Saussurea involucrata polysaccharide was provided by Infinitus Co., Ltd. (Guangzhou, China). PPAR-α inhibitor (GW6471) was purchased from Selleck (Shanghai, China).

The homogeneity and molecular weight of SIP were measured using a gel permeation chromatography-eighteen angles laser light scattering instrument (GPC-MALLS) (Wyatt Technology Corporation, Santa Barbara, CA, United States). In brief, 10 mg of SIP was placed in 1 mL of .1 M NaNO₃ solution in a small beaker, which was then placed on a magnetic stirrer (Crystal Technology & Industries, Dallas, TX, United States) and stirred overnight to dissolve the SIP. The SIP sample was diluted fourfold with .1 M NaNO₃ solution and then filtered through a .22-μm nylon filter (Millipore Corp., Billerica, MA, United States). The filtered sample was then subjected to gel permeation chromatography using a GPC-MALLS system. The samples were resolved by three tandem columns (300 × 8 mm, Shodex OH-pak SB-805, 804 and 803; Showa Denko K.K., Tokyo, Japan) with .1 M NaNO₃ solution used as a mobile phase under a flow rate of .4 mL/min. The data were collected and analyzed with the Astra 6.1 software (Wyatt Technology Corporation, Santa Barbara, CA, United States).

The monosaccharide composition of SIP was analyzed using high-performance anion-exchange chromatography (HPAEC). In brief, 1 mL of 2 M TFA was added to 5 mg of polysaccharide sample, and the mixture was incubated in a 121°C oil bath for 2 h. The sample was soaked in methanol and blown dry with nitrogen to remove residual TFA, and then freeze-dried. The residue was re-dissolved in deionized water and filtered through a .22-μm microporous membrane filter for HPAEC analysis. The sample extracts were analyzed by high-performance anion-exchange chromatography (HPAEC) using a CarboPac PA-20 anion-exchange column (3 by 150 mm; Dionex) connected to a pulsed amperometric detector (PAD; Dionex ICS 5000 system). The column was run at a flow rate of .5 mL/min. Data were acquired by an ion chromatography system (ICS5000, Thermo Scientific) and processed using chromeleon 7.2 CDS (Thermo Scientific).

To make the hydrogel for topical application, .6 g carbopol, 5 g glycerinum, and 3 g propylene glycol were mixed in 20 g of water and slowly stirred until to obtain a uniform suspension. Next, .6 g of trietha-nolamine (TAE) was added to the suspension and with slow stirring to produce a hydrogel matrix. To make the SIP hydrogel, .1 or .5 g of SIP was dissolved in 80 g of water, and the SIP solution was slowly added to the hydrogel matrix to generate a homogenous mixture containing either .1% or .5% SIP. These SIP-containing hydrogel preparations will henceforth be referred to as .1% SIP hydrogel and .5% SIP hydrogel.

Healthy male C57BL/6 mice (seven-week-old, 18–22 g body weight) were purchased from Gempharmatech Co., Ltd. (Jiangsu, China). The mice were kept at 22°C ± 1°C with 50%–55% relative humidity and a 12-h light/dark cycle. After 1 week of acclimatization, the hairs on the dorsal skin were shaved and the mice were then randomly divided into four groups. One group was designated as the healthy control group, and it received no treatment. The other three groups were treated with UVB, and one of these groups was treated with hydrogel only, whereas the other two groups were each treated with either the .1% or .5% SIP hydrogel. Hydrogel was completely wiped from the dorsal skin of mice prior to UVB irradiation, and then the animals were exposed to UVB irradiation at a dose of 500 mJ/cm² for five consecutive days and the daily transdermal water loss (TEWL) was determined using a VapoMeter (Delfin Technologies, Finland). Besides, a cuticle hydration meter (MoistureMeter Epid, Delfin Technologies, Finland) was used to determine epidermal water content. During these 5 days of exposure to UVB, hydrogel or SIP hydrogel was applied to the dorsal skin twice daily, and the skin was photographed with a camera prior to the daily UVB exposure.

All animal experiments were conducted according to international ethical guidelines and the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and all protocols were approved by the Animal Care and Use Committee of Wenzhou Medical University.

The fixed slices were treated with 3% H₂O₂ and then blocked with 5% BSA. The specimens were finally incubated with anti-Filaggrin (Santa Cruz, California, United States) or anti-Involucrin (1: 500; Absin, Shanghai, China) antibody overnight at 4°C, and then incubated with horseradish peroxidase-conjugated secondary antibody at 37°C for 4 h followed by reaction with 3,3-diaminobenzidine (DAB).

The slices were first treated with 3% H₂O₂ and then blocked with 5% BSA. Next, the slices were treated with the proliferation marker Ki67 (Cell Signaling Technology, Beverly, MA, United States) overnight at 4°C. After that, the slices incubated with Alexa Flour 488 anti-rabbit secondary antibody at 37°C for 1 h. The immunofluorescence assays for γ-H2AX and CPD used γ-H2AX (1:1,000) and anti-CPD (1:200) (Cell Signaling Technology, Beverly, MA, United States) as primary antibodies. The immunofluorescence assays for K16 used Keratin 16 (1:500) (Proteintech, Chicago, IL, United States) as primary antibody. Finally, samples were stained with antifade reagent containing DAPI (Invitrogen, Life Technologies, California, United States) and examined with confocal laser scanning microscope (Olympus, Tokyo, Japan).

The skin samples were embedded in Tissue Tek OCT medium (Sakura, Tokyo, Japan) and stored at −80°C until completely solidified, after which the samples were cut into 10-μm slices. The slices were incubated with Hoechst solution (Dojindo, Tokyo, Japan) and then in Nile Red Fluorescent Dye Working Solution (Glpbio, Montclair, CA, United States). Finally, the samples were placed in glycerin gelatin and examined under confocal laser scanning microscope.

HaCaT cells were treated without or with SIP (5, 10, 20, and 50 μg/mL) for 24 h, and then washed with PBS. The cells were then exposed to UVB (200 mJ/cm²) in PBS using a VL6-M Biotronic device (Vilber Lourmat, Marne La Vallee, France). After UVB exposure, the cells were incubated with fresh DMEM at 37°C in a 5% CO₂ incubator for differenttimes depending on the experiment.

HaCaT cells were treated without or with SIP (5, 10, 20, and 50 μg/mL) for 24 h followed by exposure to UVB (200 mJ/cm²) in PBS. After that, the cells were stained with annexin V & FITC apoptosis detection kit (Dojindo, Tokyo, Japan) and then quantified with a flow cytometer equipped with an ACEA NovoCyte (Agilent, Santa Clara, CA, United States).

HaCaT cells were pretreated without or with SIP (5, 10, 20, 50 μg/mL) for 24 h, followed by exposure to UVB (200 mJ/cm²) in PBS. After washing with PBS, the cells were incubated for 6 h in DMEM, fixed in 4% paraformaldehyde and permeabilized in .3% Triton X-100. This was followed by blocking with 5% BSA-and overnight incubation with γ-H2AX and anti-CPD or Ki67 at 4°C. After that, the cells were finally treated with an appropriate secondary antibody and mounted with an antifade reagent.

HaCaT cells were pretreated without or with SIP (5, 10, 20, 50 μg/mL) for 24 h, followed by exposure to UVB (200 mJ/cm²) in PBS. The cells were then washed with PBS and incubated for 6 h in DMEM. After that, the cells were stained with a Calcein-AM/PI double staining kit (Dojindo, Tokyo, Japan) and then examined under a fluorescence microscope.

HaCaT cells were cultured in a 6-well plate in the presence of 10 mM Ca²⁺ for 24 h and then treated without or with SIP (5, 10, 20, 50 μg/mL) for 24 h followed by exposure to UVB (200 mJ/cm²). The cells were then washed with PBS and incubated for 6 h in DMEM followed by oil-red O staining using a commercial staining kit (Solarbio, Beijing, China).

Total RNA from a skin sample was extracted with TRIzol^® Reagent (Thermo Fisher Scientific, Waltham, MA, United States) whereas total RNA form HaCaT cells was extracted with an RNA isolation kit (Biomiga, San Diego, CA, United States) according to the manufacturer’s protocols using. For each sample, 1 μg of total RNA was reverse transcribed using Prime Script RT reagent Kit (Takara, Dalian, China). Quantitative real-time PCR (qPCR) was performed on an LC96 system (Roche, Basel, Switzerland) using the SYBR Green Master Mix (Applied Biosystems, Foster City, CA). The comparative Ct approach was utilized to assay the mRNA relative levels while the endogenous control was GAPDH while the2^−ΔΔCt method was employed to quantify the relative target gene expression. The sequences of the primers used are shown in Table 1.

Extracts of a skin sample or HaCaT cells were prepared and then centrifuged at 12,000 × g to precipitate the insoluble materials. The supernatant was retained and the concentration of protein in the supernatant was measured by a BCA Protein Assay Kit (Beyotime, Jiangsu, China). A sample of supernatant was resolved by SDS-PAGE using a 12% gel. The protein bands in the gel were transferred to a polyvinylidene difluoride (PVDF) membrane and blocked with 5% skim milk, followed by overnight incubation with anti-Filaggrin, anti-Involucrin, anti-PPARα, anti-PPARγ, anti-LXR (Proteintech, Chicago, IL, United States), or anti-ABCA1 (Absin, Shanghai, China). After that, the blot was washed and incubated with the appropriate secondary antibody (Cell Signaling Technology, Beverly, MA, United States). Finally, the blot was subjected to a detection assay using a chemiluminescence substrate (Pierce, Rockford, IL, United States), and images of the blot were acquired using an Amersham Imager (GE Healthcare Biosciences, Pittsburgh, PA, United States).

GraphPad Prism 6.0 software (GraphPad, San Diego, CA, United States) was used for all statistical analyses. The statistical differences between groups were performed using one-way analysis of variance (ANOVA), and correction for multiple comparisons was made using Dunnett’s test.

The molecular weight of SIP was determined to be 120 kDa according to GPC-MALLS analysis (Figure 1A). The HPAEC chromatogram for the mixed standard monosaccharide is shown in Figure 1B. The horizontal coordinate is the retention time (Time, min), and the vertical coordinate is the ion response value (Response, nC). According to the retention time of standard monosaccharides, SIP was found to be primarily composed of galacturonic acid, arabinose, glucose, rhamnose, and galactose (44.21: 16.37: 12.19: 10.06: 9.7). (Figure 1C).

To investigate the photoprotective effect of SIP against UVB radiation and its healing effects on the recovery of the damaged epidermal barrier, the shaved dorsal skin of mice was exposed to repeated UVB radiation followed by topical application of hydrogel without or with SIP. Compared with the UVB group, SIP hydrogel application dramatically reduced dryness, peeling, scaling, and erythema caused by UVB radiation (Figure 2A). Additionally, decreased UVB-induced trans epidermal water loss was also evident as revealed by TWEL measurement (Figure 2B). After repeated UVB irradiation, HE staining of skin samples revealed abnormal epidermal thickening and the development of hyperplasia-like symptoms. The epidermis thickness was considerably lowered by increasing the dose of the SIP hydrogel applied (Figure 2C). Immunofluorescence analysis of the epidermal cells in the proliferation cycle tagged with the proliferation-related antigen Ki67 revealed a substantially higher number of Ki67-labeled cells in the UVB group compared with the healthy control group. However, for SIP hydrogel group, the number of Ki67-labeled cells was significantly reduced (Figure 2D). Abnormal proliferation of epidermal cells usually leads to epidermal hyperplasia. To further investigate the effect of SIP on epidermal hyperplasia induced by UVB, the marker of epidermal hyperplasia K16 was stained. Immunofluorescence analysis showed that compared with the UVB group, SIP hydrogel treatment significantly reduced the fluorescence intensity of k16 in epidermal, suggesting that SIP could effectively prevent UVB-induced epidermal hyperplasia (Figure 2E). The dynamic balance of keratinocyte proliferation and differentiation maintains the epidermal barrier, which is necessary for skin moisturization. To test if SIP could hasten the epidermal barrier repair process following UVB exposure, the expression levels of the keratinocyte differentiation markers, filaggrin and involucrin were measured by immunohistochemistry and western blot. The expression of these proteins in the epidermis was inhibited by UVB, but the inhibition was attenuated following the application of SIP hydrogel (Figures 2F, G). SIP kept the skin moist following UVB-irradiation by reducing UVB-induced dehydration and hyperplasia of the epidermis, enabling it to prevent or reduce the extent of UVB-induced disorders in epidermal proliferation and differentiation, thereby assisting with the recovery of the barrier function.

Another major mechanism for UVB-induced skin dryness is the induction of oxidative stress, which can damage cellular DNA and trigger severe inflammatory reactions. The degree of DNA damage can be evaluated by measuring the level of CPD, the major photoproduct of UVB-induced DNA damage, and γ-H2AX, a DNA damage-sensing molecule. Compared with the healthy control group, a significantly higher γ-H2AX and CPD content was found in the UVB group, where the two markers were present in 60% of the epidermal cells. However, both γ-H2AX and CPD contents were significantly reduced in the .1% or .5% SIP hydrogel group, where only 40% of the epidermal cells displayed the two markers (Figure 3A). Many inflammatory lesions are characterized by abnormal aggregation and activation of mast cells, which will release many cytokines such as IL-1 and TNF-α as a result of their own activation, leading to chronic inflammation. UVB radiation led to abnormal aggregation of mast cells in the skin, while mast cell aggregation was significantly reduced in the SIP hydrogel group (Figure 3B). Furthermore, SIP hydrogel application also lowered the mRNA levels of the pro-inflammatory factors IL-1β and TNF-α in a dose-dependent manner (Figure 3C). Taken together, the findings indicated that SIP could protect the skin against UVB-induced DNA damage while decreasing UVB-induced inflammatory responses.

Disorder in the synthesis of epidermal lipids can lead to the loss of the epidermal permeability barrier, which will consequently interfere with the ability of the skin to maintain its moisture and lead to skin dryness. The majority of epidermal intercellular lipids are neutral lipids, and the metabolism of neutral lipids is essential to the modulation of the skin permeability barrier. Nile red is a lipophilic fluorescent dye that is typically used to identify neutral lipids. A lower content of neutral lipids was found in the epidermis of the UVB group relative to that of the healthy control group, whereas for the SIP hydrogel group, the content of neutral lipids was comparable with that of the healthy control group (Figure 4A). This suggested that SIP could prevent the loss of epidermal intercellular lipid induced by UVB, thereby, maintaining the epidermal barrier. Peroxisome proliferator-activated receptors (PPARs) and liver X receptors (LXRs) are important regulators of keratinocyte proliferation and differentiation, as well as lipid synthesis. UVB radiation inhibits the expression of PPAR-α in the epidermis, and this may be the underlying mechanism for UVB-induced skin dryness. SIP antagonized the inhibitory effect of UVB on PPAR-α protein as shown by western blot analysis (Figure 4B). These data indicated that SIP might contribute to mitigating effect on UVB-induced skin dryness through the inhibition of disorder in epidermal lipid synthesis, the regulation of epidermal proliferation and differentiation, and the regulation of lipid synthesis via PPAR-α.

Oxidative stress is the most direct injury sustained by keratinocytes following exposure to UVB irradiation, and ROS are byproducts of UV-absorbing cells. The production of ROS and the occurrence of oxidative stress in HaCaT cells induced by UVB irradiation were both reduced following treatment with SIP as revealed by flow cytometry (Figure 5A). In line with the ROS results, SIP treatment reduced the γ-H2AX and CPD content of UVB-irradiated HaCaT cells in a dose-dependent manner (Figure 5B). Furthermore, qPCR results of IL-1β and TNF-α showed that SIP reversed the UVB-induced upregulation of mRNA levels of pro-inflammatory factors, indicating that SIP could relieve the UVB-induced inflammatory response in HaCaT cells (Figure 5C). Ultraviolet radiation has a cytotoxic effect on skin cells, leading to cell death. The use of calcein-AM/PI fluorescence staining on HaCaT cells revealed more dead cells after UVB radiation and a lower proportion of life/death ratio compared with the healthy control group (Figure 5D). Treatment with a high dose of SIP (20, 50 μg/mL) was found to lower the number of dead cells while increasing the life/death ratio by more than twofold. Besides, Ki67 fluorescence staining revealed a lower number of Ki67-labeled cells for the UVB irradiated cells compared to the non-treated cells. However, upon SIP treatment, the number of Ki67-labeled cells increased for the UVB irradiated cells (Figure 5E). Overall, SIP could reduce UVB-induced keratinocyte damage, including oxidative stress and inflammation responses, thereby preventing cell death and protecting cell proliferation-associated activity.

To elucidate the mechanism by which SIP might have prevented the UVB-induced disorder in epidermal intercellular lipid synthesis, the neutral lipid content in HaCaT cells was examined by Nile red fluorescence and oil-red O staining. Both staining revealed reduced lipid content in HaCaT cells following UVB irradiation, but the irradiated cells that were also treated with SIP displayed a normal level of lipid content (Figures 6A, B). Activation of the PPAR-α pathway has previously been shown to increase keratinocyte differentiation and boost lipid production, thus increasing epidermal barrier function. Therefore, we examined the expression levels of PPAR-α and its downstream target genes filaggrin, involucrin, and ABCA1. UVB was found to inhibit the protein expression of PPAR-α and its target genes, but SIP could prevent this effect (Figure 6C). ELOVLs are key enzymes in the synthesis of long-chain fatty acids. Compared with the healthy control group, UVB treatment caused a reduction of ELOVL1, ELOVL4, and ELOVL6 mRNA levels in the cells, but UVB + SIP treatment not only prevented the loss of these mRNA levels (Figure 6D). High concentrations of SIP also increased the mRNA levels of ELOVL1 and ELOVL4 beyond the levels of the control. The result indicated that SIP could abolish the inhibitory effect of UVB on the expression of genes involved in fatty acid synthesis. SIP could prevent the occurrence of keratinocyte differentiation and lipid synthesis disorder induced by UVB, and the PPAR-α signaling pathway may be involved in the overall protective effect of SIP against UVB-induced skin dryness.

To elucidate the molecular mechanism by which SIP might have alleviated UVB-induced skin dryness, HaCaT cells were treated with SIP in the presence of a specific PPAR-α inhibitor (GW6471) and then exposed to UVB irradiation (200 mJ/cm²). After blocking PPAR-α, the protein levels of ABCA1 or the mRNA levels of ELOVL1, ELOVL4, and ELOVL6 in the SIP-treated HaCaT cells did not increase, but the levels of filaggrin and involucrin still increased (Figures 7A, B). Furthermore, blocking PPAR-α also resulted in the loss of the ability of SIP to prevent the UVB-induced decrease in neutral lipid (Figure 7C). Thus, it appeared that the protective effect of SIP against UVB-induced inhibition of keratinocyte differentiation and lipid synthesis might be entirely or partially dependent on the activation of PPAR-α.

Through the action of PPAR-α, SIP could regulate the differentiation of keratinocytes and the disorder of epidermal intercellular lipid synthesis in UVB-irradiated HaCaT cells. To further investigate the role of PPAR-α in the protective effect of SIP against UVB-induced skin dryness, HaCaT cells were treated with SIP in the presence of a specific PPAR-α inhibitor (GW6471) and then exposed to UVB irradiation (200 mJ/cm²). These cells displayed no increase in ROS production or γ-H2AX and CPD contents as shown by flow cytometry analysis and immunofluorescence analysis, respectively (Figures 8A, B). Besides, in the presence of GW6471, SIP treatment did not reduce the mRNA levels of IL-1β and TNF-α in UVB-irradiated HaCaT cells (Figure 8C). SIP also did not reduce the number of dead cells increased by UVB irradiation or increase the number of Ki67-labeled cells decreased by UVB irradiation in the presence of GW6471 as confirmed by calcein-AM/PI fluorescence and Ki67 fluorescence staining (Figures 8D, E). The findings suggested that the involvement of PPAR-α in the anti-oxidative and anti-inflammatory activities of SIP might form an important mechanism by which SIP could reduce the extent of UVB-induced cell death.