Spermidine Attenuates Oxidative Stress-Induced Apoptosis via Blocking Ca2+ Overload in Retinal Pigment Epithelial Cells Independently of ROS

Retinal pigment epithelial (RPE) cells occupy the outer layer of the retina and perform various biological functions. Oxidative damage to RPE cells is a major risk factor for retinal degeneration that ultimately leads to vision loss. In this study, we investigated the role of spermidine in a hydrogen peroxide (H2O2)-induced oxidative stress model using human RPE cells. Our findings showed that 300 μM H2O2 increased cytotoxicity, apoptosis, and cell cycle arrest in the G2/M phase, whereas these effects were markedly suppressed by 10 μM spermidine. Furthermore, spermidine significantly reduced H2O2-induced mitochondrial dysfunction including mitochondrial membrane potential and mitochondrial activity. Although spermidine displays antioxidant properties, the generation of intracellular reactive oxygen species (ROS) upon H2O2 insult was not regulated by spermidine. Spermidine did suppress the increase in cytosolic Ca2+ levels resulting from endoplasmic reticulum stress in H2O2-stimulated human RPE cells. Treatment with a cytosolic Ca2+ chelator markedly reversed H2O2-induced cellular dysfunction. Overall, spermidine protected against H2O2-induced cellular damage by blocking the increase of intracellular Ca2+ independently of ROS. These results suggest that spermidine protects RPE cells from oxidative stress, which could be a useful treatment for retinal diseases.


Introduction
Age-related macular degeneration (AMD), a multifaceted disease with demographic, environmental, and genetic risk factors, is among the most common causes of irreversible blindness in the world [1][2][3]. AMD progression occurs over an extended time period and its incidence rapidly increases in patients over 70 years old [1,4]. There are two major types of AMD: exudative or "wet" and non-exudative or "dry" [1,5]. Most patients with

Spermidine Attenuated H 2 O 2 -Induced Cytotoxicity in ARPE-19 Cells
Oxidative stress was generated by the addition of various concentrations of H 2 O 2 for 24 h and cell viability was measured via 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. As shown in Figure 1A, H 2 O 2 led to cytotoxicity at concentrations of 200 µM or more while cell viability dropped to approximately 74% at 300 µM. The IC 50 value of H 2 O 2 was 408.12 µM. Spermidine was also examined for cytotoxic effects and concentrations over 20 µM were found to be toxic ( Figure 1B). To assess the protective effects of spermidine on H 2 O 2 -induced cytotoxicity, we treated cells with spermidine for 1 h prior to being treated with 300 µM H 2 O 2 for 24 h. Figure 1C showed that pre-treatment with 10 µM spermidine significantly attenuated the decrease of cell viability induced by H 2 O 2 . However, spermidine does not have a protective effect against H 2 O 2 -induced cytotoxicity below 10 µM or over 20 µM. Morphologically, control ARPE-19 cells showed even density, spindle-shaped adherent monolayer growth, and stretched shapes ( Figure 1D). In contrast, H 2 O 2 -treated cells were sparse in density with many cells detached, whereas pre-treatment with spermidine improved to the cell morphology such that it was comparable to controls. Next, we evaluated which mode of cell death was involved in H 2 O 2 -induced cytotoxicity and whether this event was regulated by spermidine. Results of flow cytometric analysis using annexin V/propidium iodine (PI) staining showed 300 µM H 2 O 2 markedly enhanced the frequency of annexin V-positive cells to approximately 25%, but this increase was significantly suppressed by spermidine ( Figure 1E,F). These results suggest that spermidine attenuates H 2 O 2 -mediated oxidative stress-induced apoptosis in ARPE-19 cells. with 10 μM spermidine significantly attenuated the decrease of cell viability induced by H2O2. However, spermidine does not have a protective effect against H2O2-induced cytotoxicity below 10 μM or over 20 μM. Morphologically, control ARPE-19 cells showed even density, spindle-shaped adherent monolayer growth, and stretched shapes ( Figure 1D). In contrast, H2O2-treated cells were sparse in density with many cells detached, whereas pre-treatment with spermidine improved to the cell morphology such that it was comparable to controls. Next, we evaluated which mode of cell death was involved in H2O2induced cytotoxicity and whether this event was regulated by spermidine. Results of flow cytometric analysis using annexin V/propidium iodine (PI) staining showed 300 μM H2O2 markedly enhanced the frequency of annexin V-positive cells to approximately 25%, but this increase was significantly suppressed by spermidine ( Figure 1E,F). These results suggest that spermidine attenuates H2O2-mediated oxidative stress-induced apoptosis in ARPE-19 cells. Data are expressed as the mean ± SD (n = 4). * p < 0.05 and *** p < 0.001 when compared to control. ### p < 0.001 when compared to H2O2-treated cells.

Spermidine Downregulated Extrinsic and Intrinsic Apoptosis Pathways in H2O2-Stimulated ARPE-19 Cells
On the basis of spermidine's suppression of H2O2-induced apoptosis in ARPE-19 cells, we investigated which apoptotic pathways were involved in this process. Figure 2A indicates that H2O2 upregulated the expression of death receptor 4 (DR4) and Bax, but downregulated the expression of anti-apoptotic Bcl-2. Spermidine reversed the altered expression of apoptosis regulator proteins following H2O2 exposure. To determine whether spermidine regulates the mitochondrial-mediated intrinsic apoptosis pathway, we evaluated mitochondrial functions, including mitochondrial membrane potential (MMP, ∆Ψm) (A-C) Cell viability was measured via 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Data are expressed as the mean ± SD (n = 5). *** p < 0.001 when compared to control. ## p < 0.01 when compared to H 2 O 2 -treated cells. (D) Morphological changes observed under an inverted microscope (scale bar; 75 µm). (E) Flow cytometry: annexin V and propidium iodine (PI). (F) The percentages of apoptotic cells were determined by counting the percentage of annexin V-positive cells. Data are expressed as the mean ± SD (n = 4). * p < 0.05 and *** p < 0.001 when compared to control. ### p < 0.001 when compared to H 2 O 2 -treated cells.

Spermidine Downregulated Extrinsic and Intrinsic Apoptosis Pathways in H 2 O 2 -Stimulated ARPE-19 Cells
On the basis of spermidine's suppression of H 2 O 2 -induced apoptosis in ARPE-19 cells, we investigated which apoptotic pathways were involved in this process. Figure 2A indicates that H 2 O 2 upregulated the expression of death receptor 4 (DR4) and Bax, but downregulated the expression of anti-apoptotic Bcl-2. Spermidine reversed the altered expression of apoptosis regulator proteins following H 2 O 2 exposure. To determine whether spermidine regulates the mitochondrial-mediated intrinsic apoptosis pathway, we evaluated mitochondrial functions, including mitochondrial membrane potential (MMP, ∆Ψm) and mitochondrial activity using 5,5 ,6,6 -tetrachloro-1,1 ,3,3 -tetraethyl-imidacarbocyanune iodide (JC-1) staining and MitoTracker Red staining, respectively. The outcome of flow cytometric analysis for JC-1 showed that H 2 O 2 greatly promoted the frequency of JC-1 monomers, which indicates mitochondrial membrane potential (MMP, ∆Ψm) loss by mitochondrial membrane depolarization. As shown, pre-treatment with spermidine markedly suppressed H 2 O 2 -induced MMP (∆Ψm) loss ( Figure 2B,C). Furthermore, Figure 2D indicates that the population of MitoTracker Red-positive cells, indicating healthy mitochondria, was decreased by H 2 O 2 , whereas it was recovered to control levels by spermidine. In addition, we investigated the effect of spermidine on changes in cytochrome c following MMP (∆Ψm) loss in H 2 O 2 -stimulated ARPE-19 cells. The immunoblotting results of Figure 2E show that expression of cytochrome c in H 2 O 2 -stimulated cells was increased in the cytoplasm as compared to the mitochondria; these results were reversed by pre-treatment with spermidine, indicating that spermidine has protective effects on cytochrome c release from the mitochondria induced by H 2 O 2 . Moreover, the activities of caspase-3, -8, and -9 were significantly increased by H 2 O 2 stimulation, while substantially decreased by spermidine ( Figure 2F-H). These results suggest that spermidine downregulated the extrinsic apoptosis pathway, including the increase in DR4 expression and caspase-8 activity upon H 2 O 2 insult. Simultaneously, spermidine also acted on the intrinsic apoptosis pathway, displaying protective effects related to mitochondrial dysfunction-mediated MMP (∆Ψm) loss, Bcl-2 downregulation, cytochrome c release, and caspase-9 activation. and mitochondrial activity using 5,5′,6,6′-tetrachloro-1,1′3,3′-tetraethyl-imidacarbocyanune iodide (JC-1) staining and MitoTracker Red staining, respectively. The outcome of flow cytometric analysis for JC-1 showed that H2O2 greatly promoted the frequency of JC-1 monomers, which indicates mitochondrial membrane potential (MMP, ∆Ψm) loss by mitochondrial membrane depolarization. As shown, pre-treatment with spermidine markedly suppressed H2O2-induced MMP (∆Ψm) loss ( Figure 2B,C). Furthermore, Figure 2D indicates that the population of MitoTracker Red-positive cells, indicating healthy mitochondria, was decreased by H2O2, whereas it was recovered to control levels by spermidine. In addition, we investigated the effect of spermidine on changes in cytochrome c following MMP (∆Ψm) loss in H2O2-stimulated ARPE-19 cells. The immunoblotting results of Figure 2E show that expression of cytochrome c in H2O2-stimulated cells was increased in the cytoplasm as compared to the mitochondria; these results were reversed by pre-treatment with spermidine, indicating that spermidine has protective effects on cytochrome c release from the mitochondria induced by H2O2. Moreover, the activities of caspase-3, -8, and -9 were significantly increased by H2O2 stimulation, while substantially decreased by spermidine ( Figures 2F-H). These results suggest that spermidine downregulated the extrinsic apoptosis pathway, including the increase in DR4 expression and caspase-8 activity upon H2O2 insult. Simultaneously, spermidine also acted on the intrinsic apoptosis pathway, displaying protective effects related to mitochondrial dysfunctionmediated MMP (∆Ψm) loss, Bcl-2 downregulation, cytochrome c release, and caspase-9 activation.  , and caspase-9 (H) were measured using caspase colorimetric assay kits. Data are expressed as the mean ± SD (n = 3). * p < 0.05 and *** p < 0.001 when compared to control. ## p < 0.01 and ### p < 0.001 when compared to H 2 O 2 -treated cells.

Spermidine Suppressed DNA Damage and Dysregulation of Cell Cycle Processes in H 2 O 2 -Stimulated ARPE-19 Cells
To assess whether spermidine can decrease H 2 O 2 -induced DNA damage, we performed immunofluorescence analysis and immunoblots for γH2AX, a sensitive marker for DNA damage. As results indicate, spermidine greatly suppressed the increase in the expression of γH2AX following H 2 O 2 insult ( Figure 3A-C). We further examined the effect of spermidine on cell cycle progression in H 2 O 2 -stimulated ARPE-19 cells. The results of flow cytometric analysis for PI staining showed that H 2 O 2 increased the distribution of cells in the G2/M phase, which was markedly decreased by spermidine treatment (Figure 3D,E). The result of Western blotting analysis for cell cycle regulators indicated that H 2 O 2 upregulated the expression of p53, p16, cyclin A, and cyclin B1, while it downregulated the expression of p27; alterations in cyclin-dependent kinase 2 (CDK2) or cell division cycle gene 2 (CDC2, or CDK1) were not induced by H 2 O 2 insult. Importantly, these changes to the expression of cell cycle regulators following H 2 O 2 insult were noticeably restored by spermidine ( Figure 3F). These results suggest that spermidine has a protective effect against H 2 O 2 -induced DNA damage and cell cycle arrest at the G2/M phase through the control of cell cycle regulators. compared to control. ## p < 0.01 when compared to H2O2-treated cells. (D) Cells probed with 100 nM MitoTracker Red and observed under a fluorescence microscope. Scale bar: 200 μm. (E) Cytosolic and mitochondrial proteins were isolated and the expression of cytochrome c was detected by Western blot analysis. Cytochrome oxidase subunit VI (COX IV) and βactin served as protein loading controls for the mitochondria and cytosol, respectively. The activities of caspase-3 (F), caspase-8 (G), and caspase-9 (H) were measured using caspase colorimetric assay kits. Data are expressed as the mean ± SD (n = 3). * p < 0.05 and *** p < 0.001 when compared to control. ## p < 0.01 and ### p < 0.001 when compared to H2O2treated cells.

Spermidine Suppressed DNA Damage and Dysregulation of Cell Cycle Processes in H2O2-Stimulated ARPE-19 Cells
To assess whether spermidine can decrease H2O2-induced DNA damage, we performed immunofluorescence analysis and immunoblots for γH2AX, a sensitive marker for DNA damage. As results indicate, spermidine greatly suppressed the increase in the expression of γH2AX following H2O2 insult ( Figure 3A-C). We further examined the effect of spermidine on cell cycle progression in H2O2-stimulated ARPE-19 cells. The results of flow cytometric analysis for PI staining showed that H2O2 increased the distribution of cells in the G2/M phase, which was markedly decreased by spermidine treatment ( Figure  3D,E). The result of Western blotting analysis for cell cycle regulators indicated that H2O2 upregulated the expression of p53, p16, cyclin A, and cyclin B1, while it downregulated the expression of p27; alterations in cyclin-dependent kinase 2 (CDK2) or cell division cycle gene 2 (CDC2, or CDK1) were not induced by H2O2 insult. Importantly, these changes to the expression of cell cycle regulators following H2O2 insult were noticeably restored by spermidine ( Figure 3F). These results suggest that spermidine has a protective effect against H2O2-induced DNA damage and cell cycle arrest at the G2/M phase through the control of cell cycle regulators.

Spermidine Had Antioxidant Capacity But Did Not Regulate Intracellular ROS Generation in H 2 O 2 -Stimulated ARPE-19 Cells
Next, to investigate whether the protective effects of spermidine on H 2 O 2 -induced apoptosis were due to the blocking of the oxidative stress, we observed the intracellular ROS levels using dichlorodihydrofluorescein diacetate (DCF-DA), a fluorescent-labeled probe. Results of flow cytometric analysis showed intracellular ROS production was significantly increased (≈90%) by H 2 O 2 after 30 min and this increment was not altered by spermidine ( Figure 4A,B). Pre-treatment with N-acetyl-L-cysteine (NAC), an ROS scavenger commonly used as a positive control, completely blocked intracellular ROS generation following H 2 O 2 . Meanwhile, H 2 O 2 slightly upregulated the expression of Kelch-like ECH-associated protein 1 (Keap1) and heme oxygenase-1 (HO-1), which are enzymes with antioxidant properties induced by oxidative stress ( Figure 4C). In spermidine-treated cells, the expression of Keap1 and HO-1 was markedly increased ( Figure 4C). On the basis of these results, we determined that spermidine does not block H 2 O 2 -induced intracellular ROS generation despite its antioxidant capacity.
2-phenylindole (DAPI) was used to counterstained the nuclei (blue). Scale bar: 75 μm. (B) The percentage of DNA-damaged cells in whole field. Data are expressed as the mean ± SD (n = 3). ** p < 0.01 and *** p < 0.001 when compared to control. ### p < 0.001 when compared to H2O2-treated cells. (C) Expression of γH2AX was determined by Western blot analysis. β-Actin was used as an internal control. (D,E) Flow cytometry: PI. (D) Representative histograms. (E) The average percentages of cells in each phase of the cell cycle are displayed (excluding sub-G1). (F) The expression of cell cycle-regulatory proteins was determined by Western blot analysis. β-Actin was used as an internal control.

Spermidine Had Antioxidant Capacity But Did Not Regulate Intracellular ROS Generation in H2O2-Stimulated ARPE-19 Cells
Next, to investigate whether the protective effects of spermidine on H2O2-induced apoptosis were due to the blocking of the oxidative stress, we observed the intracellular ROS levels using dichlorodihydrofluorescein diacetate (DCF-DA), a fluorescent-labeled probe. Results of flow cytometric analysis showed intracellular ROS production was significantly increased (≈ 90%) by H2O2 after 30 min and this increment was not altered by spermidine ( Figure 4A,B). Pre-treatment with N-acetyl-L-cysteine (NAC), an ROS scavenger commonly used as a positive control, completely blocked intracellular ROS generation following H2O2. Meanwhile, H2O2 slightly upregulated the expression of Kelch-like ECHassociated protein 1 (Keap1) and heme oxygenase-1 (HO-1), which are enzymes with antioxidant properties induced by oxidative stress ( Figure 4C). In spermidine-treated cells, the expression of Keap1 and HO-1 was markedly increased ( Figure 4C). On the basis of these results, we determined that spermidine does not block H2O2-induced intracellular ROS generation despite its antioxidant capacity.

Spermidine Decreased Cytosolic Ca 2+ Levels Released Due to Endoplasmic Reticulum (ER) Stress in H2O2-Stimulated ARPE-19 Cells
In order to determine whether spermidine is involved in the recovery of ER damage in H2O2-stimulated cells, we stained cultures with ER-Tracker Red. We observed the fluorescence expression of ER-Tracker Red was markedly suppressed in H2O2-treated cells

Spermidine Decreased Cytosolic Ca 2+ Levels Released Due to Endoplasmic Reticulum (ER) Stress in H 2 O 2 -Stimulated ARPE-19 Cells
In order to determine whether spermidine is involved in the recovery of ER damage in H 2 O 2 -stimulated cells, we stained cultures with ER-Tracker Red. We observed the fluorescence expression of ER-Tracker Red was markedly suppressed in H 2 O 2 -treated cells as compared with control cells; spermidine substantially reversed these effects ( Figure 5A). Next, we investigated the effect of spermidine on intracellular Ca 2+ ([Ca 2+ ]i) levels resulting from ER damage. ER stress frequently results in the release of Ca 2+ from the interior of the ER, inducing cytosolic Ca 2+ accumulation and triggering cell death [21].  Figure 5A, NAC preserved the expression of ER-Tracker Red, indicating that the blocking of ROS repaired ER function. BAPTA-AM also improved the expression of ER-Tracker Red as compared to the H 2 O 2 -treated cells, but this expression was minor compared to spermidine or NAC ( Figure 5A). As expected, the results of fluo-4 AM staining showed that NAC pretreatment significantly suppressed fluorescence intensity compared with H 2 O 2 -treatment alone ( Figure 5C,F). The [Ca 2+ ]i levels were also elevated by NAC treatment in the absence H 2 O 2 ( Figure 5C,F). Additionally, intracellular ROS levels were markedly suppressed by BAPTA-AM treatment, suggesting that ROS and [Ca 2+ ]i interacted by some mechanism (Figure 5D,G). These results suggest that H 2 O 2 promotes intracellular ROS generation, resulting in increased [Ca 2+ ]i levels due to ER stress. Interestingly, our results show that spermidine can act as a specific [Ca 2+ ]i chelator in an ROS-independent manner.
as compared with control cells; spermidine substantially reversed these effects ( Figure  5A). Next, we investigated the effect of spermidine on intracellular Ca 2+ ([Ca 2+ ]i) levels resulting from ER damage. ER stress frequently results in the release of Ca 2+ from the interior of the ER, inducing cytosolic Ca 2+ accumulation and triggering cell death [21]. Re  Figure 5A, NAC preserved the expression of ER-Tracker Red, indicating that the blocking of ROS repaired ER function. BAPTA-AM also improved the expression of ER-Tracker Red as compared to the H2O2-treated cells, but this expression was minor compared to spermidine or NAC ( Figure 5A). As expected, the results of fluo-4 AM staining showed that NAC pre-treatment significantly suppressed fluorescence intensity compared with H2O2-treatment alone ( Figure 5C,F). The [Ca 2+ ]i levels were also elevated by NAC treatment in the absence H2O2 ( Figure 5C,F). Additionally, intracellular ROS levels were markedly suppressed by BAPTA-AM treatment, suggesting that ROS and [Ca 2+ ]i interacted by some mechanism (Figure 5D,G). These results suggest that H2O2 promotes intracellular ROS generation, resulting in increased [Ca 2+ ]i levels due to ER stress. Interestingly, our results show that spermidine can act as a specific [Ca 2+ ]i chelator in an ROSindependent manner.

Blocking of Cytosolic Ca 2+ Levels Attenuated H 2 O 2 -Induced Cytotoxicity in ARPE-19 Cells
To examine the role of spermidine as a specific intracellular Ca 2+ chelator, we investigated the effect of BAPTA-AM on H 2 O 2 -induced cellular alteration. Pre-treatment with BAPTA-AM significantly improved cell viability and decreased apoptosis following H 2 O 2 treatment ( Figure 6A-C). Furthermore, BAPTA-AM markedly inhibited H 2 O 2 -induced cell cycle arrest at the G2/M phase ( Figure 6D,E). Additionally, the mitochondrial function of BAPTA-AM-treated cells was also greatly improved compared to H 2 O 2 -stimualted cells ( Figure 6F-H). These results suggest that ER stress-mediated intracellular Ca 2+ increases play a critical role in H 2 O 2 -induced cytotoxicity.

Blocking of Cytosolic Ca 2+ Levels Attenuated H2O2-Induced Cytotoxicity in ARPE-19 Cells
To examine the role of spermidine as a specific intracellular Ca 2+ chelator, we investigated the effect of BAPTA-AM on H2O2-induced cellular alteration. Pre-treatment with BAPTA-AM significantly improved cell viability and decreased apoptosis following H2O2 treatment ( Figure 6A-C). Furthermore, BAPTA-AM markedly inhibited H2O2-induced cell cycle arrest at the G2/M phase ( Figure 6D,E). Additionally, the mitochondrial function of BAPTA-AM-treated cells was also greatly improved compared to H2O2-stimualted cells ( Figure 6F-H). These results suggest that ER stress-mediated intracellular Ca 2+ increases play a critical role in H2O2-induced cytotoxicity.

Discussion
RPE cells are polarized epithelial cells that play a key role in retinal physiology including forming the outer blood-retinal barrier; transport of ions, water, nutrients, and metabolic end products; phagocytosis; and production of various growth factors [22]. Interestingly, the RPE is an ideal environment for the production of ROS [23]. The RPE contains an abundance of photosensitizers that generate reactive oxygen intermediates as a result of photochemical reactions and cellular metabolism [24]. Additionally, the process of phagocytosis by the RPE itself involves oxidative stress and leads to production of reactive oxygen intermediates [25]. Dysfunction and cell death in RPE cells are hallmarks of AMD; mechanistically, oxidative stress and reactive oxygen intermediates are believed to contribute to RPE cell death in AMD [26]. To identify the pathological mechanism of RPE dysfunction in AMD, numerous studies have evaluated RPE cell death in response to oxidative stress using pro-oxidants such as H 2 O 2 and tert-butyl hydroperoxide (tBH) [27][28][29][30]. In the present study, we found that H 2 O 2 led to cytotoxicity above 200 µM in ARPE-19 cells and accompanied apoptotic morphological changes. Furthermore, our findings indicate that H 2 O 2 -induced cytotoxicity caused apoptotic cell death in ARPE-19 cells. On the basis of these results, we established a model of oxidative stress-mediated RPE cell death using H 2 O 2 , and our findings demonstrated that spermidine significantly suppressed H 2 O 2 -induced RPE apoptosis (Figure 1). Apoptosis can be executed by extrinsic or intrinsic pathways, which are mediated through cellular membrane death receptors and mitochondria, respectively [31]. The extrinsic pathway involves the formation of a death-inducing signaling complex and the recruitment of initiator caspases, which in turn activate the downstream effector caspases [32]. Meanwhile, the intrinsic pathway can be triggered by viral infection, UV, growth factor deprivation, inordinate ROS, or DNA damage, which lead to activation of proapoptotic Bcl-2 family proteins located in outer mitochondrial membrane [33]. Bax promotes the opening of the mitochondrial transition pores and subsequently induces the release of cytochrome c into the cytoplasm, which leads to the formation of an apoptosome and activation of effector caspases [33]. Both pathways activate the effector caspases, leading to the cleavage or degradation of cellular substrates including poly (ADP-ribose) polymerase and histones, ultimately leading to apoptotic cell death [31,32]. In the present study, we found that spermidine downregulated the activation of the extrinsic apoptosis pathway induced by H 2 O 2 , including DR4 expression and caspase-8 activity. Simultaneously, our findings demonstrated that spermidine also acts on intrinsic apoptosis pathway, including MMP (∆Ψm) loss, Bcl-2 down-expression, cytochrome c release to cytoplasm, and caspase-9 activation (Figure 2). Among the wide variety of factors that are instrumental in the etiology and pathogenesis of AMD, mitochondrial damage in RPE cells contributes significantly to RPE dysfunction [34]. In AMD, mitochondria are fragmented with a higher number of lesions, altered ATP synthase activity, as well as compromised protein expression and nuclear-encoded protein import [35,36]. Our findings demonstrate that oxidative stress induced by H 2 O 2 caused mitochondrial damage and resulted in MMP (∆Ψm) loss as well as cytochrome c release; these conditions led to the activation of the intrinsic apoptosis pathway, which was suppressed by spermidine.
Cells are able to block the cell cycle transiently or irreversibly in response to stressful conditions. Apoptosis and cell cycle arrest commonly occur in response to DNA damage [37,38]. One response to DNA damage is the expression of γH2AX, which is an early sign of DNA damage induced by stalled replication [38]. The formation of γH2AX foci takes place immediately after the generation of a DNA break, as well as replication stalling or single-stranded DNA breaks [38]. When DNA is damaged, the G2 checkpoint inhibits cells from entering mitosis, thereby arresting the cell cycle at the G2/M phase, indicating that the damage of intracellular DNA is difficult to repair [39]. Progression of cell cycle is regulated by CDKs and their activity is coordinated by the binding of their essential regulatory subunits, cyclins [40]. The cyclin B1/CDC2 complex regulates cell cycle progression from the G2 to M phase, and cyclins accumulate steadily during the G2 phase while being rapidly eliminated as cells exit mitosis [41]. In this study, the suppression of γH2AX by spermidine demonstrated protective effects against H 2 O 2 -induced DNA damage in RPE cells. Furthermore, our findings provided data showing that H 2 O 2 upregulated the expression of p53, p16, CDC2, cyclin A, and cyclin B1, whereas these changes to the expression of p53 and p16 following H 2 O 2 were slightly downregulated by spermidine (Figure 3). In addition, our results suggest that spermidine has preventive effects against H 2 O 2 -induced cell cycle arrest at G2/M phase through partially control of cell cycle regulators ( Figure 3). These results correspond with the findings of Liu et al. [42], reporting that oxidative damage by H 2 O 2 triggered G2/M phase arrest of ARPE-19 cells through the regulation of cyclin B1. On the basis of our presented data, we can support the claim that spermidine suppresses oxidative stress-mediated DNA damage induced by H 2 O 2 , which in turn prevents cell cycle arrest in the G2/M phase.
As previous discussed, oxidative stress contributes to RPE cell death in AMD. In RPE cells, oxidative stress by ROS generation is mainly derived from the photo-oxidation of mitochondria, nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, and the decline in ability to repair damaged organelles [23,43,44]. Several studies have suggested that H 2 O 2 enhances the production of intracellular ROS in ARPE-19 cells [26,45,46], and our results agree with these reports (Figure 4). Of note, spermidine does not block H 2 O 2 -induced intracellular ROS generation despite its antioxidant actions. Although Rider et al. [18] demonstrated that spermidine inhibits the action of ROS and acts as an endogenous ROS scavenger, few studies have focused on the ROS scavenging effect of spermidine in RPE. In light of these reports, our findings are significant as we show that while spermidine has antioxidant capacity, it does not act as an ROS scavenger. On the basis of these results, we hypothesized that H 2 O 2 -induced oxidative stress is derived from not only intracellular ROS, but from other sources as well.
RPE cells are a target and source of various cytokines whose intracellular signaling cascades influence [Ca 2+ ]i levels [22]. In addition, changes in [Ca 2+ ]i are involved in normal RPE function, including transcellular fluid and ion transport, cell differentiation, and photoreceptor outer segment phagocytosis [22]. Homeostatic disorders in the calcium signaling system could represent a mechanism underlying apoptosis as changes in [Ca 2+ ]i provide a chemical signal for early cell death [47,48]. As the ER stores [Ca 2+ ]i for various physiological functions [49], experimental evidence suggests that ER dysfunction induced by oxidative stress in RPE cells is a key risk factor exacerbating the progression of AMD [50,51]. Yao et al. [52] Figure 6). These data suggest that ER stress-mediated [Ca 2+ ]i plays a critical role in H 2 O 2 -induced cytotoxicity and that spermidine acts as a [Ca 2+ ]i chelator.

Cell Viability Analysis
To measure the cytotoxicity of spermidine, we assessed cell viability via MTT assay. The cells were treated with various concentrations (0, 1, 10, 20, and 30 µM) of spermidine for 24 h. In order to assess the effect of spermidine, NAC, or BAPTA-AM upon oxidative stress, we pre-treated the cells with or without 10 µM spermidine, 5 mM NAC, or 5 µM BAPTA-AM for 1 h before being incubated for 24 h in the presence or absence of 300 µM H 2 O 2 . Afterwards, the cells were incubated with 0.5 mg/mL of MTT solution for 3 h before being dissolved in DMSO. Optical density was detected at 540 nm by a microplate reader (VERSA Max, Molecular Device Co., Sunnyvale, CA, USA) as previously described [54]. The cellular morphology was observed using an inverted microscope (Carl Zeiss, Oberkochen, Germany).

Flow Cytometric Analysis
Cells were pre-treated with or without 10 µM spermidine, 5 mM NAC, or 5 µM BAPTA-AM for 1 h before incubation for 24 h in the presence or absence of 300 µM H 2 O 2 . To measure apoptosis, we stained cells with FITC annexin V/PI for 20 min, according to the manufacturer's protocol. The fluorescence intensity was detected using a flow cytometer (BD Biosciences), and FITC annexin V+/PI-cell populations were considered apoptotic [55]. In order to quantify the phase distribution of the cell cycle, we stained cells with 40 µg/mL PI for 30 min and analyzed them by flow cytometry [56]. To assess the MMP (∆Ψm), we loaded cells with 10 µM JC-1 for 20 min, and the frequency of JC-1 aggregates and monomers were analyzed [57]. For the intracellular calcium assay, cells were incubated with 1 µM fluo-4 AM probe for 30 min, and the fluorescence intensity was measured by flow cytometry [58].

Intracellular ROS Detection
Intracellular ROS production was assessed by DCF-DA staining as previously described [59]. In brief, cells were pre-treated with or without 10 µM spermidine, 5 mM NAC, or 5 µM BAPTA-AM for 1 h before incubation with 300 µM H 2 O 2 for 30 min. Subsequently, 10 µM DCF-DA was added to the cell culture for 20 min and the stained images were acquired using a fluorescence microscope (Carl Zeiss).

Fluorescence Image Analysis
Cells were pre-treated with or without 10 µM spermidine, 5 mM NAC, or 5 µM BAPTA-AM for 1 h before incubation for 24 h in the presence or absence of 300 µM H 2 O 2 . To assess the function of mitochondria and the endoplasmic reticulum, we stained cells with 100 nM MitoTracker Red and 1 µM ER-Tracker Red probe, respectively, before observation using a fluorescence microscope (Carl Zeiss) per the manufacturer's instructions.

Western Blot Analysis
ARPE-19 cells were pre-treated with or without 10 µM spermidine for l hour before incubation with 300 µM H 2 O 2 for 24 h. Cells were harvested, lysed, and total proteins were analyzed using the Bradford protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA). In a parallel experiment, mitochondrial and cytosolic proteins were extracted using a mitochondria isolation kit according to the manufacturer's instructions. Equal amount of protein underwent sodium dodecyl sulfate polyacrylamide gel electrophoresis and were transferred to polyvinylidene difluoride membranes (Schleicher & Schuell, Keene, NH, USA). After blocking for 1 h, the membranes were incubated with specific primary antibodies at 4 • C overnight before incubation with the corresponding secondary antibodies for 1 h as previously described [60]. Information on the antibodies used is provided in Supplementary Table S1. The chemiluminescent bands were visualized by a Fusion FX Imaging System (Vilber Lourmat, Torcy, France).

Measurement of Caspases Activities
The cells were pre-treated with or without 10 µM spermidine for l hour before incubation with 300 µM H 2 O 2 for 24 h and lysed. Caspase-3, -8, and -9 activities were determined using colorimetric assay kits according to the manufacturer's instruction.

Immunofluorescence Analysis
Cells were transferred to a 4-well chamber slide (SPL Life Sciences Co., Pocheon, Korea) and incubated for 24 h, pre-treated with or without 10 µM spermidine for l hour, and then treated with 300 µM H 2 O 2 for 24 h. The cells were then incubated with a γH2AX antibody (Cell Signaling Technology, Beverly, MA, USA, Cat No. 9718) at 4 • C overnight before being probed with an Alexa Fluor 594-labeled donkey anti-rabbit IgG antibody for 1 h in darkness. DAPI was used to counterstain the nuclei. The cells were mounted and observed using a fluorescence microscope (Carl Zeiss).

Statistical Analysis
All the experiments were performed by conducting each assay at least three times. The data were analyzed using GraphPad Prism 5.03 (GraphPad Software Inc., La Jolla, CA, USA) and are expressed as the means ± standard deviation (SD). The statistical analyses were conducted using analysis of variance (ANOVA) and Tukey's post hoc test to examine between-group differences; p < 0.05 was considered significant.

Conclusions
In conclusion, ARPE-19 cells treated with H 2 O 2 underwent apoptosis through both the intrinsic and extrinsic pathways in response to cellular damage to the DNA, mitochondria, and ER. Pretreatment with spermidine caused a marked decrease in apoptosis and cell cycle arrest through the downregulation of ER stress-mediated [Ca 2+ ]i overload, which is ROS-independent ( Figure 7). However, further studies are needed to identify the role of spermidine on in the regulation of calcium channels in RPE. Our findings reveal that spermidine acts as a [Ca 2+ ]i blocker in oxidative stress-induced RPE injury and offer a deeper understanding of spermidine's clinical potential for the treatment of retinal disorders. dria, and ER. Pretreatment with spermidine caused a marked decrease in apoptosis and cell cycle arrest through the downregulation of ER stress-mediated [Ca 2+ ]i overload, which is ROS-independent ( Figure 7). However, further studies are needed to identify the role of spermidine on in the regulation of calcium channels in RPE. Our findings reveal that spermidine acts as a [Ca 2+ ]i blocker in oxidative stress-induced RPE injury and offer a deeper understanding of spermidine's clinical potential for the treatment of retinal disorders. Figure 7. Spermidine attenuates H2O2-induced cellular dysfunction via suppression of Ca 2+ signaling pathways in ARPE-19 cells. Oxidative stress by H2O2 triggers apoptosis through the intrinsic and extrinsic pathways in ARPE-19 cells. Spermidine markedly attenuated mitochondrial and nuclear dysfunction upon oxidative stress, inhibiting apoptosis. Spermidine suppresses intracellular Ca 2+ levels and repairs ER damage, creating the anti-apoptotic effects. Nevertheless, while spermidine has antioxidant capacity, it does not down-regulate intracellular ROS generation during oxidative stress. Taken together, spermidine suppresses oxidative stress-induced cellular dysfunction via suppression of Ca 2+ signaling pathways in ARPE-19 cells independently of ROS generation.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1: Table S1: Primary and secondary antibodies used for immunoblotting.