**2. Results and Discussion**

#### *2.1. Purification and Monosaccharide Determination of Fucoidan Fractions*

Fucoidan fractions of LJSF1-LJSF4 were purified from *S. japonica* (Figure 1A), among which LJSF4 has the highest sulfate content of 30.72%. As shown in Figure 1B, its monosaccharide compositions were determined to be 79.49% of fucose and 16.76% of galactose, thus it is a fucose-rich fucoidan. In our recently published work, the structures of LJSF4 were analyzed by different spectroscopic methods and its excellent anti-inflammatory activities were studied in vitro and in vivo [25]. In order to achieve the application of this fucoidan, knowledge of its various bioactivities is required. Thus, its anti-UVB activity and mechanism both in vitro and in vivo were investigated in this study.

**Figure 1.** Purification and monosaccharide determination of fucoidan fractions: (**A**) elution profile of *S. japonica* crude fucoidan on DEAE-Sepharose Fast Flow anion exchange chromatography; (**B**) monosaccharide compositions of LJSF4 determined by gas chromatography.

#### *2.2. E*ff*ect of LJSF4 on UVB-Irradiated HaCaT Cells*

2.2.1. Effect of LJSF4 on Intracellular ROS Generation and Cell Death in UVB-Irradiated HaCaT Cells

Prior to evaluate the anti-UVB effect of LJSF4 on HaCaT cells, we detected its potential toxicity on HaCaT cells through the MTT viability assay. As shown in Figure 2A, the viability of HaCaT cells without LJSF4 treatment was 100%, and those treated with LJSF4 were above 90%. In addition, compared with the control group, cells treated with 50 μg/mL of LJSF4 even had higher cell viability. These results indicate that LJSF4 had no cytotoxicity on HaCaT cells and even exhibited positive effect on the cells at the concentration of 50 μg/mL.

Exposure of cells to UVB irradiation results in ROS excessive production and cellular damage, which is the cause of skin cancer and photoaging [26]. Therefore, suppressing the excessive production of ROS can protect the skin from the damage of UVB [27]. In this study, the HaCaT cell damage induced by UVB irradiation was examined by measuring the intracellular ROS generation and cell death. As shown in Figure 2B, the ROS level of UVB-irradiated cells was significantly increased compared to those of non-irradiated cells, while the ROS levels of those treated with different concentrations of LJSF4 were significantly decreased in a dose-dependent manner. In addition, the viability of cells irradiated by UVB was significantly decreased, and dose-dependently increased in LJSF4 treated cells (Figure 2C). These results indicate that LJSF4 possessed a protective effect against UVB-induced HaCaT cell damage. The anti-UVB activities of fucoidan extracted from other brown seaweed of *Costaria costata*, *Fucus evanescens* and *Undaria pinnatifida* have been investigated [22–24]. However, these seaweed resources are not as rich as *S. japonica,* which is the most abundant economic seaweed around the world [28]. In addition, fucoidan extracted from *C. costata* increase the cell viability by 1.77% and 4.94% at concentrations of 0.01 and 0.1 μg/mL, respectively, while the fucoidan was mildly cytotoxic at the concentration of 1 μg/mL [24]. However, LJSF4 was able to increase the cell viability by 16.14% at the concentration of 100 μg/mL with no cytotoxic effect. Therefore, LJSF4 possessed an excellent anti-UVB activity on HaCaT cells.

**Figure 2.** Protective effect of LJSF4 on UVB-induced HaCaT cells: (**A**) cytotoxicity of LJSF4 on HaCaT cells; (**B**) intracellular ROS level of UVB-irradiated HaCaT cells; (**C**) the viability of UVB-irradiated HaCaT cells. All experiments were performed in triplicate. Data are expressed as means ± standard error of the means (S.E.M). \* *p* < 0.05, \*\* *p* < 0.01 as compared to the UVB-treated group and # *p* < 0.05, ## *p* < 0.01 as compared to the control group.

#### 2.2.2. Effect of LJSF4 on Apoptosis in UVB-Irradiated HaCaT Cells

Apoptosis is a strong response of cells to UVB-induced damage. Cells eventually undergo apoptosis to eliminate severely damaged cells. However, even if serious damage occurs, some cells can still escape apoptosis, then turn into cancer cells. Studies have found that inhibiting apoptosis plays an important part in the formation of cancer [29].

In order to measure the protective effect of LJSF4 from UVB irradiation, HaCaT cells were stained with Hoechst 33342, which is a cell-permeable DNA dye. Later, the nuclear morphology of cells was observed by fluorescence microscopy. The cell images are shown in Figure 3. We found that the amount of apoptotic body of cells with UVB irradiation was almost 2.5 fold greater than that of cells without UVB irradiation, while those of cells pre-treated with 25 to 100 μg/mL of LJSF4 were decreased in a dose-dependent manner significantly. Masaki et al. reported that UVB-induced apoptosis was related to the increase in intracellular ROS generation. Inhibiting the production of ROS contributed to improving the intracellular defense against oxidative stress, thereby reducing cell apoptosis [27]. Our study evaluated the levels of ROS and apoptosis with or without LJSF4 treatment, and intervened with different concentrations of LJSF4. This result indicates that LJSF4 possessed excellent protective effect against UVB-induced HaCaT cell apoptosis.

**Figure 3.** The apoptotic body formation levels in UVB-irradiated HaCaT cells: (**A**) nuclear morphology of non UVB-irradiated HaCaT cells; (**B**) nuclear morphology of UVB-irradiated HaCaT cells; (**C**) nuclear morphology of cells treated with 25 of μg/mL LJSF4 and irradiated with UVB; (**D**) nuclear morphology of cells treated with 50 of μg/mL LJSF4 and irradiated with UVB; (**E**) nuclear morphology of cells treated with 100 of μg/mL LJSF4 and irradiated with UVB; (**F**) reactive apoptotic body formation. Apoptosis levels were measured using Image J software. All experiments were performed in triplicate. Data are expressed as means ± standard error of the means (S.E.M). \*\* *p* < 0.01 as compared to the UVB-treated group and ## *p* < 0.01 as compared to the control group.

2.2.3. Effect of LJSF4 on Bax/Bcl-xL and Cleaved Caspase-3 Levels in UVB-Irradiated HaCaT Cells

Apoptosis is a regulated mechanism of cell suicide, usually manifested as nuclear condensation, wrinkling, membrane foaming as well as chromosomal DNA fragmentation [29–31]. According to the previous studies, the formation of apoptosome is an intrinsic apoptotic signaling pathway which could activated by oxidative stress [32–34]. Excessive ROS production as a second messenger regulates apoptotic signaling pathway following UV irradiation. In this case, the balance of anti-apoptotic molecules of Bcl-xL and pro-apoptotic molecules of Bax indicated whether apoptosis is promoted or suppressed [35]. Eventually, their imbalance triggered the caspase cascade [36]. In addition, cleaved caspase-3 directly induces cell death as a critical executor of apoptosis [37].

The Bax/Bcl-xL and cleaved caspase-3 levels of UVB-irradiated HaCaT cells were measured by Western blot analysis. As shown in Figure 4, the UVB irradiation increased the Bax level and decreased the Bcl-xL level, while pre-treatment with LJSF4 significantly increased the Bcl-xL level and decreased the Bax level in a dose-dependent manner. Moreover, LJSF4 reduced UVB-induced cleaved caspase-3 activation in UVB-irradiated HaCaT cells. These results indicate that LJSF4 inhibited UVB-induced apoptosis by the regulation of Bax, Bcl-xL and cleaved caspase-3 levels. The anti-UVB activity of Sargachromenol extracted from the brown seaweed of *Sargassum micracanthum* has been investigated, and the similar pathway by regulation of Bax, Bcl-xL and cleaved caspase-3 levels has been reported [38]. In addition, fucoidans isolated from different brown algae have been reported to inhibit UVB-induced photoaging through other signaling pathways, such as by inhibiting the MAPK pathways related to NF-κB and AP-1, by inhibiting MMP-1 expression via blocking the signal pathways of p38, JNK, and ERK [23,39]. Thus, the *S. japonica* fucoidan showed a different anti-UVB pathway to those of other reported fucoidan via regulating the expression of Bax, Bcl-xL and cleaved caspase-3.

**Figure 4.** Effect of LJSF4 on Bax/Bcl-xL, and cleaved caspase-3 levels in UVB-irradiated HaCaT cells: (**A**) the effect of LJSF4 on UVB-induced apoptosis related protein expression; (**B**) the relative amounts of Bax/Bcl-xL and cleaved caspase-3 levels compared with GAPDH. The Bax level of control group and the cleaved caspase-3 level of experimental group treated with 100μg/mL of LJSF4 were not detected. All experiments were performed in triplicate. Data are expressed as means ± standard error of the means (S.E.M). \*\* *p* < 0.01 as compared to the UVB-treated group and ## *p* < 0.01 as compared to the control group.

#### *2.3. E*ff*ect of LJSF4 on UVB-Irradiated Zebrafish*

Zebrafish has become a popular animal model in the field of biological activity, due to the fact that its tissues and organs are very similar to mammals at the genetic, physiological, behavioral, and anatomical levels [40]. In previous reports, researchers have successfully used the zebrafish model to explore the protective effects of natural compounds on UVB-induced oxidative stress [41]. Therefore, we chose the same model to evaluate the protective effect of LJSF4 on UVB-irradiated photoaging in vivo. Zebrafish were pre-treated with LJSF4 of 25 to 100 μg/mL, respectively, and subsequently exposed to 50 mJ/cm<sup>2</sup> of UVB.

Excessive ROS produced by UVB stimulation plays a critical role in destroying keratinocytes through cell damage. ROS can be detected by using 2, 7-dichlorodihydrofluorescein diacetate (DCFH-DA) staining in living embryos. As shown in Figure 5A, the control group without exposure to UVB had no fluorescent generation, whereas that with exposure to UVB generated fluorescence, which indicated the generation of ROS on UVB-irradiated zebrafish. However, the zebrafish treated with different concentrations of LJSF4 showed a dose-dependent reduction in the production of ROS before being exposed to UVB irradiation. This result indicates that ROS levels decreased by 237.64% in zebrafish treated with 100 μg/mL of LJSF4. Then, cell death was determined by staining the embryos with acridine orange. The result indicates that the fluorescence intensity of the LJSF4 pretreatment group at the concentration of 100 μg/mL was decreased to almost the same level to that of the control group (Figure 5B). In addition, the protective effect of the fucoidan on zebrafish was determined against ROS-activated NO generation. As shown in Figure 5C, compared to that of the control group with no UVB irradiation, the production of NO was significantly increased 3.64 fold, while the pretreatment with LJSF4 inhibited the production of NO in a dose-dependent manner significantly. A similar result of dose-dependent inhibition for lipid peroxidation by LJSF4 treatment was also obtained (Figure 5D). This is similar to the results of previous studies, which reported that zebrafish exposed to UVB increased ROS generation, cell death levels, NO generation, and lipid peroxidation levels compared to that without UVB irradiation [12,42]. Furthermore, Wang et al. (2017) demonstrated that DPHC, isolated from *I. okamurae*, could decrease ROS levels by 85.21% with the concentration of 100 μM [41]. In short, the results show that LJSF4 extracted from *S. japonica* could excellently inhibit the generation of inflammation and reduce the destruction of cellular components by decreasing ROS levels, thereby further indicating the photoprotective effect of LJSF4 in zebrafish.

**Figure 5.** The effect of LJSF4 on UVB-irradiated zebrafish: (**A**) effect of LJSF4 on UVB-irradiated ROS generation in zebrafish; (**B**) effect of LJSF4 on UVB-irradiated cell death levels in zebrafish; (**C**) effect of LJSF4 on UVB-irradiated NO level in zebrafish; (**D**) effect of LJSF4 on UVB-irradiated lipid peroxidation levels in zebrafish. Zebrafish embryos at 2 days post-fertilization (dpf) were used for the anti-UVB study. Data are expressed as means ± standard error of the means (S.E.M). \*\* *p* < 0.01 as compared to the UVB-treated group and ## *p* < 0.01 as compared to the control group.
