mean ± SEM. *p<0.05 by t-test; **p<0.001 by t-test.

Figure 4. Mitochondrial respiration in MLG with Rev-erbα stabilized (SD MLG) and WT MLG

(A) Mitochondrial respiration of intact WT and SD MLG cells was determined using high-resolution respirometry. The graph depicts the plot comparing the oxygen consumption rate (OCR or Flux) per 10⁶ cells from a representative experiment. (B) Averaged values of basal respiratory capacity, Oligo/leak and FCCP/maximal respiratory capacity (MRC) results of mitochondrial respiration from above experiments (n=4 independent determinations) for WT and SD MLG (mouse lung fibroblasts) cells. (C) Averaged values of basal respiratory capacity, Oligo/leak and FCCP/MRC results of mitochondrial respiration from above experiments (n=4 independent determinations) for WT and SD MLG (mouse lung fibroblasts) cells after normalization to Citrate Synthase specific activity. (D)Averaged values of Coupling efficiency [defined as the (Basal-Oligo/leak)/Basal], respiration driving ATP synthesis/Oligomycin sensitive OCR [defined as Basal-Oligo/Leak OCR] and Spare respiratory capacity [Defined as FCCP/MRC-Basal OCR] in SD versus WT cells derived from the hi-resolution respirometry experiments. All OCRs were normalized for the Specific Citrate Synthetase activity. (n=4 independent determinations) (E) Mitochondrial content (Mito Tracker Green) Flowcytometry was used to measure the fluorescence generated by stained mitochondria after normalization against WT cells. Values are expressed as the relative fold change in median fluorescence intensity (MFI) normalized to that of WT MLG cells. (n=5 independent determinations) (F) Confocal microscopic images depicting the fluorescence intensity of Mito Tracker Deep Red, actin (in green) and nuclear staining with DAPI (blue) of single WT and SD mouse lung fibroblasts at 60× magnification. Scale bar =10μm (G) Mitochondrial area/cell was determined by analyzing 12 fields (n=5 with 2 independent experiments performed under conditions similar to the mitochondrial respiration experiments); each field containing 11–16 cells/field were analyzed to calculate mitochondrial area per cell. All data are presented as mean ± SEM. *p<0.05 by t-test or one-way ANOVA with Bonferroni correction for multiple comparisons.

Figure 5. Mitophagy is reduced in Rev-erbα stabilized (SD) compared with WT MLG

(A) Confocal microscopic images depicting the fluorescence intensity of MitoTracker Deep Red (red), autophagosome marker LC3B (green), co-localization of autophagosomes and mitochondria from individual z-planes, compressed to 2D for display (yellow; highlighted with arrows). Scale bar =10μm (B) Left panel: Quantification of endogenous LC3B-II and relative co-localization with MitoTracker Deep Red with balfilomycin incubation (left panel) and with/without bafilomycin (right panel). Total mitophagy was normalized to number of cells in the field to calculate mitophagy area per cell. Right panel: Quantification of endogenous LC3B-II and relative co-localization with MitoTracker Deep Red with or without balfilomycin incubation. Total mitophagy was normalized to number of cells in the field to calculate mitophagy area per cell. *p<0.05 by t-test; n= 3 independent experiments. (C) Representative immunoblot of Parkin in WT and stabilized Rev-erbα (SD) MLG cells (n=3 independent determinations). Densitometric analysis of Parkin protein normalized to WT MLG cells and GAPDH as a loading control. (D) Representative immunoblots of beclin-1 and LC3b protein expression in whole cell lysates from SD or WT mouse lung fibroblasts (MLG) [normalized to calnexin and commassie blue as loading control]. Densitometric analysis of beclin-1 and LC3b normalized to normalized to calnexin and commassie blue as loading controls respectively. (n=4 independent determinations) All data presented as the mean ± SEM. *p<0.05 by t-test

Figure 6. Cells with stabilized Rev-erbα are protected against oxidative stress

(A) Left panel: Fluorescence intensity of MitoSOX expressed as relative fold change in median fluorescence intensity (MFI) normalized to WT MLG cells. (n=3 independent experiments) Right Panel: Fluorescence intensity of TMRE in expressed as relative fold change in median fluorescence intensity (MFI) normalized to WT MLG cells. (n=3 independent experiments) (B) Left panel: FoxO-1 gene expression in mouse embryonic fibroblasts (MEF) cells (n=4 independent determinations) with stabilized Rev-erbα (SD) or control (WT) cells. Right panel: Representative immunoblot of FoxO-1 and phospho-FoxO1 (p-FoxO1; phosphorylated at Serine 256) in WT and stabilized Rev-erbα (SD) MEF cells (n=3 independent determinations) with stabilized Rev-erbα (SD) or control (WT) MEF cells. Bar graph representing the relative levels of p-FoxO1 protein as a fraction of total FoxO1 protein. Calnexin immunoblot is shown as a loading control. (C) Hmox-1, MnSOD and Catalase gene expression in MLG cells (n=3 independent determinations) with stabilized Rev-erbα (SD) or control (WT) cells. (D) Upper Panel Phase microscopy images depicting cell survival in stabilized Rev-erbα (SD) and WT MEF after oxidative stress induced by either 250 μM or 500 μM hydrogen peroxide (middle and right panel n=3–4) for 1 hr versus control cells (Left panel; n=3 independent experiments). Scale bar = 100μm. Lower Panel: Bar graph represents number of healthy/live cells in each group normalized to cells that were not exposed to hydrogen peroxide for each genotype. The cell densities were comparable at the beginning of the exposure to H₂O₂ between the two genotypes. (E) Images depicting oxidative damage to DNA (using 8-oxo-DG immunostaining in green with nuclear co-localization with DAPI in blue). Bottom panel shows Rev-erbα (SD) and the top panel depicts WT MEF after oxidative stress induced by 500µM hydrogen peroxide for 1hr or under control conditions. (F) Cell death quantified by Trypan Blue exclusion assay in SD and WT MLG cells exposed to hyperoxia (95%O₂/5%CO₂) or to room air (21%O₂ /5%CO₂). Experiments were repeated in triplicate for five separate determinations. All data presented as the mean ± SEM. *p<0.05 by t-test. **p<0.001 by t-test.

Figure 7. Stabilization of Rev-erbα increases mitochondrial respiration in a Pgc1α independent manner

(A) Relative Pgc1α mRNA levels in WT or SD MLG cells (n=3 independent determinations). (B) Representative immunoblot of PGC1α in WT and stabilized Rev-erbα (SD) MLG. Bar graph representing quantification of protein expression (normalized to WT cells and calnexin as loading control) All data presented as the mean ± SEM. *p<0.05 by t-test and Mann-Whitney for (n=4 independent determinations)