The dried root bark of E. sessiliflorus (Wuga bark) was purchased from Linding Herbal Plantation Ltd. (China); DPPH, ABTS, and PTIO reagents were obtained from MACKLIN (China); thiazolyl blue tetrazolium bromide (MTT), 3,3′-diaminobenzidine (DAB), and DCFH-DA were purchased from Sigma-Aldrich (United States); DMEM, fetal bovine serum (FBS), and 0.25% trypsin were purchased from Gibco (United States); DAPI and TUNEL kits were obtained from Beyotime (China); hydroxyproline (HYP) assay kits were purchased from Nanjing Jiancheng (China); neutral protease (50,000 U/g), RIPA lysate, H&E and Masson staining kits, MMP-1 and MMP-9 ELISA kits, and hydrogen peroxide (H₂O₂) were obtained from Solarbio (China); and all other chemicals were obtained from Beijing Chemical Factory (China).

The dried root barks of E. sessiliflorus (Wuga bark) were ground into dried defatted Eleutherococcus powder; the crude protein in Eleutherococcus was extracted by alkaline extraction and acid precipitation; the enzyme conditions were as follows: neutral protease, pH 7.0, an enzyme time of 2 h, an enzyme temperature of 45°C, a liquid–solid ratio of 120:1, and an enzyme additive quantity of 3200 U/g so as to obtain the protein enzyme solution of E. sessiliflorus; The peptides of different molecular weights were separated and purified by filtration chromatography on Sephadex G-25 and an AKTA protein purification chromatograph (GE-100, United States), and the molecular weights of the peptides were initially determined by tricine-SDS-PAGE gel analysis.

The precise molecular weight and amino acid sequence of the peptide were identified by LC-MS/MS; the E. sessiliflorus peptide with the identified amino acid sequence was synthesized by Dangang Biotechnology Co., Ltd. (China).

We analyzed the free radical scavenging ability of each peptide using chemical antioxidant assays like DPPH, ABTS, and PTIO. Briefly, the DPPH solution was added to the peptides for 30 min at room temperature (RT), the OD value was measured at 519 nm, and the radical scavenging rate of DPPH was calculated; the ABTS solution was added to the peptides for 10 min at RT, the OD value was measured at 734 nm, and the radical scavenging rate of ABTS was calculated; the PTIO solution was added to the peptides for 24 h at 37°C, the OD value was measured at 565 nm, and the radical scavenging rate of PTIO was calculated.

To the sample, 250 μM FeCl₂ was added. The mixture was placed in a water bath at 50°C for 30 min, and 500 μM phenanthrozine solution was added. The mixture was left to stand at RT for 5 min. The OD value was measured at 562 nm, and the ferrous ion chelation rate was calculated.

Human immortalized keratinocytes (HaCaT) were obtained from NewgainBio (Cat. No. CH1031), and the cells were cultured in DMEM with 10% FBS in a 5% CO₂ incubator at 37°C. We first determined the optimal dose of P-8-R in the pre-test and then divided the experimental groups into four groups: control, P-8-R, UVB, and UVB + P-8-R. Control: cells were not exposed to UVB; P-8-R: cells were not exposed to UVB and treated with 0.2 μg/mL P-8-R for 24 h; UVB: as mentioned below; UVB + P-8-R: cells were pretreated with 0.2 μg/mL P-8-R for 1 h and then treated with P-8-R for 24 h after UVB radiation.

We slightly modified the UVB modeling method (Ding et al., 2023). Briefly, we used UVB lamps (PL-S 9W; Royal Dutch Philips Electronics Ltd., Holland, wavelength range: 300–320 nm; peak value: 311 nm) and a UV intensity detector (TM213; Tenmars Electronics Ltd., Taiwan). For UVB irradiation, the DMEM in the cell plate was discarded and PBS (pH 7.3) was added, and the cell plate was placed at a vertical distance of 10 cm from the UV light source for irradiation with a power of 500 μW/cm², an irradiation time of 60 s, and a final intensity of 30 mJ/cm². After UVB irradiation, DMEM was added again, and the sample was incubated for 24 h.

Cell viability was determined by the MTT reduction assay. Briefly, after the treatment period, the cells were incubated with 0.5 mg/mL MTT for 4 h. The formazan precipitate was dissolved in 200 μL DMSO, and the absorbance at 570 nm was detected by using a microplate reader (BioTek, United States). The cell viability was expressed as a percentage of the control.

HaCaT cells were seeded in 6-well plates and incubated for 24 h at 37°C with 5% CO₂. After the treatment period, the HaCaT cell morphology was observed by using an inverted optical microscope (Olympus, Japan).

The DCFH-DA assay was used to quantify ROS production (Liu et al., 2023). Briefly, after the treatment period, the DCFH-DA was added to the treated plates at a final concentration of 10 μM, which were then incubated in the dark at 37°C for 30 min. Fluorescence images were obtained by using an upright fluorescence microscope (Olympus, Japan), and relative fluorescence was measured using ImageJ 1.6 software.

The comet assay was used to analyze the recombinant protein for UVB-induced DNA damage (Chaiprasongsuk et al., 2019). Briefly, after placing the first layer of the HaCaT cell suspension, 0.7% agarose was added and solidified at 4°C for 10 min. The slides were then placed in the pre-cooled lysis buffer, lysed at 4°C for 2 h, and electrophoresed for 20 min. The slides were then removed and placed in PBS to neutralize the strong base. Each slide was stained with EB and then observed under a fluorescence microscope. The percentage of DNA content in the comet tail was analyzed using the CASP software. Tail DNA % was used as an index of DNA damage (Duan et al., 2019).

The levels of MMP-1 and MMP-9 in the supernatant of HaCaT cells were measured using MMP-1 and MMP-9 ELISA kits (Cat. No. SEKH-0251 and SEKH-0257; Solarbio). Briefly, after the treatment period, the supernatant was collected and centrifuged at 3,000 g for 10 min to discard the precipitate. The standard curves of MMP-1 and MMP-9 were then plotted, and the absorbance of the supernatant was detected at 450 nm by using a microplate reader (BioTek, United States).

IF staining was used to detect the specific expression of MMP-1 and MMP-9 in the cells. We treated HaCaT cells with 4% paraformaldehyde fixation, 0.2% Triton X-100 permeabilization, and 1% BSA closure and finally added the following primary antibodies: MMP-1 (1:200; Cat. No. AF0231; Beyotime) and MMP-9 (1:200; Cat. No. K001664P; Solarbio) and then incubated the cells overnight at 4°C. The following day, goat anti-rabbit IgG-FITC secondary antibody (1:500; Cat. No. A0562; Beyotime) was added, and finally the nuclei were stained with DAPI, and images were obtained.

This animal experiment was approved by the Laboratory Animal Ethics Committee of Beihua University. BALB/c mice (male, weight: 20–25 g, n = 60) were purchased from Changchun Yisi Laboratory Animal Technology Co. All mice were kept in a temperature-controlled environment with a 12-h light/dark cycle with free access to water and food. After 1 week, the mice were depilated on their backs over an area of 6 cm² until the skin was completely exposed. The optimal dose of P-8-R was determined to be 0.1 mg/cm² by a pre-test. After that, the mice were randomly divided into four groups. In the control group, saline was applied directly without UVB irradiation. In the P-8-R group, only 0.1 mg/cm² P-8-R was applied. In the UVB group, the UVB modeling method was slightly modified (Divya et al., 2015). Briefly, the backs of the mice were irradiated at a vertical distance of 5 cm from the UVB source with an irradiation power of 1,000 μW/cm² and a final intensity of 120 mJ/cm² three times per week for 4 weeks. For the UVB + P-8-R group, 0.1 mg/cm² P-8-R was injected into the backs of the mice after 1 week, 2 h after each UVB irradiation. At the end of the experiment, all mice were euthanized, and the dorsal skin tissues were excised for various experiments.

The amounts of H₂O₂ and HYP in the skin are measured using the appropriate kits [(Cat. No. A030-2-1; Solarbio) and (Cat. No. A030-2-1; Nanjing Jiancheng), respectively]. Briefly, the supernatant from the centrifugation of the skin tissue homogenate is reacted with appropriate reagents. Finally, the absorbance of the reactants at the corresponding wavelength is measured with a microplate reader (BioTek, United States), and the levels are calculated.

The skin tissue from the back of mice was fixed in 4% paraformaldehyde, dehydrated through a graded ethanol series, cleared in xylene, embedded in paraffin, and sectioned at 5 μm by using a microtome (Leica, Germany). Tissue sections were stained with hematoxylin and eosin (H&E) for epidermal hyperplasia studies and with Masson’s stain for collagen fiber analysis. All staining procedures were performed according to their respective staining schemes without modification. Histopathological changes were examined under a light microscope (Olympus, Japan).

TUNEL staining of the skin was carried out using an in situ cell death detection kit (Cat. No. C1091; Beyotime). Briefly, the sections were digested with proteinase K at 37°C for 20 min, and then the TUNEL reaction mixture was incubated at 37°C in the dark. The sections were then coverslipped with Converter-POD at 37°C for 30 min and then incubated with DAB for 3 min to observe the reaction. Finally, the sections were counterstained with hematoxylin for 30 s. The staining result was expressed as TUNEL-positive.

We used IHC to detect the specific expression of MMP-1 and MMP-9 in skin tissues. After skin sections were deparaffinized, hydrated, cleared of endogenous peroxidase, antigenically repaired, and closed with 5% BSA, MMP-1 and MMP-9 primary antibodies were added and incubated overnight at 4°C in the dark. Goat anti-rabbit IgG-HRP (1:300; Cat. No. SE134; Solarbio) was added the following day, followed by DAB staining, hematoxylin re-staining, and differentiation. Finally, the sections were observed and photographed using an orthogonal microscope.

After treatment, HaCaT cells and skin tissue were extracted using RIPA lysis solution (Cat. No. P00138; Beyotime) containing 1% PMSF to extract proteins and quantified using BCA protein kits (Cat. No. PC0020; Solarbio). Protein samples (35 μg) were separated by SDS-PAGE using A 12% resolving gel and then transferred to PVDF membranes (Millipore, United States). The PVDF membranes were blocked for 1 h in PBS containing 1% BSA, and the PVDF membrane was incubated overnight at 4°C with primary antibodies, including MMP-1, MMP-9, Bax (1:1,000; Cat. No. 2772; CST), Bcl-2 (1:1,500; Cat. No. ab196495; Abcam), cleaved caspase-3 (1:1,000; Cat. No. 9661; CST), and β-actin (1:5,000; Cat. No. AF5003; Beyotime). The membranes were then incubated with the appropriate goat anti-rabbit secondary antibodies for 1 h at RT, followed by three washes in TBST. Proteins were detected using the ECL reagent. The loading protein was normalized to β-actin, and the target protein were exposed using a high-sensitivity chemiluminescence imaging system (Bio-Rad, United States).

Data were expressed as mean ± SD and were statistically analyzed with Prism software (GraphPad Software Inc., United States) and analyzed with ANOVA for single-factor analysis. The Student’s t-test was used for inter-group comparison. p < 0.05 was considered significant.

We purified three peptides, P1, P2, and P3 (Figure 1B), from the hydrolysate of E. sessiliflorus neutral protease using Sephadex G-25. Tricine-SDS-PAGE and MS analyses showed that the molecular weights of P1, P2, and P3 were 1,327.5 Da, 1,228.6 Da, and 781.9 Da, respectively. The amino acid sequences of P1, P2, and P3 were P1: FNNRTKKHPW (F-10-W), P2: PHWWEYRR (P-8-R), and P3: GKKTWY (G-6-Y), respectively (Figure 1C,E). In addition, we evaluated the ability of three peptides to scavenge oxygen and nitrogen radicals by the in vitro chemical antioxidant assay (DPPH, ABTS, and PTIO), and the experimental results are shown in Figures 2A–D. P-8-R has the strongest ability to scavenge oxygen and nitrogen radicals and can be used for subsequent antioxidant pharmacological studies.

To investigate the antioxidant effect of P-8-R, we established a UVB-induced HaCaT cell model. The MTT results are shown in Figure 3C. The cell viability of the cells in the UVB group was decreased, while P-8-R was able to significantly increase the cell viability. The results of cell morphology analysis showed that both the control and P-8-R groups had normal cell morphology without any abnormal changes; the cells in the UVB group showed morphological changes such as cell shrinkage, cell membrane folds, and cytoplasmic tightness, but the cell morphology in the UVB + P-8-R group did not change remarkably compared with that in the UVB group, which indicated that P-8-R had a strong cytoprotective effect. In order to study the mechanism of action of P-8-R, we examined the oxidative stress and apoptosis of HaCaT cells by DCFH-DA staining and comet assay, respectively, and the results are shown in Figure 3D. Compared with the control group, the fluorescence intensity of intracellular DCF in the UVB group was significantly enhanced, but P-8-R could obviously weaken the fluorescence intensity of intracellular DCF, which indicated that P-8-R could protect HaCaT cells from oxidative stress by scavenging ROS. After that, we detected the cellular DNA damage by comet assay, and the results showed that the trailing cell rate of cells in the UVB group increased with DNA breakage and structural damage, while the trailing cell rate of cells in the UVB + P-8-R group decreased, which indicated that P-8-R was able to inhibit UVB-induced DNA damage in HaCaT cells. Additionally, we detected the key proteins in the mitochondria-dependent apoptosis pathway by Western blotting, and the results are shown in Figure 3G. The protein expression rate of Bcl-2/Bax in cells of the UVB group was significantly decreased, and the protein expression of cleaved caspase-3 was remarkably elevated. However, P-8-R could increase the protein expression rate of Bcl-2/Bax and downregulate the protein expression of cleaved caspase-3, which indicated that P-8-R could inhibit HaCaT cell apoptosis through the mitochondria-dependent apoptosis pathway.

We examined the levels of MMP-1 and MMP-9 in the supernatants of HaCaT cells, and the results are shown in Figure 4. Compared with the control group, the levels of MMP-1 and MMP-9 in the cell supernatants of the UVB group were remarkably increased, whereas compared with the UVB group, the levels of MMP-1 and MMP-9 in the cell supernatants of the UVB + P-8-R group were notably decreased, indicating that P-8-R was able to inhibit UVB-induced MMP-1 and MMP-9 protein overexpression in skin epidermal cells and reduce the protein secretion of MMP-1 and MMP-9.

The results of H&E staining and epidermal thickness measurements are shown in Figure 5B, C. The skin structure of mice in the control and P-8-R groups was intact, and the stratum corneum, epidermis, and dermis were clearly visible, with clear epidermal–dermal boundaries. In the UVB group, the skin of mice was significantly thickened, and the epidermal–dermal boundary disappeared, indicating that the UVB-induced ROS were capable of contributing to pathological changes in the skin. The mice in the UVB + P-8-R group showed a decrease in the thickness of the skin and epidermis, and the epidermal–dermal boundary was clearer. To further study the damage caused by oxidative stress in the skin, we detected the level of H₂O₂ in the skin, and the results showed that P-8-R was able to remarkably reduce the level of H₂O₂ in the skin and prevent the damage caused by oxidative stress in the skin (Figure 5D). Moreover, TUNEL and Western blotting results showed that P-8-R could inhibit skin cell apoptosis through the mitochondria-dependent apoptotic pathway, as opposed to the in vitro experiments. In short, P-8-R could reduce the pathological changes caused by UVB-induced oxidative stress in mouse skin through antioxidant and anti-apoptotic effects.

We detected the collagen density and content in the skin of mice by Masson staining as well as the HYP kit, and the results showed that P-8-R could keep the collagen density of the skin basically normal (Figure 6A–D) (Figure 7). We further examined the expression and secretion of MMP-1 and MMP-9 in the skin of mice. As shown in Figure 6E–H, P-8-R could notably downregulate the UVB-induced overexpression and secretion of MMP-1 and MMP-9, which suggests that P-8-R could suppress the overexpression of MMP-1 and MMP-9 proteins, thereby reducing the loss of collagen in mouse skin and preventing the skin from photoaging. All in all, P-8-R could inhibit UVB-induced oxidative stress in mouse skin and provide protection against aging caused by collagen loss.