Pyropia yezoensis were cultured in Provasoli enriched seawater (Provasoli 1968) medium at 15°C under irradiation of 50 μmol photons m⁻² s⁻¹ provided by cool-white fluorescent lighting with a 12 h light/12 h dark cycle. Following established protocols for PyNVs isolation (You et al. 2021, Kim and Park 2022), the samples were mixed with phosphate-buffered saline (PBS; pH 8.0) in a 1:1 (w/w) ratio and homogenized using a blender (Model 7011G; Waring Laboratory Science, Torrington, CT, USA) at 22,000 rpm for 5 min at room temperature. The homogenate was centrifuged at 3,000 ×g for 30 min at room temperature to remove large debris, followed by centrifugation at 10,000 ×g for 1 h at room temperature. The supernatant was filtered through a 0.45 μm filter (Sartorius AG, Göttingen, Germany). Final ultracentrifugation was performed at 100,000 ×g for 2 h at room temperature using an Optima XE-90 ultracentrifuge (Beckman Coulter, Brea, CA, USA), according to established EV isolation protocols (Théry et al. 2018). The isolated PyNVs were resuspended in PBS (pH 8.0) and stored at −80°C until further use. The homogenized P. yezoensis extract using a blender was filtered through 0.22 μm and used as a control for the isolated PyNVs. The concentration of P. yezoensis extracts used in the experiments was conducted according to the highest concentration of PyNVs based on protein.

The size distribution and concentration of PyNVs were measured using a NanoSight NS300 (Malvern Panalytical, Worcestershire, UK) equipped with a 488 nm laser. Samples were diluted 1:1,000 in PBS. The instrument was configured with a camera level of 13 and a detection threshold of 4. For each sample, three 60-s videos were recorded at 25°C. Data were analyzed using NTA 3.4 software (Malvern Panalytical), and particle concentrations were corrected for the dilution factor. Concentrations are reported as mean ± standard error of the mean from three technical replicates. Total protein concentration of PyNVs was measured using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. The morphology and structural details of PyNVs were examined through cryo-transmission electron microscopy (Cryo-TEM) using a Glacios microscope (Thermo Fisher Scientific) operated at 200 kV.

Human keratinocyte cells (HaCaT) and human dermal fibroblasts (HDF) were maintained in high-glucose Dulbecco’s modified Eagle medium (DMEM; Gibco, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin-streptomycin-amphotericin mixture (Gibco) at 37°C in a humidified atmosphere containing 5% CO₂, as described by Werner et al. (2007). Human umbilical vein endothelial cells (HUVECs) were cultured in endothelial cell growth medium-2 (Lonza, Basel, Switzerland) under the same conditions, following established protocols (DiPietro 2016).

Cell viability was assessed using established protocols (Zhou et al. 2020). HaCaT cells (3 × 10⁵ cells well⁻¹), HDF cells (4 × 10⁴ cells well⁻¹), and HUVECs (6 × 10⁴ cells well⁻¹) were seeded in 24-well plates. After 24 h, cells were treated with PyNVs at concentrations ranging from 0.5 × 10⁹ to 2.0 × 10¹⁰ particles mL⁻¹ in serum-free medium and incubated for 24 h at 37°C. Cell viability was evaluated using the water-soluble tetrazolium salt-1 (WST-1) assay (EZ Cytox; DoGen Bio, Seoul, Korea). The absorbance was measured at 450 nm using a multi-mode microplate reader (PerkinElmer, Waltham, MA, USA) after 2 h of WST-1 reagent incubation, following methods described by Wang et al. (2019).

For proliferation assays, HaCaT cells (6 × 10⁴ cells well⁻¹) and HDF cells (1 × 10⁴ cells well⁻¹) were seeded in 24-well plates and treated with 2% exosome-depleted FBS and PyNVs (0.5–7.5 × 10⁹ particles mL⁻¹), as described by Zhang et al. (2015). Cell proliferation was measured at 12, 24, and 48 h using the WST-1 assay as detailed above.

Cell migration was evaluated using a wound healing assay following modified protocols from Cappiello et al. (2018). HaCaT cells (5 × 10⁴ cells well⁻¹) were seeded on silicone inserts (SPLScar Block; SPL Life Sciences, Pocheon, Korea) placed in 6-well plates and incubated at 37°C for 24 h in a 5% CO₂ incubator. After cell attachment, the silicone inserts were removed using sterile forceps to create a defined wound area (500 μm width). Cells were washed twice with PBS (pH 8.0) and cultured in fresh medium containing 2% exosome-depleted FBS and PyNVs (0.5–7.5 × 10⁹ particles mL⁻¹). Wound closure was monitored for 48 h using a JuLI Stage live cell imager (NanoEnTek, Seoul, Korea), with images captured every 30 min. The wound area was quantified using ImageJ software (NIH, Bethesda, MD, USA), and the closure rate was calculated as described by Tracy et al. (2016).

Total RNA from HDF cells was isolated using WelPrep Total RNA Isolation Reagent (Welgene, Gyeongsan, Korea) according to the manufacturer’s protocol. RNA purity and concentration were determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific), following methods described by Kim et al. (2017). First-strand cDNA was synthesized from 1 μg of total RNA using the QuantiTect Reverse Transcription Kit (Qiagen, Hilden, Germany). Gene-specific primers were designed to amplify fibronectin, elastin, COL1A1 (collagen type I alpha 1 chain), MMP-1 (matrix metalloproteinase-1), and β-actin. Primer pairs are as follows: fibronectin (5′-ACA ACA CCG AGG TGA CTG AGA C-3′, 5′-GGA CAC AAC GAT GCT TCC TGA G-3′), Elastin (5′-GGT TGT GTC ACC AGA AGC AGC T-3′, 5′-CCG TAA GTA GGA ATG CCT CCA AC-3′), COL1A1 (5′-GAT TCC CTG GAC CTA AAG GTG-3′, 5′-AGC CTC TCC ATC TTT GCC AGC A-3′), MMP-1 (5′-ATG AAG CAG CCC AGA TGT GGA G-3′, 5′-TGG TCC ACA TCT GCT CTT GGC A-3′), and β-actin (5′-CAC TGT GCC CAT CTA CG-3′, 5′-CTT AAT GTC ACG CAC GAT TTC-3′). Quantitative polymerase chain reaction (qPCR) was performed using WizPure qPCR Master Mix (Wizbiosolutions, Seongnam, Korea) on a Rotor-Gene Q 2plex RT-PCR system (Qiagen). The polymerase chain reaction (PCR) conditions were optimized based on previous studies (Lee et al. 2015): initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Melting curve analysis was performed to confirm specific amplification. Gene expression was normalized to β-actin using the 2-ΔΔCT method as described by Livak and Schmittgen (2001).

For enzyme-linked immunosorbent assay (ELISA) analysis, HDF cells were seeded at 1.5 × 10⁴ cells per well in 24-well plates and cultured for 24 h, following protocols adapted from Singh et al. (2023). Cells were then treated with PyNVs (0.5–7.5 × 10⁹ particles mL⁻¹) for 72 h. Culture supernatants were collected and centrifuged at 3,000 ×g for 10 min at 4°C. Collagen type I production was measured using the Procollagen Type I C-peptide (PIP) EIA Kit (Takara Bio, Kusatsu, Japan), and MMP-1 levels were determined using the Human Total MMP-1 DuoSet ELISA Kit (R&D Systems, Minneapolis, MN, USA), as described by Schuster et al. (2023). Standard curves were generated using serial dilutions (0–1,000 ng mL⁻¹) of the provided standards. The absorbance was measured at 450 nm using a microplate reader (PerkinElmer).

Angiogenic capacity was evaluated using a tube formation assay following methods established by DiPietro (2016) and Morbidelli et al. (2021). Growth factor-reduced Matrigel (Corning, Corning, NY, USA) was thawed overnight at 4°C and applied to 48-well plates (150 μL well⁻¹) at 4°C. The plates were incubated at 37°C for 1 h to allow polymerization. HUVECs (5 × 10⁴ cells well⁻¹) were seeded onto the Matrigel layer and treated with PyNVs (1.0–7.5 × 10⁹ particles mL⁻¹). After 16 h of incubation, tube formation was observed using an inverted microscope (Olympus, Tokyo, Japan), and five random fields per well were photographed. Tube formation parameters (total tube length, number of nodes, meshes, and branches) were quantified using the Angiogenesis Analyzer plugin for ImageJ software, as described by Akhtari et al. (2024).

To analyze the protein composition of nanovesicles (NVs) isolated from P. yezoensis, samples were collected at each step of the NV isolation process and subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). A total of 20 μg of each sample was mixed with 4× Laemmli sample buffer and heated at 95°C for 5 min. The samples were then loaded onto a 12% SDS-polyacrylamide gel and electrophoresed at 120 V for 90 min. After electrophoresis, the gel was stained with Coomassie brilliant blue R-250 and destained using a methanol:acetic acid:water (4:1:5, v/v/v) solution. A molecular weight marker (Precision plus protein; Bio-Rad, Hercules, CA, USA) was used to estimate the molecular weight of the protein bands. In-gel digestion with trypsin and liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis were analyzed using an LTQ Orbitrap XL mass spectrometer (Thermo Fisher Scientific). The resulting tandem mass spectra were searched using the Mascot search engine (version 2.7.0; Matrix Science, London, UK) for protein identification.

All experiments were performed in triplicate and repeated at least three times independently, following established statistical procedures for biological studies (Potekaev et al. 2021). Data are presented as mean ± standard deviation. Statistical analyses were performed using GraphPad Prism 6.0 software (GraphPad Software Inc., San Diego, CA, USA). Normal distribution was confirmed using the Shapiro-Wilk test as recommended by Tottoli et al. (2020). Differences between groups were analyzed using two-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. p-values were considered statistically significant at three levels: *p < 0.05, **p < 0.01, and ***p < 0.001.

P. yezoensis cultivated under optimized conditions was used to isolate NVs using differential ultracentrifugation method (Fig. 1A). Nanoparticle tracking analysis revealed that the isolated PyNVs exhibited a size distribution ranging from 95 to 195 nm, with an average diameter of 140.1 nm (Fig. 1B). The mean particle concentration was 2.77 × 10¹¹ ± 4.71 × 10⁹ particles mL⁻¹. Cryo-TEM analysis confirmed the presence of spherical vesicles with intact lipid bilayer structure, approximately 110 nm in diameter (Fig. 1C). The production yield of PyNVs, determined by dividing the number of isolated PyNVs by the weight of P. yezoensis used, was 2.29 × 10¹¹ particles g⁻¹ of P. yezoensis (Fig. 1D). The purity of isolated PyNVs was validated by a particle-to-protein ratio of 1.43 × 10⁸ particles μg⁻¹, which falls within the acceptable range for high-quality EV preparations (Fig. 1E). The protein composition of PyNVs was analyzed through SDS-PAGE and proteomic profiling. SDS-PAGE was performed on protein fractions collected during the NV isolation procedure. As shown in Supplementary Fig. S1A, the purified PyNVs fraction (lane 4) exhibited multiple distinct protein bands ranging approximately from 50 to 250 kDa, indicating the successful enrichment of protein-containing NVs. To further characterize the PyNVs protein cargo, LC-MS/MS analysis was conducted. A total of 232 proteins were identified and annotated (Supplementary Table S1). These proteins were classified into six major categories based on their functional roles and known associations with EVs, as summarized in Supplementary Fig. S1B. A total of 232 proteins were categorized into six groups: enzymes (76 proteins), cytosolic proteins (4 proteins), membrane proteins (8 proteins), heat shock proteins (5 proteins), cytoskeletal proteins (5 proteins), and others (134 proteins).

To assess the potential cytotoxicity of PyNVs, three cell types critical to wound healing—HaCaT, HDF, and HUVECs—were exposed to increasing concentrations (1 × 10⁸ to 2 × 10¹⁰ particles mL⁻¹) of PyNVs for 24 h. WST-1 assay results demonstrated that PyNVs and P. yezoensis extracts (140 μg mL⁻¹) exhibited no significant cytotoxicity across the tested concentration range in all three cell types (Fig. 2), supporting their safety for therapeutic applications.

Angiogenesis represents a critical early phase of wound healing, providing essential oxygen and nutrients while promoting cell survival. Using an in vitro tube formation assay with HUVECs, we evaluated the angiogenic potential of PyNVs. Angiogenic activity was enhanced in a dose-dependent manner in the PyNV-treated group, and treatment with 7.5 × 10⁹ particles mL⁻¹ of PyNV significantly enhanced angiogenic activity (Fig. 3A). Quantitative analysis revealed significant increases in total tube length (Fig. 3B), number of nodes (Fig. 3C), mesh formation (Fig. 3D), and branch points (Fig. 3E) compared to untreated controls. However, P. yezoensis extracts (50 μg mL⁻¹) did not affect angiogenic activity even when used at high concentrations.

To investigate the regenerative potential of PyNVs, we examined their effects on cell proliferation and migration. Both HaCaT and HDF cells treated with PyNVs (5 × 10⁸ to 7.5 × 10⁹ particles mL⁻¹) showed dose-dependent increases in proliferation. The most pronounced effect was observed at 7.5 × 10⁹ particles mL⁻¹, with significant enhancement in proliferation of both cells at 48 h (Fig. 4A & B).

In the wound healing assay, HaCaT cells treated with higher concentrations of PyNVs (1 × 10⁹ and 7.5 × 10⁹ particles mL⁻¹) demonstrated markedly enhanced migration capabilities. The wound closure rates reached 96.9 and 97.6%, respectively at these concentrations within 12 h, significantly surpassing the control group’s 76.9% closure rate (Fig. 4C). In contrast, high concentration of P. yezoensis extract (50 μg mL⁻¹) did not affect skin cell proliferation or migration compared to the control. These findings suggest that PyNVs effectively promote both proliferation and directional migration of skin cells.

To elucidate the molecular mechanisms underlying PyNVs-mediated wound healing, we analyzed the expression of key ECM-related genes. Real-time PCR analysis revealed that high concentrations of PyNVs (1 × 10⁹ and 7.5 × 10⁹ particles mL⁻¹) significantly upregulated multiple ECM genes. At 7.5 × 10⁹ particles mL⁻¹, we observed remarkable increases in expression: fibronectin (3.5-fold) (Fig. 5A), elastin (8.5-fold) (Fig. 5B), COL1A1 (10-fold) (Fig. 5C), and MMP-1 (2.6-fold) (Fig. 5D).

To validate these findings at the protein level, we examined the secretion of collagen and MMP-1 in HDF cells treated with various concentrations of PyNVs. ELISA analysis demonstrated a dose-dependent increase in collagen secretion (Fig. 5E), while MMP-1 levels remained stable (Fig. 5F). Importantly, the collagen/MMP-1 ratio increased approximately 1.6-fold in the high-concentration group compared to controls (Fig. 5G), indicating that PyNVs promote ECM production while maintaining balanced matrix remodeling.