Human hTERT-immortalized foreskin fibroblast (BJ-5ta) cells were purchased from ATCC (CRL-4001) and maintained in a 4:1 mixture of DMEM and Medium 199 (WelGENE Inc.), supplemented with 10% fetal bovine serum (FBS; HyClone; Cytiva) and 1% penicillin/streptomycin at 37°C in a humidified incubator with 5% CO₂. To isolate the exosomes, the cells were incubated in serum-free DMEM. The conditioned medium was then collected and centrifuged at 300 × g for 10 min and 2,000 × g for 20 min at 4°C, followed by filtration to remove the cells and cellular debris using 0.22-μm filters. The clarified supernatant was collected and concentrated using a sterile-membrane T-series cassette (Pall Life Sciences) with tangential flow filtration using a membrane with a molecular weight cut-off (MWCO) of 100 kDa (Sartorius AG). The mixture was then centrifuged at 100,000 × g for 3 h at 4°C (Fig. S1). The morphology of the BJ-5ta Exo was imaged using a field-emission scanning electron microscope (FE-SEM; Sigma HD, Carl Zeiss Meditec AG). The size distribution was determined by nanoparticle tracking analysis (NTA) using ZetaView (Particle Metrix). In total, two positive exosome markers, CD63 and ALIX, were used, while Calnexin was used as the negative protein marker.

Prior to UVB irradiation, the cells were treated with BJ-5ta Exo at a concentration of 10⁴ particles/ml for 6 h in a serum-free 4:1 mixture of DMEM and Medium 199 (WelGENE Inc.). UVB irradiation was performed at 30 mJ/cm² for <1 min in a thin layer of phosphate-buffered saline (PBS) using a UVB-emitting system from Biospectra (Vilber Lourmat Sto). Following irradiation, the cells were incubated in serum-free medium, and both the cells and cell supernatants were used for further analysis.

The cells were seeded and cultured in 96-well plates (Corning, Inc.) until they reached a confluency of 90%. The cells were treated with BJ-5ta Exo (0, 10³, 10⁴, 10⁵, 10⁶ and 10⁷ particles/ml) for 24 h or exposed to various doses of UVB radiation (20, 30, and 40 mJ/cm²), followed by incubation at 37°C with 5% CO₂ for 24 h. In addition, the cells were treated with BJ-5ta Exo (10⁴ particles/ml) for 6 h before being exposed to UVB radiation (30 mJ/cm²) and then incubated at 37°C with 5% CO₂ for 24 h. Cell viability was assessed using a WST-8 assay kit (QuantiMax™, Biomax). The absorbance was measured at 450 nm using a microplate spectrophotometer (SpectraMax 340; Moleular Devices, Inc.).

The intracellular production of ROS was measured using the Cellular ROS Detection Assay kit (cat. no. ab113851, Abcam). The cells were pre-treated with various concentrations of BJ-5ta Exo for 6 h, washed with 1X assay buffer (Abcam), and incubated in 20 μM in 2,7-dichlorofluorescin diacetate (DCFDA) solution (Abcam) (100 μl) for 45 min at 37°C with 5% CO₂ in the dark. Subsequently, the cells were exposed to UVB radiation (30 mJ/cm²) and incubated in complete medium containing 10% fetal bovine serum (FBS), but lacking phenol red (WelGENE Inc.) for 2 h. The samples were observed using a fluorescence microscope (DMi8, Leica Microsystems GmbH), and fluorescence readings were obtained using a spectrophotometer (SpectraMax 340; Molecular Devices, Inc.) at wavelengths of 485 and 535 nm.

SA-β-gal staining was performed according to the instructions provided with the SA-β-gal Staining kit (cat. no. 9860, Cell Signaling Technology, Inc.). After washing with PBS (pH 6.0), the cells were fixed in 4% paraformaldehyde (4% PFA) for 10 min at room temperature (RT) and stained at 37°C for overnight in a dry incubator without CO₂. The SA-β-gal-positive cells were observed using an optical microscope (DMi8, Leica Microsystems GmbH), and the count was determined by examining 400 cells per dish using the Image Pro Plus (IPP) 6.0 software (Media Cybernetics, Inc.). The proportions of cells exhibiting SA-β-gal activity are presented as percentages of the total cells count in each dish.

The BJ-5ta fibroblasts were pre-treated with BJ-5ta Exo at a concentration of 10⁴ particles/ml for 1 h and then irradiated with UVB at a dose of 30 mJ/cm². Following 24 h of incubation at 37°C with 5% CO₂, the cells were fixed with cold ethanol (70%) and then treated with propidium iodide (PI; MilliporeSigma) and RNase A (Thermo Fisher Scientific, Inc.). The cell cycle was assessed using a BD FAC Symphony A1 flow cytometer (BD Biosciences) and analyzed using FlowJo software v10.

An Annexin V-FITC apoptosis detection kit (V13241, Invitrogen; Thermo Fisher Scientific, Inc.) was used to determine the number of apoptotic cells. The cells were pre-treated with BJ-5ta Exo at concentrations of 10⁴ particles/ml for 1 h and then irradiated with UVB (30 mJ/cm²). Following exposure, the cells were incubated for 24 h at 37°C with 5% CO₂, and then collected and centrifuged at 252 × g for 10 min at RT, washed three times with cold PBS and resuspended in 100 μl binding buffer solution. Finally, the cells were incubated with Annexin V-FITC (5 μl) and PI (5 μl) at room temperature for 15 min in the dark. The fluorescence of the cells was immediately assessed using a flow cytometer (BD Biosciences). In the FACS diagram, the early apoptotic cells and the late apoptotic cells are respectively represented in the lower right quadrant and upper right quadrant. The total apoptotic rates were calculated using the following formula: Total apoptosis rate (%)=early apoptosis rate + late apoptosis rate.

Gallic acid (GA), a phenolic antioxidant found in numerous types of plants, a positive control was used due to its antioxidant activity (27). The BJ-5ta fibroblasts were pre-treated with BJ-5ta Exo or GA at a concentration of 10⁴ particles/ml for 6 h and then irradiated with UVB at a dose of 30 mJ/cm². Following 1 h of incubation at 37°C with 5% CO₂, total RNA was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Single-strand cDNA synthesis was performed using reverse transcription using PrimeScript TM RT Master Mix (Takara Bio, Inc.). The resulting cDNA was subjected to qPCR on a CFX96 thermocycler (Bio-Rad Laboratories, Inc.) using qPCR PreMIX SYBR-Green (Enzynomics). The following thermal cycling conditions were used: PCR initial activation step for 15 min at 95°C; three-step cycling: Denaturation for 10 sec at 95°C, annealing for 15 sec at 60°C, elongation for 30 sec at 72°C; for 45 cycles. Gene expression levels were calculated as a cycle threshold (Ct) value using the 2^−ΔΔCq quantification method and normalized to that of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (28). The primers used for qPCR are listed in Table I.

Whole protein lysates were extracted from BJ-5ta cells using RIPA buffer (Thermo Fisher Scientific, Inc.), and the protein content was quantified using Bradford reagent (MilliporeSigma). Equal amounts of protein (10 μg) were separated on a 10% SDS-PAGE gel and transferred to nitrocellulose membranes (Cytiva, Amersham, United States), which were then blocked in 5% skim milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) for 2 h at RT and probed overnight at 4°C with primary antibodies listed in Table II. The membranes were then incubated with HRP-conjugated anti-mouse (1:5,000, PI-2000-1, Vector Laboratories, Inc.) or anti-rabbit (1:5,000,PI-1000-1, Vector Laboratories, Inc.) secondary antibodies at room temperature for 1 h. Immunodetection was performed using an Amersham ECL kit (GE Healthcare; Cytiva) according to the manufacturer's protocol. The protein bands were visualized using a ChemiDoc™ MP Imaging System (Bio-Rad Laboratories, Inc.) and analyzed using ImageJ software V1.8.0 (National Institutes of Health).

The cells were fixed with 4% PFA for 30 min, washed with PBS, blocked with 3% bovine serum albumin (BSA) and 0.2% Triton X-100 in PBS at RT for 1 h, and incubated overnight at 4°C with primary antibodies listed in Table II. After washing with PBS, the cells were incubated with anti-rabbit IgG-FITC secondary antibodies (1:3,000, ab6717, Abcam) for 1 h at RT in the dark. The cell nuclei were counterstained with 4°C counterstained 4′,6-diamidino-2-phenylindole (DAPI; cat. no. AR-6501-01, ImmunoBioScience Corp.) at RT for 30 min, and the stained cells were observed using a confocal microscope (LSM 880, Zeiss AG).

Female SKH-1 hairless mice (7 weeks old, 17-22 g) were purchased from Saeron Bio, Inc. The mice were acclimatized for 1 week under the following conditions: A temperature of 23±2°C, 55±10% humidity, and a 12-h-light/12-h-dark cycle. All animal experiments were conducted in accordance with the principles of laboratory animal care at the National Institutes of Health and with the approval of the Ethics Committee for Laboratory Animals at Chung-Ang University (IACUC no. A2022053). The mice were randomly assigned to four groups (n=6 per group) as follows: Group 1, normal control; group 2, UVB only; group 3, UVB + BJ-5ta Exo 10⁶ (particles/ml); and group 4, UVB + BJ-5ta Exo 10⁸ (particles/ml). The mice were exposed to UVB irradiation using a BIO-SPECTRA (Vilber Lourmat Sta), three times a week for 8 weeks. Initially, the dose was set at 100 mJ/cm² for 2 weeks, and this was then increased to 150 mJ/cm² for 4 weeks. In addition, the mice were irradiated with UVB at 150 or 200 mJ/cm² for 2 weeks. Immediately after UVB irradiation, 100 μl BJ-5ta Exo were injected subcutaneously into the backs of the mice, three times a week. The normal mice were subcutaneously injected with saline three times a week. The wrinkles on the backs of the mice were photographed using a DSLR camera (Canon EOS 700D) and PRIMOS CR (SnTLab).

The mice were euthanized by CO₂ asphyxiation. The mice were placed in a new cage, and immediately euthanized by the displacement of air with 100% CO₂ (30% chamber volume/min), within 5 min and decapitated for tissue collection (29,30).

Morphological changes in the dorsal skin were observed using a DSLR camera (Canon EOS 700D) and PRIMOS CR (SnT Lab Co., Ltd.). at 12 weeks. Mouse skin roughness (RA) was analyzed using PRIMOS CR software version 5.8E (SnT Lab Co., Ltd.). Skin biopsies were fixed in 10% formalin for 24 h. Paraffin-embedded 3-μm-thick sections were cut, mounted on POLYSINE^® Slides (Thermo Fisher Scientific, Inc.), dewaxed in xylene, and then dehydrated in an ethanol series. Hematoxylin and eosin (H&E) staining was performed to examine the histological features and skin thickness. Masson's trichome (MT) staining and Verhoeff-van Gieson (VVG) staining were performed to evaluate collagen and elastin fibers, respectively. H&E staining was performed as follows: Hematoxylin solution was applied for 5 min, followed by eosin solution (Muto Pure Chemical Co., Ltd.) for 2 min, both at RT. Additionally, Masson's trichrome staining was conducted using the following steps: The mordant solution for 25 min, 0.75% orange G solution (Muto Pure Chemical Co., Ltd.) for 1 min, Masson's B stain solution for 25 min, 2.5% phosphotungstic acid solution for 20 min, and aniline blue (Muto Pure Chemical Co., Ltd.) for 8 min, all at RT. For VVG staining, the procedure included an initial incubation in Verhoeff's solution for 1 h, followed by washing with tap water. Subsequently, samples were exposed to 2% FeCl3 (MilliporeSigma) for 5 min, followed by treatment with 5% sodium thiosulfate (MilliporeSigma). For counterstaining, slides were stained with VVG's solution for 5 min. The samples subsequently underwent dehydration using 95% alcohol, followed by two changes of 100% alcohol. For IHC, the sliced sections were subjected to antigen retrieval with Tris-EDTA at 4°C for 15 min, incubated with BLOXALL blocking Solution (Vector Laboratories, Inc.) at RT for 30 min, and then incubated overnight at 4°C with the primary antibodies listed in Table II, including procollagen type I, collagen type I, MMP-1, elastin and filaggrin antibodies. The slides were then incubated with HRP using the ImmPRESS^® Excel Amplified Polymer Staining kit (Vector Laboratories, Inc.). The staining was developed using the 3,3′-diaminobenzidine (DAB) peroxidase substrate kit (Vector Laboratories, Inc.). To identify nuclei, the slides were counterstained with hematoxylin at RT for <1 min. The stained slides were photographed using a slide scanner (Pannoramic MIDI; 3DHISTECH Ltd.), observed using Case Viewer software (V 2.7; 3DHISTECH Ltd.), and analyzed using ImageJ software (V1.8.0; National Institutes of Health).

The hydration of the dorsal skin was assessed using a Corneometer^® CM 825 (Courage + Khazaka Electronic GmbH). The amount of transepidermal water loss (TEWL) was measured using a Tewameter^® TM 300 (Courage + Khazaka Electronic GmbH) after 8 weeks.

The reconstructed human skin model Neoderm^®-ED was purchased from Tego Science, Inc. Neoderm^®-ED was removed from medium-containing agar and transferred onto 12-well plates for equilibration at 37°C (5% CO₂) for 1 day. Under treatment with BJ-5ta Exo, Neoderm^®-ED was irradiated with UVB (128 mJ/cm²), and the tissue samples were incubated at 37°C with 5% CO₂ for 48 h.

The quantitative measurement of procollagen-type I and MMP-1 production from reconstructed human skin (Neoderm^®-ED) in the supernatant was conducted using a procollagen type I ELISA kit (cat. no. MK101, Takara Bio, Inc.) and a Human MMP-1 (Sandwich ELISA) ELISA kit (cat. no. ab215083, Abcam), following the manufacturer's instructions.

Data are presented as the mean ducted using a procollagen type I ELISA kit independent experiments. Data analyses were performed using unpaired one-way analysis of variance (ANOVA) followed by the Bonferroni post hoc test. Statistical analysis was performed using GraphPad Prism 7.0 software (GraphPad Software Inc.). All experiments were repeated at least three times.

BJ-5ta-Exo exhibited a spherical morphology (Fig. 1A). The number and size of the total particles were quantified using NTA as follows: The mean size of the exosomes was 215.4±116.1 nm and the number was 2.33×10¹⁰±2.22×10⁹/ml (Fig. 1B). The two positive markers for the exosomes (ALIX and CD63) were abundant in the BJ-5ta-Exo, while the negative protein marker, calnexin, was absent in the BJ-5ta Exo (Fig. 1C). The BJ-5ta cells were treated with BJ-5ta Exo (10³ to 10⁷ particles/ml) for 24 h, and cell viability was then measured using WST-8 assay. BJ-5ta Exo at doses of up to 10⁶ particles/ml was not toxic to the BJ-5ta cells (Fig. 1D). UVB irradiation inhibited cell viability by 36 and 46% at doses of 30-40 mJ/cm², respectively (Fig. 1E). However, BJ-5ta Exo (10⁴ particles/ml) prevented the inhibition of cell viability induced by UVB (Fig. 1F). In addition, UVB significantly decreased the percentage of cells expressing Ki67. Conversely, BJ-5ta Exo reversed this trend (Fig. 1G).

UVB-mediated ROS generation can trigger genes related to skin photoaging and lead to oxidative stress. UVB significantly increased intracellular ROS levels compared to the control group, while BJ-5ta Exo attenuated ROS generation induced by UVB (Fig. 2A and B). UVB irradiation induced a decrease in the mRNA levels of antioxidant enzymes (SOD-1, SOD-2, CAT and GPX); however, BJ-5ta Exo prevented the downregulation of these genes (Fig. 2C). Furthermore, UVB irradiation resulted in the downregulation of NF-E2-related factor 2 (Nrf2), which is responsible for inducing the transcription of antioxidant and cytoprotective genes. However, BJ-5ta Exo attenuated this decrease in Nrf2 expression (Fig. 2D). In addition, it was confirmed that BJ-5ta Exo inhibited the UVB-mediated decrease of Nrf2 in the nucleus (Fig. 2E). These results thus suggested that BJ-5ta Exo suppressed UVB-induced oxidative stress by increasing the expression levels of Nrf2.

UVB exposure caused an increase in γH2AX expression, while BJ-5ta Exo suppressed this increase (Fig. 3A). Using ICC staining (Fig. 3B), it was found that BJ-5ta Exo inhibited the UVB-induced γH2AX foci and nuclear accumulation. BJ-5ta Exo also prevented the UVB-induced upregulation of RAD51 and cleaved PARP-1 (Fig. 3C). In addition, the apoptotic cells were analyzed using Annexin V and PI solution followed by flow cytometric analysis. UVB irradiation increased total apoptosis by 20.9% compared to the control group. However, BJ-5ta Exo significantly reduced apoptosis, resulting in an 11.39% inhibition compared to the UVB-alone group (Fig. 3D). Cell cycle analysis also revealed that UVB irradiation decreased the cell populations in the G0/G1 phase (from 82.73±1.38 to 62.30±5.31%). However, BJ-5ta Exo partially prevented the G2-phase cell cycle alteration compared with the UVB group (from 16.90±4.20 to 9.32±0.41) (Fig. 3E). Subsequently, it was confirmed that BJ-5ta Exo efficiently reversed the UVB-mediated increase in p53 and p21 expression (Fig. 3F). The number of cells expressing SA-β-Gal and the expression levels of p16 were also significantly increased after UVB irradiation compared to the control cells; however, these effects were attenuated by BJ-5ta Exo (Fig. 3G and H).

UVB increased COX-2 and iNOS expression (4.0-fold and 1.4-fold, respectively, compared to the control group), while BJ-5ta Exo significantly reduced the COX-2 and iNOS levels by 0.3-fold compared to the UVB-only group (Fig. 4A). The levels of p-NF-kB (Ser⁵³⁶) were found to be elevated by UVB, although this effect was reversed by BJ-5ta Exo (Fig. 4B). Furthermore, ICC staining confirmed that BJ-5ta Exo significantly inhibited the UVB-induced translocation of NF-κB to the nucleus (Fig. 4C).

BJ-5ta Exo suppressed the UVB-induced decrease in collagen type I levels, while increasing MMP-1 expression (Fig. 5A). The TGF-β1 signaling pathway plays a key role in the regulation of ECM biosynthesis, including procollagen (31). As shown in Fig. 5B, UVB significantly decreased TGF-β1 expression, as well as the phosphorylation levels of Smad2 (Ser^465/467) and Smad3 (Ser^423/425), which were reduced by 0.2-fold, while the level of Smad7 increased by 1.3-fold compared to the control group. However, BJ-5ta Exo inhibited the UVB-induced suppression of the TGF-β1/Smad pathway. The UVB-induced production of MMPs can be mediated by protein kinase cascades, such as MAPK and AP-1 (32). BJ-5ta Exo prevented the UVB-mediated increase in the levels of p-p38 (Thr¹⁸⁰/Tyr¹⁸²), p-JNK (Thr¹⁸³/Tyr¹⁸⁵) and p-ERK(Thr²⁰²/Tyr²⁰⁴) (Fig. 5C). In addition, BJ-5ta Exo suppressed the UVB-induced phosphorylation of AP-1 subunits (c-Fos on Ser³² and c-Jun on Ser⁷³) (Fig. 5D).

To examine whether BJ-5ta Exo reduce wrinkle formation caused by UVB irradiation in SKH-1 hairless mice, BJ-5ta Exo were subcutaneously injected into the dorsal skin of each group of mice for 8 weeks following irradiation with UVB (Fig. S2A). Ultimately, there was no significant difference in the body weights of the mice treated with various concentrations of BJ-5ta Exo alongside UVB exposure, indicating the safety of BJ-5ta Exo for the mice (Fig. S2B). A visual assessment indicated that the administration of BJ-5ta Exo reduced the UVB-induced wrinkle formation on the back skins of the mice compared to the saline-treated UVB-irradiated mice (Fig. 6A). To quantitatively measure the roughness, relevant parameters in the images were analyzed using PRIMOS CR software version 5.8E, an optical three-dimensional skin measurement system. The roughness values in the UVB-saline group were significantly higher (1.4-fold) compared with those in the control group. By contrast, the mice treated with BJ-5ta Exo at concentrations of 1×10⁶ particles/ml and 1×10⁸ particles/ml exhibited a reduction in total wrinkle area of 0.3- and 0.4-fold, respectively, compared to the saline-treated group (Fig. 6E). BJ-5ta Exo also restored skin hydration and inhibited moisture evaporation compared to the UVB-saline group (Fig. 6F and G). As shown in Fig. 6H and I-K, the saline-treated UVB-irradiated mice had a thicker epidermal surface and a marked decrease in collagen (blue) and elastin (black) fibers in the dermis compared to the control mice, while the treated mice exhibited a significant increase in the abundance and density of collagen and elastin fibers in the dermis, and a significant decrease in the thickness of the epidermal layer. In addition, using IHC staining, it was observed that the expression levels of procollagen type I, collagen type I and elastin were restored by 1.4, 1.3, and 1.3-fold, respectively, in the BJ-5ta Exo 10⁶-treated group, and by 1.5, 1.3, and 1.3-fold, respectively, in the BJ-5ta Exo 10⁸-treated group, compared to the UVB-saline group. By contrast, MMP-1 expression decreased by 0.3-fold in the BJ-5ta Exo 10⁶-treated group and 0.4-fold in the BJ-5ta Exo 10⁸-treated group (Fig. 6L and M). These results suggest that BJ-5ta Exo can ameliorate skin photodamage by increasing ECM components.

To further investigate the effects of BJ-5ta Exo on UVB-induced photodamage in human skin, the Neoderm-ED reconstructed human skin model was employed. The thickness of the epidermis increased as a result of UVB exposure. However, BJ-5ta Exo effectively reduced the increased epidermal thickness caused by UVB (Fig. 7A and E). UVB radiation causes an abnormal reduction in procollagen type I and elastin, as well as the production of MMP-1. Using IHC staining for collagen type I, elastin and MMP-1, it was confirmed that BJ-5ta Exo inhibited the UVB-induced downregulation of collagen type I and elastin, while also suppressing the UVB-induced increase in MMP-1 (Fig. 7B-D). In addition, ELISA assay used to quantify procollagen type I and MMP-1. Compared to the UVB-alone group, the groups treated with BJ-5ta Exo at concentrations of 10⁶ and 10⁸ particles/ml exhibited an elevated expression of procollagen type I, with an increased rate of 34.53 and 35.30%, respectively. Additionally, they exhibited a decrease in MMP-1 levels by 6.85 and 20.14%, respectively. (Fig. 7F and G). These observations indicated that BJ-5ta Exo effectively restored the tissue damage induced by UVB in the layers of the reconstructed human skin model.

BJ-5ta Exo treatment contributes to the removal of oxidative stress by inhibiting ROS generation and preventing the decrease in the expression levels of SOD1 and SOD2, GPX and catalase. It also provides protection against UVB radiation by activating the DNA repair system through the regulation of both γH2AX and RAD51. BJ-5ta Exo prevent UVB-induced collagen degradation by activating the TGF-β1/Smads pathway and inhibiting the MAPK/AP-1 pathway. Furthermore, BJ-5ta Exo suppress UVB-induced cellular senescence by inhibiting the expression of SA-β-gal and p16. In a UVB-induced photoaging model, it was confirmed that BJ-5ta-Exo treatment decreased the level of TEWL, wrinkle formation and MMP-1 expression. On the other hand, BJ-5ta Exo increased the levels of collagen type-I and elastin in the dorsal skin (Fig. 8).