Sod1 Loss Induces Intrinsic Superoxide Accumulation Leading to p53-Mediated Growth Arrest and Apoptosis

Oxidative damages induced by a redox imbalance cause age-related changes in cells and tissues. Superoxide dismutase (SOD) enzymes play a major role in the antioxidant system and they also catalyze superoxide radicals (O2•−). Since the loss of cytoplasmic SOD (SOD1) resulted in aging-like phenotypes in several types of mouse tissue, SOD1 is essential for the maintenance of tissue homeostasis. To clarify the cellular function of SOD1, we investigated the cellular phenotypes of Sod1-deficient fibroblasts. We demonstrated that Sod1 deficiency impaired proliferation and induced apoptosis associated with O2•− accumulation in the cytoplasm and mitochondria in fibroblasts. Sod1 loss also decreased the mitochondrial membrane potential and led to DNA damage-mediated p53 activation. Antioxidant treatments effectively improved the cellular phenotypes through suppression of both intracellular O2•− accumulation and p53 activation in Sod1-deficient fibroblasts. In vivo experiments revealed that transdermal treatment with a vitamin C derivative significantly reversed the skin thinning commonly associated with the upregulated p53 action in the skin. Our findings revealed that intrinsic O2•− accumulation promoted p53-mediated growth arrest and apoptosis as well as mitochondrial disfunction in the fibroblasts.


Introduction
Reactive oxygen species (ROS) are mainly generated from mitochondrial respiration and they non-specifically oxidize cellular molecules including proteins, nucleic acid, and lipids, resulting in oxidative damage in organisms [1]. Redox balance is physiologically regulated through the production and degradation of ROS by antioxidant systems to protect cells from oxidative damage. Extrinsic excess ROS induces the DNA damage response (DDR) associated with oxidative DNA damage and promotes a canonical ATM-p53 cascade that regulates the cell fate [2]. ROS also dysregulates mitochondrial function through a reduction of membrane potential and respiration [3]. In this context, maintenance of redox balance in cells plays an important role in the determination of cellular fate and function, including apoptosis, cell cycle arrest, differentiation, and energy metabolism [4].
In the present study, we investigated the cellular phenotypes of Sod1 −/− fibroblasts to clarify the biological significance of Sod1 and the pathophysiological role of intracellular O 2 ·− . We also investigated the involvement of the DDR and p53 activation under an intrinsic O 2 ·− accumulation.
Finally, we have discussed the anti-aging effect of an antioxidant administered both in vitro and in vivo.

Sod1 Deficiency Induced Apoptotic Cell Death with Increased Superoxide Accumulation in Fibroblasts
In order to investigate the biological significance of the SOD1 enzyme in cells, we analyzed the cellular phenotypes of Sod1-deficient primary dermal fibroblasts. Western blot analysis revealed the complete loss of the SOD1 protein in Sod1 −/− cells ( Figure 1A). Interestingly, the concentration of the SOD2 protein, an alternative intracellular SOD localized in mitochondria, remained unchanged in Sod1 −/− cells, suggesting that SOD1 loss did not induce the compensatory expression of SOD2 protein in the cells ( Figure 1A). Likewise, expression levels of other antioxidant enzymes, including glutathione peroxidase 1 and catalase, were not upregulated in Sod1 −/− cells (data not shown). In cell culture experiments, Sod1 −/− fibroblasts showed the marked loss of cell viability under a 20% O 2 concentration ( Figure 1B). We next analyzed the incorporation of BrdU to measure the proliferative ability of the Sod1 −/− fibroblasts. As shown in Figure 1C, Sod1 loss significantly impaired the incorporation of BrdU at culture day 2, indicating the disturbance of cell proliferation. Furthermore, Sod1 depletion markedly increased the expression of cleaved caspase3 ( Figure 1D) and annexin V positive cells ( Figure 1E,F), indicating the induction of apoptotic cell death. These results demonstrated that Sod1 deficiency induced proliferative decline and apoptosis in dermal fibroblasts. These data indicate the mean ± SD; ** p < 0.01, *** p < 0.001.  Figure 2A). Interestingly, MitoSOX staining, which is a specific fluorescent dye for O 2 ·− in mitochondria, also revealed a significant, 4-fold enhancement in the mitochondrial O 2 ·− level in Sod1 −/− compared to Sod1 +/+ fibroblasts ( Figure 2B), These results suggested that SOD1 regulates the O 2 ·− balance in both the cytoplasm and the mitochondria. . These data indicate the mean ± SD; * p < 0.05, ** p < 0.01.

Sod1 Loss Caused p53 Upregulation Associated with Mitochondrial Dysfunction in Fibroblasts
Since mitochondrial ROS induces the loss of mitochondrial membrane potential (ΔΨm) [3], we measured ΔΨm using a JC-1 dye in Sod1 −/− fibroblasts. As expected, Sod1 −/− fibroblasts showed a 2.2-fold increase in the number of mitochondria with low ΔΨm ( Figure 3A,B). Since decreased ΔΨm induces apoptosis [17], our findings suggested that O 2 ·− accumulation in mitochondria resulting from Sod1 deficiency results in apoptosis through mitochondrial dysfunction. Tumor suppressor p53 plays a crucial role in various cellular functions such as apoptosis, cell cycle arrest, energy metabolism, and senescence [4]. Since DNA damage caused by excess irradiation and ROS stimulates upregulation and phosphorylation of p53 via DDR resulting in apoptosis [18], we analyzed the p53 level in Sod1 −/− fibroblasts. Western blot analysis clearly demonstrated markedly increased protein levels and phosphorylation at Ser 18 of p53 in Sod1 −/− fibroblasts ( Figure 3C). Quantitative PCR analysis revealed that Sod1 −/− deficiency had a tendency to increase p53 mRNA levels, but not significantly ( Figure 3D), suggesting that p53 upregulation is not only regulated by mRNA levels but may be due to p53 stabilization in Sod1 −/− fibroblasts. Furthermore, we found phosphorylated H2AX at Ser 139 (γH2AX), a DNA damage marker, and identified the upregulation of p21, a target gene of p53 ( Figure 3C). These data indicated that the DNA damage caused by Sod1 deficiency also induced a proliferative defect and apoptosis via p53 activation. (B) The relative percentage of mitochondria with low ΔΨm in Sod1 −/− and Sod1 +/+ fibroblasts (n = 4); (C) Western blotting of DNA damage response proteins such as p53, p53 phosphorylation at Ser 18 , γH2AX and p21 in Sod1 −/− and Sod1 +/+ fibroblasts; (D) p53 mRNA expression was analyzed by quantitative PCR in Sod1 −/− and Sod1 +/+ fibroblasts (n = 4). These data indicate the mean ± SD; * p < 0.05.

A Vitamin C Derivative Rescued Viability of Sod1-Deficient Fibroblasts through a Suppression of O 2 ·− Generation and p53 Upregulation
In vitro data indicate that the suppression of intracellular O 2 ·− generation and p53 activation by To investigate the downstream molecular events by Sod1 deficiency, we assessed the p53 expression levels in APPS-treated Sod1 −/− fibroblasts. Treatment with APPS significantly inhibited p53 upregulation and phosphorylation at Ser 18 ( Figure 4C). We also cultured Sod1 −/− fibroblasts in the presence of another antioxidant, N-acetyl cysteine (NAC). As expected, treatment with NAC significantly accelerated cell viability and suppressed the intracellular O 2 ·− level in Sod1 −/− fibroblasts ( Figure 4D,E). Furthermore, to exclude the direct action of APPS on p53 expression, we investigated the inhibition ability of APPS on DNA damage-induced p53 upregulation. When NIH3T3 cells were cultured with camptothecin (CPT), a DNA damage-inducer, p53 protein was significantly upregulated in a dose-dependent manner ( Figure 4F). The APPS treatment, however, failed to inhibit p53 upregulation by CPT treatment, indicating that APPS is not able to suppress p53 expression in response to DNA damage. These results revealed that a VC derivative effectively normalized the cellular viability, O 2 ·− accumulation, and p53 upregulation via the direct anti-oxidant activity in

A Vitamin C Derivative Treatment Reversed Skin Atrophy Induced by Sod1 Loss in Vivo
We have previously reported that transdermal treatment with APPS reverses skin thinning in Sod1 −/− mice [8,19]. Treatment with APPS also significantly reduced lipid peroxidation in the skin of Sod1 −/− mice [19]. In the present study, we confirmed the beneficial effects of transdermal APPS treatment on skin atrophy in Sod1 −/− mice ( Figure 5A). In order to elucidate the rescue mechanism of APPS on skin pathology in Sod1 −/− mice, we measured p53 expression using quantitative PCR. As shown in Figure 5B  . These data indicate the mean ± SD; * p < 0.05.

Discussion
In the present study, we demonstrated that Sod1 −/− fibroblasts showed decreased proliferative ability and increased apoptosis ( Figure 1). Furthermore, Sod1 loss induced O 2 ·− accumulation in both the cytoplasm and the mitochondria (Figure 2). Okado-Matsumoto reported that Sod1 is distributed in the cytoplasm and in the intermembrane space of mitochondria [5] and Muller et al. found that complex III released O 2 ·− to both sides of the inner membrane in mitochondria isolated from skeletal muscles [20]. Recently, Jang et al. also reported increased ROS generated from isolated mitochondria in the skeletal muscle of Sod1 −/− mice [21], suggesting that Sod1 physiologically regulates not only cytoplasmic O 2 ·− generation, but also mitochondrial O 2 ·− release. In our preliminary experiments, we demonstrated that Sod2 −/− fibroblasts showed an increased O 2 ·− accumulation in the mitochondria and in the cytoplasm, resulting in decreased proliferation. Interestingly, Sod2 deficiency did not induce apoptosis in spite of increased O 2 ·− accumulation in the fibroblasts. Taken together, these results suggested that Sod1 might protect from the apoptosis induced by intracellular O 2 ·− accumulation in fibroblasts.
Since oxidative DNA damage induced by such irritants as irradiation and chemicals promotes p53 upregulation and phosphorylation via a canonical ATM-p53 cascade [2], we investigated DDR such as phosphorylated H2AX (γH2AX) and p53 activation. As expected, Sod1 −/− fibroblasts showed increased γH2AX with upregulation and phosphorylation of p53 at Ser 18 ( Figure 3C), whose residue is phosphorylated by ATM [2]. Moreover, p21, a p53 target gene, was significantly upregulated in Sod1-deficient fibroblasts ( Figure 3C), thus suggesting that intracellular O 2 ·− accumulation stimulated the canonical DDR cascade including p53 activation leading to growth arrest and apoptosis in Sod1 −/− fibroblasts. Accumulating evidence demonstrates that cellular p53 levels are primarily regulated by ubiquitin-mediated proteasomal degradation [22]. ATM phosphorylates both p53 at Ser 15 (mouse Ser 18 ) and MDM2, an endogenous E3-ligase for p53, at Ser 395 , resulting in an impairment of MDM2-mediated p53 degradation [22]. Since ROS-induced DDR promotes ATM activity, ATM-mediated MDM2 phosphorylation may stabilize p53 levels in Sod1 −/− cells ( Figure 3C). In contrast, Gajjar et al. recently reported that p53 mRNA-MDM2 interaction controls MDM2 nuclear trafficking and p53 activation following DNA damage. They demonstrated that ATM-dependent phosphorylation of MDM2 at Ser 395 resulted in its recruitment to p53 mRNA, thereby stimulating p53 upregulation [23,24]. Thus, DDR-mediated p53 mRNA-MDM2 interaction may increase p53 mRNA in Sod1 −/− skin tissues ( Figure 5B). Further analyses are therefore needed to clarify the molecular mechanism of the upregulation of p53 in Sod1-deficient mice. Interestingly, Sod1 −/− fibroblasts also showed mitochondrial O 2 ·− accumulation and low ΔΨm ( Figure 3A,B) [26]. Since both malic enzymes are important for NADPH production, lipogenesis, and glutamate metabolism, the upregulation of p53 may also directly impair mitochondrial function via ME1 and ME2 repression leading to ΔΨm decline. In a mouse model, p53 activation (p53m allele) by genetic engineering induces accelerated aging including skin atrophy [27]. Gannon et al. reported that Mdm2 loss in keratinocytes induced p53 upregulation resulting in epidermal stem cell senescence and premature aging phenotypes in mice [28]. In skin tissue, p53 also plays a crucial role in cellular fate and homeostasis. In the present study, we demonstrated that Sod1 loss induced p53 upregulation in the fibroblasts and the skin due to intracellular O 2 ·− accumulation. We also demonstrated that antioxidants effectively rescued O 2 ·− -p53 mediated cellular and skin pathologies in vitro and in vivo. These data indicate the possibility that antioxidant treatment is a promising strategy to delay skin aging by remarkably reducing the p53 upregulation in the skin.
It is widely known that VC is a soluble vitamin and an essential cofactor for post-translational modifications in collagen formation [29]. We previously reported that the oral administration of VC significantly improved low-turnover osteoporosis [9] as well as reduced behavioral deficits in the AD model mice [30]. However, the oral administration of VC failed to improve skin atrophies of Sod1 −/− mice (unpublished results), indicating the low bioavailability and stability of VC in the skin tissues. Since VC exhibits both low stability and liposolubility, it is difficult to transdermally absorb into the skin. Actually, we showed that a transdermal treatment of VC itself failed to improve the skin atrophy of Sod1 −/− mice [19]. To increase the stability and liposolubility of VC, various VC derivatives were developed for dermatologic application. One of the VC derivatives, APPS, is conjugated to a phosphate group and a long hydrophobic chain to enhance its stability and liposolubility. Transdermal treatment with APPS effectively normalized skin thickness in Sod1 −/− mice through the suppression of oxidative damage and p53 upregulation ( Figure 5). Treatment with APPS in wild-type mice showed no adverse effects in the skin tissues. Therefore, VC derivatives, including APPS, are powerful anti-aging agents for skin through an antioxidant effect.

Animals
Sod1 −/− mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). All of the genotypes of Sod1 mice were assessed by PCR using genomic DNA isolated from the tail tip as described previously [9]. The animals were housed under a 12 h light/dark cycle and were fed ad libitum. The experimental procedures were approved by the Animal Care and Use Committee of Chiba University.

Western Blotting
The

Flow Cytometry
The accumulation of intracellular O 2 ·− was detected using dihydroethidium (DHE, Life Technologies Corporation, Gaithersburg, MD, USA) and MitoSOX (Life Technologies Corporation), which are specific detectors of the O 2 ·− concentration in the cytoplasm and mitochondria, respectively [31]. To measure ΔΨm, the cells were stained with JC-1 dye, a ΔΨm probe (Life Technologies Corporation). The cells were incubated with 10 μM DHE, 5 μM MitoSOX, or 10 μM JC-1 for 30 min at 37 °C. Following incubation, the cells were trypsinized and resuspended in PBS. Apoptosis was measured using an FITC Annexin V Apoptosis Detection Kit I (BD Biosciences, San Jose, CA, USA) according to the manufacturer's instructions. The fluorescence intensities were assessed using a flow cytometer (BD FACSCanto™ II, BD Biosciences, San Jose, CA, USA).

Histology
The fibroblasts were pre-incubated for 24 h with 1 mM N-acetylcysteine (NAC). The accumulation of intracellular O 2 ·− was detected using DHE. The cells were incubated with 10 μM DHE for 30 min at 37 °C. Following incubation, the cells were washed three times with buffer and then photographed using a Leica DFC300 FX camera (Leica Microsystems, Wetzlar, Germany) and the software application, Leica IM50 v4.0. For the histological morphology analysis, the skin specimens from back tissues of Sod1 −/− mice (5 months old) were dissected, fixed overnight in a 20% formalin neutral buffer solution (Wako Pure Chemical Industries, Ltd, Osaka, Japan), embedded in paraffin, and sectioned on a microtome at 4 μm thick by standard techniques. The hematoxylin and eosin staining was performed as described previously [8]. The thickness of the skin tissue was measured using Leica QWin V3 image software (Leica Microsystems).

Statics
The statistical evaluations were performed using the two-tailed Student's t-test for unpaired values. Any differences between the data were considered to be significant when the p values were less than 0.05. The data are represented as the means plus or minus the standard deviation.

Conclusions
Our results revealed that Sod1 deficiency induced intracellular O 2 ·− accumulation in the cytoplasm and mitochondria resulting in decreased proliferation and increased apoptotic cell death in fibroblasts.
Our findings also demonstrated that Sod1 loss triggered the DNA damage response including p53 activation and promoted the mitochondrial dysfunction associated with low mitochondrial membrane potential. These results suggested that both mitochondrial dysfunction and p53 activation can cause the impairment of cell proliferation and the induction of apoptosis in Sod1 −/− fibroblasts. Furthermore, antioxidant treatment effectively suppressed the p53 activation and improved the cellular and skin phenotypes caused by the Sod1 deficiency in vitro and in vivo.