Effect of Fermented Fish Oil on Fine Particulate Matter-Induced Skin Aging

Skin is exposed to various harmful environmental factors such as air pollution, which includes different types of particulate matter (PM). Atmospheric PM has harmful effects on humans through increasing the generation of reactive oxygen species (ROS), which have been reported to promote skin aging via the induction of matrix metalloproteinases (MMPs), which in turn can cause the degradation of collagen. In this study, we investigated the effect of fermented fish oil (FFO) derived from mackerel on fine PM (particles with a diameter < 2.5 µm: PM2.5)-induced skin aging in human keratinocytes. We found that FFO inhibited the PM2.5-induced generation of intracellular ROS and MMPs, including MMP-1, MMP-2, and MMP-9. In addition, FFO significantly abrogated the elevation of intracellular Ca2+ levels in PM2.5-treated cells and was also found to block the PM2.5-induced mitogen-activated protein kinase/activator protein 1 (MAPK/AP-1) pathway. In conclusion, FFO has an anti-aging effect on PM2.5-induced aging in human keratinocytes.


Introduction
Reactive oxygen species (ROS) have unpaired electrons and unstable bonds, which are properties that can lead to cellular damage [1,2] and the regulation transcription factors such as AP-1 and NF-kB [3,4]. Cellular ROS can accumulate in response to exogenous stimuli, including air pollution [4].
The skin is the largest organ in the human body and acts as the first-line defense barrier against harmful stimuli such as ultraviolet (UV) light and air pollution, including particulate matter (PM). Depending on particle size, PM can be classified as ultrafine (particles with a diameter < 0.1 µm: PM 0.1 ), fine (particles with a diameter < 2.5 µm: PM 2.5 ), and coarse (particles with a diameter < 10 µm: PM 10 ) [5]. PM can lead to the development of various skin diseases, including skin aging, alopecia, and skin cancer, through the induction of oxidative stress [6,7]. In addition, PM-induced oxidative stress promoted via the production of ROS and subsequent increase in the activity of matrix metalloproteinases (MMPs) [6,8], including MMP-1, MMP-2, and MMP-9, can cause skin aging as a consequence of the degradation of collagen [9,10].
The generation of ROS has been reported to affect skin aging by increasing the expression of MMP-1 in keratinocytes [11]. Several studies have reported that UVB-induced ROS causes skin photoaging via the promotion of the activity of MMP-1 in human keratinocytes and dermal

PM 2.5 -Induced Keratinocyte Senescence
Next, we measured β-galactosidase activity, a marker of cellular senescence [17], to detect HaCaT cell senescence using flow cytometry and confocal microscopy after SPiDER-βGal staining. PM 2.5 -treated cells showed increased β-galactosidase activity in the cytosol, whereas FFO treatment was found to decrease the PM 2.5 -induced β-galactosidase activity (Figure 2a,b).

Effect of FFO on PM2.5-Induced MMP-1 Activation and MMP Expression
In this study, we examined cellular MMP-1 activity, as MMP-1 is known to be primarily responsible for skin aging via collagen degradation [18]. The treatment of cells with PM2.5 significantly increased the activation of MMP-1 after 6, 12, and 24 h (Figure 3a), whereas pretreatment with FFO decreased the PM2.5-induced activation of MMP-1 (Figure 3b). The expression of MMP-1 mRNA and protein was also increased in PM2.5-treated cells, whereas it was decreased in the cells pretreated with FFO (Figure 3c,d). As MMP-2 and MMP-9 have also been reported to be involved in skin aging through the degradation of collagen [19], we analyzed the expression levels of MMP-2 and MMP-9 protein and found that these were also increased by PM2.5 treatment and decreased by FFO pretreatment (Figure 3e).

Effect of FFO on PM 2.5 -Induced MMP-1 Activation and MMP Expression
In this study, we examined cellular MMP-1 activity, as MMP-1 is known to be primarily responsible for skin aging via collagen degradation [18]. The treatment of cells with PM 2.5 significantly increased the activation of MMP-1 after 6, 12, and 24 h (Figure 3a), whereas pretreatment with FFO decreased the PM 2.5 -induced activation of MMP-1 (Figure 3b). The expression of MMP-1 mRNA and protein was also increased in PM 2.5 -treated cells, whereas it was decreased in the cells pretreated with FFO ( Figure 3c,d). As MMP-2 and MMP-9 have also been reported to be involved in skin aging through the degradation of collagen [19], we analyzed the expression levels of MMP-2 and MMP-9 protein and found that these were also increased by PM 2.5 treatment and decreased by FFO pretreatment (Figure 3e). Mar. Drugs 2019, 17 4

Effect of FFO on PM 2.5 -Induced MAPKs and Intracellular Ca 2+ Levels
MAPKs, which enhance the expression of MMP-1, are activated by an increase in the intracellular Ca 2+ levels [20]. Our western blot analysis indicated that PM 2.5 treatment induced the activation (phosphorylation) of extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK), whereas this activation was decreased following FFO pretreatment. Furthermore, PM 2.5 treatment induced the activation (phosphorylation) of MAPK kinase (MEK) 1/2 and SAPK/ERK kinase (SEK) 1 (Figure 4a), and significantly increased the levels of intracellular Ca 2+ , whereas FFO decreased the PM 2.5 -induced increase in Ca 2+ levels (Figure 4b,c).

Effect of FFO on PM2.5-Induced MAPKs and Intracellular Ca 2+ Levels
MAPKs, which enhance the expression of MMP-1, are activated by an increase in the intracellular Ca 2+ levels [20]. Our western blot analysis indicated that PM2.5 treatment induced the activation (phosphorylation) of extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK), whereas this activation was decreased following FFO pretreatment. Furthermore, PM2.5 treatment induced the activation (phosphorylation) of MAPK kinase (MEK) 1/2 and SAPK/ERK kinase (SEK) 1 (Figure 4a), and significantly increased the levels of intracellular Ca 2+ , whereas FFO decreased the PM2.5-induced increase in Ca 2+ levels (Figure 4b,c).

Effect of FFO on PM 2.5 -Induced Transcription Factor Activator Protein 1 (AP-1) Expression
The nuclear transcription factor, AP-1, regulated by MAPKs, increases the expression of MMPs [12,21]. The activation of MAPKs results in the heterodimerization of c-Jun/c-Fos and the formation of an AP-1 complex [10]. As shown in Figure 4, PM 2.5 treatment increased MAPK activity and intracellular Ca 2+ levels. Western blot analysis of c-Fos and phospho-c-Jun levels indicated that FFO significantly decreased the PM 2.5 -induced increase in phospho-c-Jun and c-Fos levels (Figure 5a). In addition, FFO reduced the PM 2.5 -induced AP-1 binding to the MMP-1 promoter (Figure 5b).

Effect of FFO on PM2.5-Induced Transcription Factor Activator Protein 1 (AP-1) Expression
The nuclear transcription factor, AP-1, regulated by MAPKs, increases the expression of MMPs [12,21]. The activation of MAPKs results in the heterodimerization of c-Jun/c-Fos and the formation of an AP-1 complex [10]. As shown in Figure 4, PM2.5 treatment increased MAPK activity and intracellular Ca 2+ levels. Western blot analysis of c-Fos and phospho-c-Jun levels indicated that FFO significantly decreased the PM2.5-induced increase in phospho-c-Jun and c-Fos levels (Figure 5a). In addition, FFO reduced the PM2.5-induced AP-1 binding to the MMP-1 promoter (Figure 5b).

Discussion
The skin is the largest organ in the human body and acts as a first-line defense barrier against harmful environmental stimuli, including PM. In several studies, PM 2.5 has been reported to have harmful effects associated with inflammatory skin diseases, skin aging, and damage to the respiratory system through the generation of intracellular ROS [6,22,23]. The accumulation of ROS has been reported to induce skin aging through the expression of MMPs, including MMP-1, MMP-2, and MMP-9 [24]. Accordingly, it is desirable to identify an effective antioxidant that could be used to prevent skin aging.
Previous studies have demonstrated that mackerel-derived FFO has a direct ROS scavenging effect and a protective effect against UVB-induced oxidative damage [16]. In the present study, we focused on the effect of mackerel-derived FFO against PM 2.5 -induced skin aging.
We initially measured the generation of intracellular ROS induced by PM 2.5 treatment, and accordingly found that pretreatment with FFO significantly reduced the induction of intracellular ROS by PM 2.5 treatment (Figure 1a,b). Given that the generation of ROS causes skin aging, we investigated keratinocyte senescence by examining the activity of the marker enzyme β-galactosidase [25]. We found that PM 2.5 treatment increased β-galactosidase activity, whereas FFO pretreatment significantly reduced the PM 2.5 -induced β-galactosidase activity (Figure 2a,b). As MMPs are known to cause skin aging through the degradation of collagen [9,10], we also examined the activity of MMP-1 and the expression levels of MMP-1, MMP-2, and MMP-9. Again, we found that FFO has an inhibitory effect on PM 2.5 -induced MMPs (Figure 3). These results indicate that PM 2.5 induces skin aging by promoting the expression of MMPs, whereas FFO has anti-aging effects via its ROS scavenging activity.
ROS activated MAPK signaling pathways, and the activation of MAPKs, in turn, induced various transcription factors, including AP-1 and NF-κB [14,26]. As a result of the translocation of activated AP-1, heterodimers comprising c-Jun and c-Fos and MMPs were synthesized [27,28]. As shown in Figure 4a, ERK and JNK were activated by PM 2.5 treatment, and this activation was reduced by FFO pretreatment. The activation of MEK and SEK, upstream of ERK and JNK, respectively, was also increased by PM 2.5 treatment, and this increase was again decreased by FFO pretreatment (Figure 4a). In addition, intracellular Ca 2+ levels, which regulate MAPKs, were significantly increased in PM 2.5 -treated cells, whereas pretreatment with FFO was found to decrease the observed PM 2.5 -induced increase in Ca 2+ levels (Figure 4b,c). Furthermore, phospho-c-Jun and c-Fos levels were increased in PM 2.5 -treated cells and decreased in cells receiving pretreatment with FFO ( Figure 5a). We also found that PM 2.5 -induced AP-1 binding to the MMP-1 promoter was reduced by FFO pretreatment (Figure 5b). These results therefore provide evidence that FFO can block the activation of the MAPK/AP-1 pathway induced by PM 2.5 in human keratinocytes. FFO showed an antioxidant effect in this study and in our previous study [16]. However, we did not assess whether FFO may induce the nuclear factor-erythroid 2-related factor 2-mediated antioxidant enzyme pathway to scavenge ROS. This will be evaluated in a future study.
In conclusion, mackerel-derived FFO has anti-aging effects against PM 2.5 -induced skin aging through a reduction in the generation of intracellular ROS and expression of MMPs (Figure 5c).

Detection of Intracellular ROS
For the detection of intracellular ROS in HaCaT cells, the cells were seeded in 6-well culture plates at a density of 1.0 × 10 5 cells/well, cultured for 16 h, and treated with 20 µg/mL FFO and 50 µg/mL PM 2.5 . After 30 min, 50 µM 2',7'-dichlorodihydrofluorescein diacetate (DCF-DA, Molecular Proves, Eugene, OR, USA) solution was added to the cells. After 15 min incubation at 37 • C, DCF fluorescence was measured using a BD LSRFortessa flow cytometer (PerkinElmer, Waltham, MA, USA). For imaging, cells were seeded in a 4-well glass chamber slide at a density of 1.0 × 10 5 cells/well, cultured for 16 h, and treated with 20 µg/mL FFO and 50 µg/mL PM 2.5 . After 30 min, 50 µM DCF-DA (Molecular Proves, Eugen, OR, USA) solution was added to the cells. After 15 min incubation at 37 • C, the cells were washed 3 times with PBS. DCF fluorescence images were obtained using a FV1200 laser scanning confocal microscope (Olympus, Tokyo, Japan).

Detection of β-Galactosidase Activity
β-galactosidase activity is a classic marker of cellular aging [30], and we accordingly measured the activity of this enzyme to detect cell senescence. Cells were seeded in 6-well culture plates at a density of 1.0 × 10 5 cells/mL, and after a 16-h incubation at 37 • C, they were treated with 20 µg/mL FFO and 50 µg/mL of PM 2.5 , followed by the addition of 2 µM SPiDER-βGal solution (Dojindo Molecular Technologies, Inc., Rockville, MD, USA) 24 h later. After a 15-min incubation at 37 • C, SPiDER-βGal fluorescence was measured using a BD LSRFortessa flow cytometer (PerkinElmer, Waltham, MA, USA). For imaging, cells were seeded in a 4-well glass chamber slide at a density of 1.0 × 10 5 cells/well, and after a 16 h incubation at 37 • C, they were treated with 20 µg/mL FFO and 50 µg/mL of PM 2.5 , followed by the addition of 2 µM SPiDER-βGal solution (Dojindo Molecular Technologies, Inc., Rockville, MD, USA) 24 h later. After a 15 min incubation at 37 • C, the cells were washed 3 times with PBS. After washing, the cells were treated with mounting medium containing DAPI to label nuclei. SPiDER-βGal fluorescence images were obtained using an FV1200 laser scanning confocal microscope (Olympus, Tokyo, Japan).

MMP-1 Activity
MMP-1 activity was measured using a Fluorokine ® E human active MMP-1 fluorescent assay kit (R&D Systems Inc., Minneapolis, MN, USA). HaCaT cells were seeded in a 60 mm culture dish at 1.0 × 10 5 cells/mL. After a 16 h incubation at 37 • C, the cells were treated with 20 µg/mL FFO, and 1 h thereafter with 50 µg/mL PM 2.5 . MMP-1 activity was assessed according to the manufacturer's instructions. Fluorescence was measured using a Spectra Max i3x microplate reader (Molecular Devices, San Jose, CA, USA).

Reverse Transcription-PCR (RT-PCR)
Cells were seeded in a 60 mm culture dish at 1.5 × 10 5 cells/mL, and after a 1 h incubation at 37 • C, cells were treated with 20 µg/mL FFO and 50 µg/mL PM 2.5 . After 24 h, we isolated total RNA from cells using an easy-BLUE™ total RNA extraction kit (iNtRON Biotechnology Inc., Seongnamsi, Korea). Subsequently, cDNA was amplified using reverse transcription reaction buffer, primers, dNTPs, and Taq DNA polymerase in a final volume of 20 µL. The amplified products were mixed with 6× blue/orange loading dye and resolved by electrophoresis on 1% agarose gels. Gels were subsequently stained with RedSafe™ nucleic acid staining solution (iNtRON Biotechnology Inc., Seongnamsi, Korea), photographed under a UV light, and analyzed using Image Quant™ TL analysis software (Amersham Biosciences, Uppsala, Sweden). The PCR conditions used were as follows: an initial denaturation at 94 • C for 5 min, followed by 30 cycles of 94 • C for 30 s, 55 • C for 30 s, and 72 • C for 1 min. The primers used in this study were as follows: human MMP-1: forward (5'-GGAGGAAATCTTGCTCAT-3') and reverse (5'-CTCAGAAAGAGCAGCATC-3'); human GAPDH: forward (5 -TCAAGTGGGGCGATGCTGGC-3') and reverse (5 -TGCCAGCCCCAGCGTCAAAG-3').

Western Blot Analysis
Protein lysates (30 µg per lane) were electrophoresed on 12% SDS-polyacrylamide gels, and then transferred to nitrocellulose membranes, which were incubated with the primary antibodies and subsequently with HRP-conjugated secondary antibodies (Invitrogen, Carlsbad, CA, USA). Thereafter, the membranes were developed using a western blotting detection kit (GE Healthcare Life Sciences, Buckinghamshire, Little Chalfont, UK) for detection of the protein bands and then exposed to X-ray film. The primary antibodies used in this study were as follows:

Measurement of Ca 2+ Levels
To determine cellular Ca 2+ levels, cells were seeded in 6-well culture plates at a density of 1.0 × 10 5 cells/well, cultured for 16 h, and then treated with 20 µg/mL FFO and 50 µg/mL PM 2.5 . After 24 h, 5 µM Fluo-4-AM (Molecular Probes, Eugene, OR, USA) solution was added to the cells. After 15 min incubation at 37 • C, Fluo-4-AM fluorescence was measured using a BD LSRFortessa flow cytometer (PerkinElmer, Waltham, MA, USA). For imaging, cells were seeded in a 4-well glass chamber slide at a density of 1.0 × 10 5 cells/well, cultured for 16 h, and then treated with 20 µg/mL FFO and 50 µg/mL PM 2.5 . After 24 h, 5 µM Fluo-4-AM (Molecular Probes, Eugen, OR, USA) solution was added to the cells. After a 15 min incubation at 37 • C, the cells were washed 3 times with PBS. After washing, Flou-4-AM fluorescence images were obtained using an FV1200 laser scanning confocal microscope (Olympus, Tokyo, Japan).

Chromatin Immunoprecipitation (ChIP) Assay
ChIP assays were performed using a SimpleChIP™ enzymatic chromatin IP kit (Cell Signaling Technology, Danvers, MA, USA). HaCaT cells were seeded in a 100 mm culture dish at 1.5 × 10 5 cells/mL, and after 16 h, were treated with 20 µg/mL FFO and 50 µg/mL PM 2.5 . All processes were performed according to the manufacturer's instructions. As an antibody, we used a c-Jun antibody (Invitrogen, Carlsbad, CA, USA). The following primers were designed from the MMP-1 gene promoter (−67 to +94 of the MMP-1 gene sequence from the transcription starting site; Bionics): sense 5 -CCTCTTGCTGCTCCAATATC-3 and antisense 5 -TCTGCTAGGAGTCACCATTTC-3 . PCR products were separated on 1% agarose gels. DNA bands were photographed under a UV light and were analyzed using Image Quant™ TL analysis software version 1.0.2 (Amersham Biosciences, Uppsala, Sweden).

Statistical Analysis
All analyses were performed in triplicate and all values are expressed as the mean ± standard error of the means. The results were assessed using one-way analysis of variance with Tukey's test using the SigmaStat version 3.5 software (Systat Software Inc., San Jose, CA, USA) to determine the statistical significance of differences between means. A p-value < 0.05 was considered to be statistically significant.

Conflicts of Interest:
The authors declare no conflict of interest.