Magnesium Supplementation Attenuates Ultraviolet-B-Induced Damage Mediated through Elevation of Polyamine Production in Human HaCaT Keratinocytes

Magnesium ions (Mg2+) have favorable effects such as the improvement of barrier function and the reduction of inflammation reaction in inflammatory skin diseases. However, its mechanisms have not been fully understood. Microarray analysis has shown that the gene expressions of polyamine synthases are upregulated by MgCl2 supplementation in human HaCaT keratinocytes. Here, we investigated the mechanism and function of polyamine production. The mRNA and protein levels of polyamine synthases were dose-dependently increased by MgCl2 supplementation, which were inhibited by U0126, a MEK inhibitor; CHIR-99021, a glycogen synthase kinase-3 (GSK3) inhibitor; and Naphthol AS-E, a cyclic AMP-response-element-binding protein (CREB) inhibitor. Similarly, reporter activities of polyamine synthases were suppressed by these inhibitors, suggesting that MEK, GSK3, and CREB are involved in the transcriptional regulation of polyamine synthases. Cell viability was reduced by ultraviolet B (UVB) exposure, which was rescued by MgCl2 supplementation. The UVB-induced elevation of reactive oxygen species was attenuated by MgCl2 supplementation, which was inhibited by cysteamine, a polyamine synthase inhibitor. Our data indicate that the expression levels of polyamine synthases are upregulated by MgCl2 supplementation mediated through the activation of the MEK/GSK3/CREB pathway. MgCl2 supplementation may be useful in reducing the UVB-induced oxidative stress in the skin.


Introduction
The important roles of skin are to maintain body hydration, temperature, moisture, and sensations [1]. Exposure of skin to high doses of ultraviolet (UV) can induce initiating events of skin inflammation and carcinogenesis. The UV spectrum is divided into three major sections: short-wave UV with 250-290 nm (UVC), middle-wave UV with 290-320 nm (UVB), and long-wave UV with 320-400 nm (UVA). UVC is blocked by the ozone layer, but UVA and UVB can reach the Earth's surface and fall on the skin, eyes, and hair of humans. UVA irradiation is preferentially absorbed by lipid rafts of cell membrane and induces the formation of singlet oxygen. UVB irradiation is absorbed by DNA and proteins, resulting in the production of reactive molecule species. High doses of UVB (>25 mJ/cm 2 ) show a cytotoxic effect in various cells including keratinocytes. The UVB-induced cell damage is caused by the generation of reactive oxygen species (ROS) including superoxide anion radical, hydroxy radical, and hydrogen peroxide [2].
Magnesium is a divalent cation most abundantly existing in the human body, and a majority of it is stored in the bone, muscle, and liver. Magnesium ions (Mg 2+ ) play a pivotal Magnesium is a divalent cation most abundantly existing in the human body, and a majority of it is stored in the bone, muscle, and liver. Magnesium ions (Mg 2+ ) play a pivotal role in the regulation of hundreds of enzymatic reactions including synthesis of DNA, RNA, and protein; glucose metabolization; and energy production [3]. In the skin, Mg 2+ has functions such as acceleration of tissue repair [4] and suppression of inflammatory response [5]. Bathing in a Mg 2+ -rich Dead Sea salt solution induces several favorable effects, including the improvement of skin barrier function, enhancement of skin hydration, and reduction of inflammation in atopic dry skin [6]. These effects may be caused by the properties of Mg 2+ such as water binding capability. However, the functional mechanism of Mg 2+ has not been fully understood.
Three major polyamines (putrescine, spermidine, and spermine) are present in living cells, and spermine is especially abundantly contained in the human epidermis [7]. These polyamines play pivotal roles in cell proliferation, differentiation, apoptosis, and protection from oxidative damage [8,9]. Putrescine, which is synthesized from ornithine by ornithine decarboxylase (ODC), is converted to spermidine and spermine by the addition of aminopropyl groups derived from decarboxylated S-adenosylmethionine, which is converted from S-adenosylmethionine (SAM) to S-adenosyl-methioninamine (dcSAM) by adenosylmethionine decarboxylase 1 (AMD1) (Figure 1). These reactions are catalyzed by spermidine synthase (SRM) and spermine synthase. Intracellular polyamine contents are strictly regulated by a combination of synthesis, degradation, uptake, and excretion. The mice targeted disruption of ODC and AMD1 genes died early in embryonic development [10,11]. The level of spermidine in skin changes during aging peaked at 10 weeks old and was markedly reduced at 26 weeks old [12]. Inflammation and macrophage activation in mouse skin are inhibited by natural polyamines [13]. In addition, polyamines can function as biological antioxidants related to radical scavenging activity [14]. In the present study, DNA microarray analysis showed that the expression levels of polyamine synthases are upregulated by MgCl2 supplementation in human HaCaT keratinocytes. Therefore, we investigated the mechanism and function of MgCl2 supplementation-induced polyamine production. The mRNA and protein levels were examined using real-time polymerase chain reaction (PCR) and Western blotting analyses, respectively. The reporter activities of polyamine synthases were assessed using luciferase assay. The production of ROS and lipid peroxide was measured using specific fluorescent In the present study, DNA microarray analysis showed that the expression levels of polyamine synthases are upregulated by MgCl 2 supplementation in human Ha-CaT keratinocytes. Therefore, we investigated the mechanism and function of MgCl 2 supplementation-induced polyamine production. The mRNA and protein levels were examined using real-time polymerase chain reaction (PCR) and Western blotting analyses, respectively. The reporter activities of polyamine synthases were assessed using luciferase assay. The production of ROS and lipid peroxide was measured using specific fluorescent probes. Our results indicate that MgCl 2 supplementation may be involved in the protection on HaCaT cells against UVB-induced damage mediated through the production of polyamines.

DNA Microarray Analysis
Cells were inoculated at a density of 1 × 10 5 /dish on 6 cm dishes and cultured for 96 h. Total RNA was extracted using NucleoSpin RNA (740955.10, Takara Bio, Shiga, Japan). DNA microarray analysis was performed using GeneChip Human Gene 2.0 ST array (a contract analysis service at Filgen Incorporation Japan, Nagoya, Japan). The cutoff point was set to twofold.

Extraction of Total RNA and Quantitative Real-Time PCR
Cells were inoculated at a density of 7 × 10 4 /well on 6-well plates and cultured for 96 h. Total RNA was extracted using TRI reagent (TR118, Molecular Research Center, Cincinnati, OH, USA). Reverse transcription and quantitative real-time PCR reactions were performed using a ReverTraAce qPCR RT Kit (FSQ-101, Toyobo, Osaka, Japan) and a Thunderbird SYBR qPCR Mix (QPS-201, Toyobo), respectively, as described previously [16]. Primer pairs used in real-time PCR are shown in Table 1.

Measurement of Polyamine Contents
Cells were inoculated at a density of 5 × 10 3 /well on 96-well plates and cultured for 96 h. The production of polyamines was measured using an intracellular polyamine detection reagent, PolyamineRED (FDV-0020, Biofunctional Synthetic Chemistry Laboratory, Cluster for PioneeringResearch, RIKEN, Saitama, Japan). The nucleus was stained with 4 ,6-diamidino-2-phenylindole (DAPI, D212, Dojindo Laboratories). The fluorescence images were obtained using a BZ-X800 fluorescence microscope (Keyence, Osaka, Japan), and the fluorescence intensities of PolyamineRED were calculated by ImageJ software.

Immunocytochemistry
Cells were inoculated at a density of 7 × 10 4 /well on 6-well plates containing sterile cover glasses and were cultured for 96 h. After fixation, permeabilization, and blocking, the cells were incubated with mouse anti-p-CREB (S133) antibody (1:100 dilution, R&D Systems, Minneapolis, MN, USA) followed by incubation with donkey anti-mouse Alexa 488 monoclonal antibody (1:100, Thermo Fisher Scientific, Waltham, MA, USA) and DAPI. The fluorescence images were obtained using an LSM700 confocal laser microscope (Carl Zeiss, Jena, Germany), as described previously [16], and the fluorescence intensities were calculated by ImageJ software.

Generations of ROS and Lipid Peroxides
Cells were inoculated at a density of 5 × 10 3 /well on 96-well plates and cultured for 96 h. The generation of ROS and lipid peroxides was detected by a 2 ,7 -dichlorodihydrofluoresceindiacetate (DCF-DA, D399, Thermo Fisher Scientific) and Liperfluo, a fluorescence dye used for specific detection of lipid peroxides (L248, Dojindo Laboratories), respectively. The fluorescence intensity was measured using an Infinite F200 fluorescence microplate reader (Tecan, Mannedorf, Switzerland).

Statistics
Results are presented as means ± SEM. Differences between groups were analyzed using one-way analysis of variance, and corrections for multiple comparison were made using Tukey's multiple comparison test. Comparisons between two groups were made using Student's t-test. Statistical analyses were performed using KaleidaGraph version 4.5.1 software (Synergy Software, Reading, PA, USA). Significant differences were assumed at p < 0.05.

Effect of Extracellular MgCl 2 Concentration on Gene Expression in HaCaT Cells
HaCaT cells were exposed to the media containing 0.8 mM MgCl 2 (normal concentration)-or 5.8 mM MgCl 2 (high concentration)-containing media for 6 h. DNA microarray analysis using total RNA isolated from HaCaT cells showed that 45 genes are upregulated and 13 genes are downregulated by high MgCl 2 concentration (Tables 2, S1 and S2). All 58 differentially expressed genes (DEGs) were analyzed by DAVID software, and the results of GO analysis indicated that (1) for metabolic process (BP), upregulated DEGs were particularly enriched in DNA replication, DNA replication initiation, DNA replication checkpoint, and mitotic DNA replication checkpoint, and downregulated DEGs in cellular calcium ion homeostasis; (2) for molecular function (MF), upregulated DEGs were enriched in DNA replication origin binding, DNA binding, and enzyme binding; (3) for cell component (CC), upregulated DEGs were enriched in nucleoplasm and nucleus, and downregulated DEGs in nuclear membrane (Table S3). KEGG analysis demonstrated that upregulated DEGs were particularly enriched in cell cycle, DNA replication, and arginine and proline metabolism, whereas downregulated DEGs were not significantly enriched in any signaling pathways (Table S4).  We selected two types of spermidine synthesis-related genes, SRM and AMD1, which were upregulated in 5.8 mM MgCl 2 -containing media. To validate the data of microarray, we performed real-time PCR with specific primers to SRM and AMD1. The mRNA levels of SRM and AMD1 were dose-dependently increased by MgCl 2 supplementation in HaCaT cells, and the effects were significant over the concentration of 3.3 mM MgCl 2 ( Figure 2A). Similarly, the mRNA levels of SRM and AMD1 were significantly increased by Mg-lactate and MgSO 4 supplementation ( Figure 2B). In addition, the mRNA levels of SRM and AMD1 were increased by MgCl 2 supplementation in NHEK/SVTERT3-5 cells ( Figure 2C). These results indicate that both SRM and AMD1 expressions may be upregulated by high Mg 2+ concentration in human keratinocytes. In contrast, neither SRM nor AMD1 expressions are significantly changed by MgCl 2 depletion in HaCaT cells ( Figure 2D).

Effect of MgCl 2 Supplementation on Polyamine Synthesis
HaCaT cells were exposed to the media containing 0.8, 3.3, 5.8, and 10.8 mM MgCl 2 for 24 h. The protein levels of SRM and AMD1 were dose-dependently increased by MgCl 2 supplementation ( Figure 3A). These results are similar to those in real-time PCR analysis. Spermidine is a polyamine compound that is present in the cytoplasm of eukaryotic cells [18]. The production of intracellular polyamines can be detected using a fluorescent probe, PolyamineRED. The fluorescence intensity of PolyamineRED was dose-dependently increased by MgCl 2 supplementation, and the relative values were 157.3 ± 8.6% (5.8 mM MgCl 2 ) and 183.8 ± 16.2% (10.8 mM MgCl 2 ) ( Figure 3B). These results indicate that MgCl 2 supplementation may upregulate the production of polyamines mediated by the elevation of spermidine-synthesis-related enzymes.

Effects of Intracellular Signaling Pathway Inhibitors on SRM and AMD1 Expressions
We recently reported that MgCl2 supplementation can activate a GSK3β/CREB pathway in HaCaT cells [17]. To clarify the regulatory mechanisms of SRM and AMD1 expressions, we investigated the effects of inhibitors for intracellular signaling cascades. MgCl2

Effects of Intracellular Signaling Pathway Inhibitors on SRM and AMD1 Expressions
We recently reported that MgCl 2 supplementation can activate a GSK3β/CREB pathway in HaCaT cells [17]. To clarify the regulatory mechanisms of SRM and AMD1 expressions, we investigated the effects of inhibitors for intracellular signaling cascades. MgCl 2 supplementation upregulated the mRNA levels of SRM and AMD1, which were completely blocked by CHIR99021, a GSK3 inhibitor, and Naphthol AS-E, a CREB inhibitor, in HaCaT cells ( Figure 4A). The activity of GSK3 could be regulated by p38/MSK1 and MEK/ERK pathways [19,20]. The MgCl 2 supplementation-induced elevation of SRM and AMD1 expressions were inhibited by U0126, a MEK inhibitor, but not by apigenin, a MSK1 inhibitor ( Figure 4B). These results indicate that MgCl 2 supplementation may activate GSK3 mediated by the MEK/ERK pathway in HaCaT cells.

Effect of MgCl2 Supplementation on Intracellular Signaling Pathways
The effects of inhibitors against intracellular signaling pathways were investigated by Western blotting. The p-ERK1/2 level was increased by MgCl2 supplementation, which was inhibited by U0126, but not by CHIR99021, in HaCaT cells ( Figure 5A). On the other hand, the p-GSK3β level was increased by MgCl2 supplementation, which was inhibited by both U0126 and CHIR99021. Active form p-CREB is translocated from the cytosol to the nucleus and activates the transcription of target genes [21]. The fluorescence intensity of p-CREB in the nuclei was increased by MgCl2 supplementation, which was inhibited by U0126, CHIR99021, and Naphthol AS-E ( Figure 5B). These results indicate that the nuclear localization of p-CREB may be upregulated by a MEK/ERK/GSK3β pathway in Ha-CaT cells.

Effect of MgCl 2 Supplementation on Intracellular Signaling Pathways
The effects of inhibitors against intracellular signaling pathways were investigated by Western blotting. The p-ERK1/2 level was increased by MgCl 2 supplementation, which was inhibited by U0126, but not by CHIR99021, in HaCaT cells ( Figure 5A). On the other hand, the p-GSK3β level was increased by MgCl 2 supplementation, which was inhibited by both U0126 and CHIR99021. Active form p-CREB is translocated from the cytosol to the nucleus and activates the transcription of target genes [21]. The fluorescence intensity of p-CREB in the nuclei was increased by MgCl 2 supplementation, which was inhibited by U0126, CHIR99021, and Naphthol AS-E ( Figure 5B). These results indicate that the nuclear localization of p-CREB may be upregulated by a MEK/ERK/GSK3β pathway in HaCaT cells.

Effect of MgCl2 Supplementation on Reporter Activities of SRM and AMD1
HaCaT cells were transiently transfected with SRM or AMD1 reporter plasmid plus the internal pSEAP2 control plasmid. The reporter activities of SRM and AMD1 were significantly increased by MgCl2 supplementation compared with those in control medium (0.8 mM MgCl2), and the values were 142.5 ± 13.2% (SRM) and 136.6 ± 2.5% (AMD1). The effects of MgCl2 supplementation were significantly inhibited by U0126, CHIR99021, and Naphthol AS-E ( Figure 6). These results are similar to those in real-time PCR and Western blotting analyses, indicating that the expressions of SRM and AMD1 may be upregulated by MgCl2 supplementation at the transcriptional level.

Effect of MgCl 2 Supplementation on Reporter Activities of SRM and AMD1
HaCaT cells were transiently transfected with SRM or AMD1 reporter plasmid plus the internal pSEAP2 control plasmid. The reporter activities of SRM and AMD1 were significantly increased by MgCl 2 supplementation compared with those in control medium (0.8 mM MgCl 2 ), and the values were 142.5 ± 13.2% (SRM) and 136.6 ± 2.5% (AMD1). The effects of MgCl 2 supplementation were significantly inhibited by U0126, CHIR99021, and Naphthol AS-E ( Figure 6). These results are similar to those in real-time PCR and Western blotting analyses, indicating that the expressions of SRM and AMD1 may be upregulated by MgCl 2 supplementation at the transcriptional level.

Protection for UVB-and H2O2-Induced Cell Damages by MgCl2 Supplementation
The viabilities of HaCaT cells were reduced by the treatments with UVB and H2O2 for 24 h, and the relative values were 60.4 ± 1.9% (100 mJ/cm 2 UVB), 35.3 ± 0.7% (150 mJ/cm 2 UVB), 54.9 ± 1.3% (250 mM H2O2), and 20.2 ± 2.3% (500 μM H2O2) ( Figure 7A). The UVB-and H2O2-induced cell damages were significantly rescued by preincubation with 5.8 or 10.8 mM MgCl2. Similar effects were observed by preincubation with 100 μM or 200 μM spermidine ( Figure 7B). To clarify the involvement of spermidine, the effect of cysteamine, a spermidine synthase inhibitor, was examined. The UVB-and H2O2-induced cell damages were significantly rescued by preincubation with 10.8 mM MgCl2, which were inhibited by the cotreatment with cysteamine ( Figure 7C). These results indicate that MgCl2 supplementation may alleviate the UVB-and H2O2-induced cell damages mediated through the production of spermidine in HaCaT cells.

Protection for UVB-and H 2 O 2 -Induced Cell Damages by MgCl 2 Supplementation
The viabilities of HaCaT cells were reduced by the treatments with UVB and H 2 O 2 for 24 h, and the relative values were 60.4 ± 1.9% (100 mJ/cm 2 UVB), 35.3 ± 0.7% (150 mJ/cm 2 UVB), 54.9 ± 1.3% (250 mM H 2 O 2 ), and 20.2 ± 2.3% (500 µM H 2 O 2 ) ( Figure 7A). The UVB-and H 2 O 2 -induced cell damages were significantly rescued by preincubation with 5.8 or 10.8 mM MgCl 2 . Similar effects were observed by preincubation with 100 µM or 200 µM spermidine ( Figure 7B). To clarify the involvement of spermidine, the effect of cysteamine, a spermidine synthase inhibitor, was examined. The UVB-and H 2 O 2 -induced cell damages were significantly rescued by preincubation with 10.8 mM MgCl 2 , which were inhibited by the cotreatment with cysteamine ( Figure 7C). These results indicate that MgCl 2 supplementation may alleviate the UVB-and H 2 O 2 -induced cell damages mediated through the production of spermidine in HaCaT cells.

Inhibition of UVB-Induced Production of ROS and Lipid Peroxide by MgCl 2 Supplementation
The production of ROS and lipid peroxide in HaCaT cells was detected using fluorescent probes DCF and Liperfluo, respectively. The fluorescence intensity of DCF was significantly increased by UVB treatment, and the relative value was 124.6 ± 4.8% ( Figure 8A). The UVB-induced elevation of ROS production was significantly inhibited by preincubation with 10.8 mM MgCl 2 , spermidine, or N-acetyl-cysteine (NAC). The effect of MgCl 2 supplementation was significantly blocked by the cotreatment with cysteamine. Similar results were observed in the measurement of Liperfluo ( Figure 8B). These results indicate that MgCl 2 supplementation may attenuate ROS-induced cell damage mediated through the production of spermidine in HaCaT cells. Then, the cells were treated with 100 and 150 mJ/cm 2 UVB or 250 and 500 μM H2O2. Cell viability was measured using CCK-8 and represented as a percentage of 0.8 mM MgCl2. n = 4-6. Error bars indicate SEM. ** p < 0.01 significantly different from 0.8 mM MgCl2. ## p < 0.01 significantly different from Veh. ϮϮ p < 0.01 significantly different from without Cys.

Inhibition of UVB-Induced Production of ROS and Lipid Peroxide by MgCl2 Supplementation
The production of ROS and lipid peroxide in HaCaT cells was detected using fluorescent probes DCF and Liperfluo, respectively. The fluorescence intensity of DCF was significantly increased by UVB treatment, and the relative value was 124.6 ± 4.8% ( Figure 8A). The UVB-induced elevation of ROS production was significantly inhibited by preincubation with 10.8 mM MgCl2, spermidine, or N-acetyl-cysteine (NAC). The effect of MgCl2 supplementation was significantly blocked by the cotreatment with cysteamine. Similar results were observed in the measurement of Liperfluo ( Figure 8B). These results indicate that MgCl2 supplementation may attenuate ROS-induced cell damage mediated through the production of spermidine in HaCaT cells.

Inhibition of UVB-Induced Production of ROS and Lipid Peroxide by MgCl2 Supplementation
The production of ROS and lipid peroxide in HaCaT cells was detected using fluorescent probes DCF and Liperfluo, respectively. The fluorescence intensity of DCF was significantly increased by UVB treatment, and the relative value was 124.6 ± 4.8% ( Figure 8A). The UVB-induced elevation of ROS production was significantly inhibited by preincubation with 10.8 mM MgCl2, spermidine, or N-acetyl-cysteine (NAC). The effect of MgCl2 supplementation was significantly blocked by the cotreatment with cysteamine. Similar results were observed in the measurement of Liperfluo ( Figure 8B). These results indicate that MgCl2 supplementation may attenuate ROS-induced cell damage mediated through the production of spermidine in HaCaT cells.  Protective effect of MgCl2 supplementatio Cells were pre-incubated in the presence of 0.8, 5.8, incubated in the absence (Veh) and presence of 100 were pre-incubated in the presence of 0.8 or 5.8 mM Then, the cells were treated with 100 and 150 mJ/cm was measured using CCK-8 and represented as a p indicate SEM. ** p < 0.01 significantly different from from Veh. ϮϮ p < 0.01 significantly different from wit

Inhibition of UVB-Induced Production of ROS Supplementation
The production of ROS and lipid peroxide cent probes DCF and Liperfluo, respectively. Th icantly increased by UVB treatment, and the re The UVB-induced elevation of ROS production tion with 10.8 mM MgCl2, spermidine, or N-a supplementation was significantly blocked by results were observed in the measurement of L that MgCl2 supplementation may attenuate RO the production of spermidine in HaCaT cells.

Discussion
Magnesium is essential for the maintenance of physiological homeostasis in the biological systems of animals. DNA microarray and KEGG pathway analyses in HaCaT cells showed that high MgCl 2 supplementation upregulates the expression levels of genes involved in cell cycle, DNA replication, and arginine and proline metabolism (Table S4). Mg 2+ plays a pivotal role in the regulation of enzymatic reactions including synthesis of DNA, RNA, and protein [3]. Therefore, it is natural that the genes involved in cell cycle and DNA replication were upregulated. Notably, we found that high MgCl 2 supplementation is involved in the regulation of arginine and proline metabolism in HaCaT cells.
Arginine is used for production of polyamines, nitric oxide, creatine, and urea. Polyamines play important roles in the maintenance of physiological function of skin. The production of polyamines is regulated by several enzymes including ODC, AMD1, and SRM ( Figure 1). MgCl 2 supplementation upregulates the expression levels of SRM and AMD1 (Figure 2A), as well as the production of polyamines ( Figure 3B). The promoter activity of ODC is upregulated by ERK1/2 and nuclear factor κB [22], CREB [23], and Sp1 [24], whereas it is downregulated by p38 MAPK [25] and Sp3 [24]. The literature concerning regulatory mechanisms of AMD1 and SRM are very limited. The expression level of AMD1 is upregulated by androgen [26], insulin [27], and polyamines [28]. The expression level of SRM is upregulated by dcSAM, a decarboxylated product of SAM catalyzed by AMD1. Here, we found that MgCl 2 supplementation upregulates the expression levels of SRM and AMD1 mediated by the activation of MEK/ERK/GSK3β/CREB pathway in HaCaT cells (Figures 4-6). The elevation of SRM and AMD1 mRNA levels were also observed in the supplementation of Mg-lactate and MgSO 4 ( Figure 2B). Therefore, the elevation of Mg 2+ concentration is suggested to be involved in the upregulation of these genes. At present, it is unknown as to how high Mg 2+ concentration can activate the signaling pathway. Mg 2+ has dual roles in the regulation of protein kinase reactions [29]. One is the nucleotide substrate such as MgATP. Another is the enzyme/metal-nucleotide complex, leading to the elevation of catalytic efficiency of enzymes. High Mg 2+ concentration affects steady-state kinetic parameters of various receptor and non-receptor protein kinases. The catalytic efficiency of ERK2 is increased by high Mg 2+ concentration [30]. The change of catalytic efficiency may be involved in the reaction caused by MgCl 2 supplementation in HaCaT cells.
The promoter region of the AMD1 gene contains the binding sites for transcriptional factors including AP-1, AP-2, CREB, SP-1, and multiple steroid receptors [31]. In contrast, there is no report concerning transcriptional factors of the SRM gene. The reporter vector of SRM used in the present study contains 1508 bp of promoter region of human SRM (HPRM44738). Binding site analysis with the TFBIND program reveals one putative CREB-binding site at the residues-1040/-1029 in the promoter region of SRM. The reporter activities of AMD1 and SRM were increased by MgCl 2 supplementation, which were completely inhibited by Naphthol AS-E, a CREB inhibitor ( Figure 6). CREB may function as transcriptional factor in MgCl 2 supplementation-induced elevation of AMD1 and SRM.
UVB irradiation to the skin of hairless mice induces the elevation of ODC activity, as well as generations of putrescine and spermidine [32]. However, the function of these polyamines has not been fully understood. The exposure of HaCaT cells to UVB (>100 mJ/cm 2 ) and H 2 O 2 (>250 µM) induced cell damage, which was partially rescued by pre-incubation of high concentration of MgCl 2 ( Figure 7A). In addition, the effect of MgCl 2 supplementation was canceled by cysteamine, a polyamine, and spermidine is able to protect the cells against UVB-and H 2 O 2 -induced damage ( Figure 7B,C). These results suggest that the elevation of polyamine levels is involved in the MgCl 2 supplementation-induced protection in HaCaT cells. Similarly, the cytoprotective effects of polyamines are reported in human skin fibroblasts [33]. Polyamines have various biological functions, including anti-inflammatory and antioxidant properties [8,9]. The antioxidant activity of spermine is over 30-fold that of vitamin E [34]. The UVB-induced generations of ROS and lipid peroxide were suppressed by the pretreatment of high Mg 2+ -or spermidine-containing media (Figure 8). Lipid peroxide induces a free radical chain reaction mechanism followed by elevation of lipid peroxidation. The production of lipid peroxidation in the human skin is increased by chronic sun light exposure and advancing age [35]. Lipid peroxidation is harmful to epidermal cells because it leads to a decrease in the cell viability [36] and induction of inflammatory response [37]. Therefore, the MgCl 2 supplementation may exert protective effects against UVB mediated through the production of polyamines and their antioxidative properties. Many antioxidants, including superoxide dismutase [38], glutathione reductase [39], and vitamin E [40], are destroyed by UV irradiation in the skin. MgCl 2 supplementation may be useful to maintain antioxidant capacity in the skin.
In conclusion, we found that the expressions of the spermidine-synthesis-related enzymes SRM and AMD1 are increased by a high concentration of Mg 2+ in HaCaT cells. The transcriptional activities of these genes were upregulated by the MSK1/GSK3/CREB pathway. The UVB-and H 2 O 2 -induced cell damages were attenuated by the pretreatment of high Mg 2+ -or spermidine-containing media. These results suggest that Mg 2+ supplementation is useful to enhance barrier function against UVB and ROS damages.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/cells11152268/s1, Table S1: List of genes with change over 2-fold in DNA microarray analysis.; Table S2: List of genes with change less than half in DNA microarray analysis. Table S3 Gene ontology analysis of differentially expressed genes in high MgCl2-treated cells.; Table S4 KEGG pathway analysis of differentially expressed genes in high MgCl 2 -treated cells; Figure S1 Original Images of Figure 1C; Figure S2 Original Images of Figure 2C; Figure S3 Original Images of Figure 8A.