Osteogenic Effects of Brown Algae Sargassum fulvellum-Derived Exosome-Like Nanovesicles via the BMP-2/4-MAPK-Smad1/5/8 Signaling Pathway

Abstract

Osteoporosis is a prevalent metabolic bone disorder that is characterized by reduced bone mass and microarchitectural deterioration, especially in women who are postmenopausal. Brown algae such as Sargassum fulvellum contain bioactive compounds with potential bone health benefits. This study examined the effects of S. fulvellum-derived exosome-like nanovesicles (SfNVs) on osteoblast differentiation and mineralization. SfNVs were isolated via differential ultracentrifugation, characterized by nanoparticle tracking analysis, bicinchoninic acid protein assay, and NanoImaging, and subsequently applied to MC3T3-E1 osteoblasts. The SfNVs (∼208 nm) were efficiently internalized without exhibiting cytotoxicity and significantly enhanced alkaline phosphatase activity, mineralized nodule formation, and the expression of osteogenic markers, such as Runx2, ALP, OPN, and ProCOL I. In addition, the SfNVs upregulated the expression of BMP-2/4 and activated the MAPK (p38, JNK, and ERK) and Smad1/5/8 signaling pathways. Collectively, these findings indicate that SfNVs promote osteoblast differentiation and mineralization via the BMP-2/4-MAPK-Smad1/5/8 axis, thereby suggesting their potential as functional marine-derived biomaterials for the nutritional prevention of bone metabolic diseases.

INTRODUCTION

Osteoporosis is a representative metabolic bone disease that is characterized by decreased bone mass and the deterioration of bone microarchitecture, which leads to an increased risk of developing fractures (Rachner et al., 2011). In particular, postmenopausal women frequently experience fractures in regions with a high proportion of trabecular bone, such as the spine, pelvis, and wrist, due to rapid bone loss caused by decreased estrogen levels (Raggatt and Partridge, 2010; Seo et al., 2021). More than 200 million people worldwide are affected by osteoporosis, which is often referred to as a “silent disease” because of the lack of noticeable symptoms (Lin and Lane, 2004). With the continued increase in the older population, the prevalence and socioeconomic burden of osteoporosis are also expected to rise, thus making the early diagnosis and prevention thereof an urgent priority (Chun, 2019).

Physiological bone remodeling is a biological process that is precisely regulated, which maintains mineral homeostasis and repairs damaged bone via the balanced interaction between osteoblasts and osteoclasts (Shin et al., 2008; Raggatt and Partridge, 2010). The disruption of this balance can result in pathological bone diseases such as osteoporosis or osteosclerosis. Recent studies have revealed that, in addition to traditional bone cells, various immune cells are involved in bone remodeling, thereby highlighting new potential targets for bone metabolism regulation (Kanis et al., 2008).

Seaweed has traditionally been used in East Asian countries, such as China, Korea, and Japan, for medicinal and nutritional purposes and has recently garnered attention as a source of functional metabolites (Kim et al., 2007; Liu et al., 2020). Sargassum fulvellum is rich in dietary fiber, particularly soluble polysaccharides, as well as essential minerals such as calcium, magnesium, and iodine. In addition, despite its low caloric content, it provides beneficial fatty acids, including omega-3s, along with a moderate amount of protein. These beneficial attributes highlight its potential as a valuable food fortification ingredient and thus as a functional food resource. In particular, brown algae are rich in bioactive polysaccharides such as alginate, fucoidan, and laminarin, which have previously been reported to exhibit antioxidant, anti-inflammatory, and immunomodulatory activities (Choi et al., 2020). Extracts from brown algae, including S. fulvellum, have been shown to promote osteoblast differentiation and inhibit osteoclast activity, thereby exerting positive effects on bone formation and density (Yamaguchi, 2013). Furthermore, some studies have suggested that these effects are mediated by the inhibition of the NF-κB pathway. Moreover, S. fulvellum-derived compounds have demonstrated anti-edema effects without inducing toxicity (Kang et al., 2008). Given that osteoporosis is a multifactorial disease that involves imbalanced bone remodeling, which is characterized by excessive osteoclast activity and insufficient osteoblast function, the ability of S. fulvellum components to modulate these processes presents a promising therapeutic avenue (Sim et al., 2024).

Recently, there has been growing interest in the functional role of extracellular vesicles with regard to intercellular communication and physiological regulation. Exosomes, as nano-sized membrane vesicles that contain various biomolecules such as RNA, proteins, and lipids, have emerged as promising carriers for bioactive substances (Peinado et al., 2012; Zempleni et al., 2019). In particular, exosomes and exosome-like nanovesicles (NVs) have attracted attention as potential platforms for drug delivery because of their high biocompatibility and targeting specificity (Antimisiaris et al., 2018; Gomari et al., 2018). Moreover, plant- and algae-derived exosomes have been indicated as biocompatible alternatives for functional materials (Jang et al., 2021; Lee et al., 2024).

Therefore, this study aimed to examine the osteogenic potential of exosome-like NVs derived from the brown alga S. fulvellum-derived exosome-like nanovesicles (SfNVs) and to clarify the molecular mechanisms involved, with a particular emphasis on the BMP-2/4-MAPK-Smad1/5/8 signaling pathway axis. Given the emerging interest in plant- and algae-derived NVs as biocompatible, sustainable bioactive molecule carriers, elucidating the role of SfNVs in bone metabolism may contribute to the development of novel marine algae-based interventions for osteoporosis prevention and the promotion of bone health.

MATERIALS AND METHODS

Materials

All reagents were purchased from Sigma-Aldrich. The α-minimum essential medium (α-MEM), penicillin-streptomycin, and fetal bovine serum (FBS), used for cell culture, were obtained from Gibco Laboratories. Other reagents and kits not mentioned here were the same as those previously described (Hwang et al., 2023; Park et al., 2023; Kim et al., 2024b).

Isolation of SfNVs derived exosome-like NVs

SfNVs were isolated from 100 g of S. fulvellum that was collected from Wando, Jeollanam-do. After washing with running water, the seaweed extract prepared in PBS was subjected to sequential centrifugations using the previously reported protocols (Eom et al., 2022; Choi et al., 2023; Hwang et al., 2023; Kim et al., 2023; Park et al., 2023; Sim et al., 2023; Kim et al., 2024a; Kim et al., 2025a; Kim et al., 2025b). To remove large particles, centrifugation was performed at 500 g for 10 min at 4°C, followed by centrifugation at 2,000 g for 20 min and then twice at 10,000 g for 30 min to eliminate the cells and debris. Ultracentrifugation was performed at 100,000 g for 60 min to collect the SfNVs. The pellet was resuspended in PBS, and the protein concentration was measured using the Pierce bicinchoninic acid (BCA, Thermo Fisher Scientific) protein assay, and the concentration was adjusted before the vesicles were used for subsequent experiments. Particle size and concentration were measured using nanoparticle tracking analysis (NTA).

Cell culture

MC3T3-E1 osteoblast cells were cultured in α-MEM that was supplemented with 10% FBS and 10,000 U/mL penicillin-streptomycin. Osteogenic differentiation was induced by adding 50 µg/mL of ascorbic acid and 10 mM β-glycerophosphate (or sodium phosphate monobasic) to the growth medium. The cultured cells were treated with SfNVs at concentrations of 0, 1, 5, and 10 µg/mL, and the medium was replaced every 3 days during incubation for up to 7 days (Hwang et al., 2023; Park et al., 2023; Sim et al., 2023).

NTA

The particle number and size of the SfNVs were measured using NTA with the NanoSight NS300 system (NanoSight, Malvern Panalytical Ltd). The instrument was calibrated using 100 nm polystyrene beads prior to analysis (Thermo Fisher Scientific). The nanoparticle concentration (particles/mL) was determined using NTA software version 3.4 (NanoSight, Malvern Panalytical Ltd), and the built-in batch process function was employed to integrate three technical replicates for each sample.

Stochastic optical reconstruction microscopy

To visualize the exosome marker proteins CD81, CD63, and CD9 present on the SfNVs lipid membrane, the EV Profiler Kit (#EV-MAN-1.0, ONI) was used in combination with direct stochastic optical reconstruction microscopy (dSTORM) for immunolabeling and imaging. The experiment was performed according to the protocol provided by the manufacturer. Prior to initiating the imaging session, 0.1 µm of TetraSpeck beads (#T7279, Thermo Fisher Scientific) were used to calibrate the channel mapping. The data were processed using NimOS software (version 1.19, ONI).

Proteomic analysis of SfNVs

The purified SfNVs were commissioned to Protia Inc. for proteomic analysis. During the analysis, the SfNV proteins were extracted and enzymatically digested, followed by LC-MS/MS analysis for protein identification. Mascot software was used to identify the protein libraries, and all identified proteins were subjected to Gene Ontology analysis using DAVID software.

DiD fluorescent labeling of SfNVs

The SfNVs (50 µg/mL in PBS) were incubated with 5 mM DiD dye at 25°C for 30 min, and then the MC3T3-E1 cells were treated with the DiD-labeled SfNVs for 6, 12, and 24 h. Their cellular uptake was determined by using a fluorescence microscope (iRiS^TM, Logos Biosystems).

MTT assay

The MC3T3-E1 cells were seeded at a density of 1×10⁵ cells/well in 48-well plates and cultured with or without SfNVs for 3 or 7 days. Thereafter, 5 mg/mL of MTT reagent was added to each well, followed by dimethyl sulfoxide to dissolve the formazan crystals. Absorbance was then measured at 570 nm using a microplate reader (TECAN) to assess cell viability and proliferation.

Alizarin Red S, Von Kossa, and Van Gieson collagen staining

The MC3T3-E1 cells were seeded at a density of 1×10⁵ cells/well in 24-well plates and treated with SfNVs for 3 or 7 days. After washing with PBS and fixing with 70% ethanol, the cells were stained with Alizarin Red S (pH 4.2) for 30 min and imaged under a light microscope. Calcium deposits were quantified by dissolving them with cetylpyridinium chloride and measuring the absorbance at 570 nm. For Von Kossa staining, similarly cultured cells were fixed and stained with 3% silver nitrate under ultraviolet light at room temperature for 30 min. The mineralized matrix was observed under a light microscope. For collagen staining, the cells were fixed with 70% ethanol, washed with PBS, and stained with Van Gieson solution (1% aqueous acid fuchsin and saturated aqueous picric acid) for 15 min at room temperature. The stained collagen fibers were imaged using a light microscope.

Determination of alkaline phosphatase (ALP) activity

The MC3T3-E1 cells were seeded at a density of 1×10⁵ cells/well in 24-well plates and treated with SfNVs for 3 or 7 days. Cells were fixed with 70% ethanol and stained using NBT/BCIP. The stained images were captured with a light microscope. ALP activity was measured using previously reported protocols (Hwang et al., 2023). Enzyme activity was expressed as U/mg protein for cellular measurements and as U/mL for the media measurements.

Quantitative real-time polymerase chain reaction

Total RNA was extracted from the MC3T3-E1 cells that were treated with or without SfNVs using the RNeasy Mini Kit (Qiagen). For cDNA synthesis, the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems) was employed and performed on a MiniAmp Thermal Cycler (Life Technologies Holding Pte Ltd.). Quantification of mRNA expression was performed using the SYBR Master Mix Kit (Applied Biosystems) and the QuantStudio^TM 1 Real-Time PCR system (Life Technologies Holding Pte Ltd.). GAPDH was used as the endogenous control gene. The sequences of primers used for quantitative real-time polymerase chain reaction (qRT-PCR) are provided in Table 1.

Table 1.

Primer sequences used for quantitative real-time polymerase chain reaction (qRT-PCR)

Gene	Sequence (5’-3’)
Runx2	Forward	CGCACGCGATGCAACACCAC
	Revers	ACTGCATGTCCCCGGGCTCA
Osteopontin	Forward	CTGGCAGCTCAGAGGAGAAG
	Revers	CAGCATTCTGTGGCGCAAG
Alkaline phosphatase	Forward	CAAGGATGCTGGGAAGTCCG
	Revers	CGGATAACGAGATGCCACCA
ProCOL 1	Forward	ACGTCCTGGTGAAGTTGGTC
	Revers	CAGGGAAGCCTCTTTCTCCT
GAPDH	Forward	TGACGTGCCGCCTGGAGAAA
	Revers	AGTGTAGCCCAAGATGCCCTTCAG
p38	Forward	AAGCCGACAGAGGTGGCAAGTT
	Revers	GGTCAAGGTGATGGTGGCAAAG
JNK1	Forward	CGCCTTATGTGGTGACTCGCTA
	Revers	TCCTGGAAAGAGGATTTTGTGGC
JNK2	Forward	GTCAGTGGGTTGCATCATGGGA
	Revers	ACTCTGCGGATGGTGTTCCTAG
ERK1	Forward	GGCTTTCTGACGGAGTATGTGG
	Revers	GTTGGAGAGCATCTCAGCCAGA
ERK2	Forward	TCAAGCCTTCCAACCTCCTGCT
	Revers	AGCTCTGTACCAACGTGTGGCT
Smad1	Forward	AAGGTGGGGAAAGTGAAAC
	Revers	CTGCTTGGAACCAAATGGGAA
Smad5	Forward	GGAACCTGAGCCACAATGAA
	Revers	CTTGCTGGGGAGTTGGGATA
Smad8	Forward	CACCGACCCTTCCAATAAC
	Revers	CTGGACAAAGATGCTGCTG
BMP-2/4	Forward	CGCACGCGATGCAACACCAC
	Revers	ACTGCATGTCCCCGGGCTCA

Western blot analysis

The SfNVs protein concentration was measured using the BCA protein assay kit, and equal amounts of protein were separated via 10% SDS-PAGE, followed by transfer onto a nitrocellulose membrane. After blocking with 3% bovine serum albumin, the membrane was incubated overnight at 4°C with the respective primary antibodies, followed by incubation with the appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies. Signal detection was performed using an enhanced chemiluminescent kit from Thermo Fisher Scientific. The Runx2, ALP, osteopontin (OPN), and ProCOL I levels were quantified relative to that of β-actin, while the phosphorylated proteins were quantified relative to their corresponding total protein levels. The primary antibodies used were against Runx2, ALP, OPN, ProCOL I, β-actin, BMP-2/4, phospho-p38, phospho-ERK, phospho-JNK, Smad1, p38, ERK, and JNK (Santa Cruz Biotechnology Inc., diluted 1:1,000), and phospho-Smad1 (Cell Signaling Technology, Rabbit mAb, diluted 1:3,000). The secondary antibodies used were m-IgGk BP-HRP and mouse anti-rabbit IgG-HRP (Santa Cruz Biotechnology Inc., diluted 1:3,000).

Immunofluorescence staining

The MC3T3-E1 cells were seeded at a density of 1×10⁴ cells/well in eight-chamber slides and treated with SfNVs for 24 h. Cells were fixed with 2.5% paraformaldehyde at room temperature for 30 min, permeabilized with 0.3% Triton X-100 in PBS for 15 min, and incubated overnight at 4°C with BMP-2/4 antibodies (Santa Cruz Biotechnology Inc.). After washing three times, images were captured by secondary antibody staining using goat antimouse IgG (Thermo Fisher Scientific, diluted 1:300), followed by nuclear staining with DAPI (Sigma-Aldrich).

Statistical analysis

All results are expressed as the mean±standard error. Statistical significance was evaluated using both the independent t-test and one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test, depending on the experimental design. All analyses were performed using SPSS 26.0 (IBM Corp.). P<0.05, P<0.01, and P<0.001 were considered statistically significant.

RESULTS

Isolation and characterization of SfNVs

SfNVs were stably isolated from the S. fulvellum extract using a differential ultracentrifugation method, which yielded a total of 15 vials (Fig. 1A). This isolation method is widely employed for purifying nanoparticles from various biological sources because of its cost-effectiveness and efficiency (Dad et al., 2021).

Fig. 1.

Isolation and characterization of Sargassum fulvellum-derived exosome-like nanovesicles (SfNVs). (A) SfNVs isolation process using ultracentrifugation. (B) Particle size distribution measured using nanoparticle tracking analysis (NTA) and a representative image of SfNVs. (C) Total protein content of SfNVs was measured with the bicinchoninic acid protein assay. (D) Each value represents the mean of five vials (n=3). (E) Distribution of the exosome marker proteins CD81 (purple), CD63 (yellow), and CD9 (blue) were visualized by direct stochastic optical reconstruction microscopy.

The size distribution of the isolated SfNVs was analyzed using NTA, which revealed an average particle diameter of approximately 208 nm. Most of the particles were distributed within the 100－250 nm range, and the total particle number determined by NTA was 1.08×10¹² (Fig. 1B). The particle number per 100 g of S. fulvellum was calculated as 1.61×10¹³, thereby indicating that the isolated SfNVs have a size comparable to those of exosomes and other cell-derived nanoparticles, which suggests their potential as carriers for cellular-level delivery.

Furthermore, to confirm the presence of proteins in the SfNVs, the BCA protein assay was used to measure the total protein concentration. The result revealed that the SfNVs contained approximately 3.67 mg/mL of protein, and the protein amount per 100 g of S. fulvellum was 18.5 mg/mL (Fig. 1C and 1D).

This result suggests the potential of SfNVs to possess functional biological activity. To further confirm this notion, quantitative proteomic analysis was performed to characterize the SfNVs protein composition, and the identified proteins are listed in Table 2. The proteins exhibited a molecular mass ranging from approximately 23－73 kDa and protein scores between 56 and 228. The major components included ribulose-1,5-bisphosphate carboxylase/oxygenase (large subunit), ATP synthase β-subunit, and several hypothetical proteins. The presence of multiple ATP synthase-related proteins indicates a potential role of SfNVs in modulating cellular energy metabolism, while the detection of ribulose-1,5-bisphosphate carboxylase/oxygenase suggests the incorporation of photosynthesis-associated components derived from the algal source. In addition, histone-related proteins were identified, thus implying that SfNVs may encapsulate nuclear-associated molecules, potentially contributing to regulatory or signaling functions.

Table 2.

List of proteins contained in SfNVs identified by proteomics analysis

Accession No.	Protein name	Score	Mass (Da)
ABU53651.1	Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit	228	54,437
KAF1856398.1	Hypothetical protein Lal_00048594	114	55,200
YP_009238059.1	CF1 beta subunit of ATP synthase	111	52,109
YP_009238261.1	CF1 beta subunit of ATP synthase	111	52,031
AAY19331.1	ATP synthase beta subunit, partial	97	51,143
RYR73949.1	Hypothetical protein Ahy_A02g008531	92	73,273
AQZ54860.1	ATP synthase beta chain, partial	87	44,363
WP_046799526.1	ATP synthase subunit beta	82	51,353
AGJ76098.1	Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit, partial	73	23,120
ABC47716.1	Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit, partial	70	49,012
PNH02976.1	Histone H2B.3	70	28,281
EOY08110.1	Histone deacetylase 2C, putative	56	32,924
XP_027070403.1	Uncharacterized protein LOC113695482	53	49,002
PUZ57787.1	Hypothetical protein GQ55_5G457100	52	58,892
XP_003607688.2	Pentatricopeptide repeat-containing protein At4g04790, mitochondrial	52	101,365
QGN73916.1	ATP synthase F1 subunit alpha	51	55,219
XP_016567275.1	Uncharacterized protein LOC107865541	48	41,419
CAA58635.1	Ribulose-1,5-bisphosphate carboxylase, partial	47	52,235
TVU28923.1	Hypothetical protein EJB05_20461, partial	46	85,717
PHT61467.1	Hypothetical protein T459_34681	46	180,304
XP_013891449.1	Hypothetical protein MNEG_15534	46	46,687
KJB30982.1	Hypothetical protein B456_005G170700	44	9,979
BAT01530.1	Os07g0487300	42	70,248
ONK74688.1	Uncharacterized protein A4U43_C03F9120	42	91,437
XP_024155957.1	Protein PLASTID TRANSCRIPTIONALLY ACTIVE 14-like	42	34,720
XP_033148238.1	Uncharacterized protein LOC117134242	42	52,767
AAO06283.1	Rbulose-1,5-bisphosphate carboxylase/oxygenase large subunit, partial	42	17,613
XP_023878992.1	Heat shock protein 90	41	80,084
TVU22195.1	Hypothetical protein EJB05_31877	41	59,098
XP_010228828.1	Uncharacterized protein LOC104581813	40	14,618
KRH29952.1	Hypothetical protein GLYMA_11G148600	40	14,008
XP_007149087.1	Hypothetical protein PHAVU_005G040100g, partial	40	21,941
SBO07518.1	Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit, partial	40	51,784
AAA19997.1	Ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit, partial	40	46,351
PLY92104.1	Hypothetical protein LSAT_1X85381	39	83,196
GFH21733.1	Histone H4	39	10,049
KAF6252479.1	Hypothetical protein COO60DRAFT_1704250	39	43,259
CAA7025778.1	Unnamed protein product	39	26,495
PRW60536.1	Mitochondrial metalloendopeptidase OMA1	39	104,721
XP_014515066.1	Cyclin-A2-4	39	54,511
KAF6171920.1	Hypothetical protein GIB67_011817	39	54,859
AMW03431.1	Branchless trichome, partial	39	19,016
GAU22916.1	Hypothetical protein TSUD_326890	39	74,750
TXG59984.1	Hypothetical protein EZV62_014557	39	125,929
EFH61576.1	NLI interacting factor family protein	39	68,719
KAF0918266.1	Hypothetical protein E2562_023348	39	26,131
KAF6142979.1	Hypothetical protein GIB67_014546	39	61,687
XP_021731985.1	UDP-glycosyltransferase 82A1-like	39	51,863

The spatial distribution of the exosome marker proteins CD81, CD63, and CD9 present on the isolated SfNVs lipid membrane was visualized using dSTORM. This technique enabled the analysis of the spatial localization of single molecules within the NV compartments. As shown in Fig. 1E, signals of CD81, CD63, and CD9 that were conjugated with the appropriate fluorescent dyes were within the size range (∼208 nm) observed in NTA, which confirmed the localization of these markers on the SfNVs lipid membrane.

By employing NTA, protein quantification, proteomics, and dSTORM, it was confirmed that these vesicles meet the internationally accepted guidelines for exosome characterization (Théry et al., 2018; Abhange et al., 2025). Furthermore, these findings confirm that the isolated vesicles exhibit previously defined exosome-like structural properties. Moreover, plant-derived NVs have been reported to exert various biological functions, including anti-inflammatory (Karabay et al., 2025), antioxidant (Majewska et al., 2025), anticancer (Karabay et al., 2025), drug delivery (Dutta et al., 2025), and anti-osteoporotic (Hwang et al., 2023) effects. In line with these reports, our study further demonstrates that the SfNVs employed herein possess similar therapeutic potential.

Cellular uptake of SfNVs

To evaluate the cellular uptake of SfNVs, the particles were labeled with the DiD fluorescent dye and applied to the MC3T3-E1 cells (Fig. 2A). After incubation for 6, 12, and 24 h, the cells were observed under a fluorescence microscope. Red fluorescence signals were detected within the cells at all time points, with signal intensity increasing over time, which indicates the time-dependent uptake and accumulation of SfNVs within the cells (Fig. 2B).

Fig. 2.

Cellular uptake of Sargassum fulvellum-derived exosome-like nanovesicles (SfNVs) by osteoblasts. (A) Schematic diagram illustrating the experimental workflow for analysis of SfNV uptake in MC3T3-E1 cells. (B) Representative fluorescence microscopy images showing the intracellular uptake of DiD-labeled SfNVs by MC3T3-E1 cells.

Effects of SfNVs on MC3T3-E1 cell proliferation, osteogenic differentiation, and mineralization

ALP activity increases during the early stages of osteogenic differentiation, thus making it a valuable marker for detecting early cellular responses (Lee et al., 1997). However, elevated ALP activity at this stage does not necessarily indicate successful mineralization. Therefore, we performed Alizarin Red S at the same time points, and many studies have reported assessing mineralization at around day 7 (Zheng et al., 2020).

To evaluate the SfNVs biological activities, we comprehensively assessed their effects on cell proliferation, osteogenic differentiation, and mineralization in MC3T3-E1 cells. Overall, SfNVs exhibited positive effects across all stages of osteoblast development.

MTT assays (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) were performed following treatment with various concentrations (1, 5, and 10 µg/mL) of SfNVs. On day 3 and day 7, cell viability remained comparable to that of the control group, thereby indicating that SfNVs did not induce cytotoxicity (Fig. 3A).

Fig. 3.

Effects of Sargassum fulvellum-derived exosome-like nanovesicles (SfNVs) on cell proliferation, osteogenic differentiation, and matrix mineralization in MC3T3-E1 cells. (A) Cell proliferation was assessed using the MTT assay. (B) Intracellular alkaline phosphatase (ALP) activity of MC3T3-E1 cells treated with SfNVs. (C) Secreted ALP activity was measured in the culture supernatant. (D) Representative images of ALP staining in MC3T3-E1 cells. (E) Collagen matrix formation was visualized by Van Gieson staining. (F) Mineralized nodules were visualized by Von Kossa staining. (G) Representative images of calcium deposition by Alizarin Red S staining. (H) Quantitative analysis of calcium deposition based on the Alizarin Red S staining results. Experimental results are expressed as the mean±standard error, and significance was evaluated using Duncan’s multiple range test. Statistically significant differences between the SfNV-treated and control (CON) groups at P<0.05 are indicated by different letters (a-c). Scale bar=500 µm.

To assess the early osteogenic differentiation of osteoblasts, ALP staining and the quantitative analysis of enzymatic activity were performed. The quantitative analysis revealed a trend of increased ALP activity on day 7, which supports the role of SfNVs in promoting early-stage differentiation (Fig. 3B). Meanwhile, ALP activity in the culture medium was significantly elevated in the 10 µg/mL SfNVs group on day 3, but no significant difference was observed when compared to that of the control on day 7 (Fig. 3C). ALP staining results further demonstrated a significant increase in the number of stained nodules on day 7 compared to that of the control (Fig. 3D). Collectively, these findings suggest that SfNVs transiently enhanced ALP activity during the early stages of osteogenic differentiation and may be involved in the regulation of ALP secretion over time.

In addition, Van Gieson staining revealed no noticeable changes in the collagen fiber structure following SfNVs treatment (Fig. 3E). To assess the mineralization-related markers, Von Kossa and Alizarin Red S staining were performed. The Alizarin Red S staining results revealed a marked increase in mineralized nodule formation in the SfNV-treated groups. Notably, significant mineralization was further supported by absorbance-based quantitative analysis, particularly on day 3 for the 5 and 10 µg/mL SfNVs groups and on day 7 at all treatment concentrations (Fig. 3F－3H).

Taken together, these results demonstrate that SfNVs promote cell proliferation, enhance osteoblast differentiation, and accelerate mineralization, thereby suggesting their potential as bioactive marine-derived materials for supporting bone formation.

Gene and protein expression of osteogenic markers

ALP, OPN, and Runx2 are key factors in osteoblast differentiation, where ProCOL I contributes to the bone tensile strength, and minerals contribute to the compressive strength (Shi et al., 1996; Komori, 2019; Si et al., 2020; Whyte, 2020). qRT-PCR and western blot analysis were used to evaluate the osteogenic marker expression. The mRNA expressions of Runx2, ALP, OPN, and ProCOL I were significantly upregulated at day 3 and day 7, with a stronger induction on day 7 (Fig. 4A and 4B). The amounts of the expressed proteins exhibited similar trends (Fig. 4C and 4D), which validates the mRNA results and confirms SfNVs role in promoting osteoblast differentiation.

Fig. 4.

Sargassum fulvellum-derived exosome-like nanovesicles (SfNVs) promote the expression of osteogenic markers at the mRNA and protein levels. (A) mRNA expression levels of osteogenic markers (Runx2, ALP, OPN, and ProCOL I) were measured using quantitative real-time polymerase chain reaction (qRT-PCR) on day 3 after SfNV treatment. (B) qRT-PCR analysis was performed for the same markers on day 7. (C) Osteogenic marker protein expression of Runx2, ALP, OPN, and ProCOL I in MC3T3-E1 cells on day 3 was assessed via western blot analysis. (D) Western blot analysis of osteogenic marker protein expression on day 7. Experimental results are expressed as the mean±standard error, and significance was evaluated using Duncan’s multiple range test. Statistically significant differences between the SfNV-treated and control (CON) groups at P<0.05 are indicated by different letters (a-c).

Promotion of osteogenesis is mediated through the dual-axis activation of the BMP-2/4-mediated Smad1/5/8 and the MAPK signaling pathways

BMP-2/4 is an upstream regulator of osteoblast differentiation via the phosphorylation of MAPKs (i.e., p38, JNK, and ERK), which then activates Smad1/5/8 phosphorylation to promote Runx2-dependent osteogenesis (Ogasawara et al., 2004; Kim et al., 2015). To determine whether SfNVs (10 µg/mL) modulate BMP-2/4 signaling, qRT-PCR and western blot analysis were performed. qRT-PCR analysis revealed that the expression of BMP-2/4, Smad1/5/8, p38, and ERK genes was significantly increased on day 7 of SfNVs treatment compared to that of the control group (Fig. 5A). Furthermore, western blot analysis showed increased BMP-2/4 protein. On day 7, the phosphorylation of all proteins, including ERK, was significantly enhanced (Fig. 5B). Immunofluorescence analysis revealed stronger intracellular BMP-2/4 fluorescence signals in the SfNVs-treated cells compared to that of the control group, thereby indicating enhanced expression and signal transduction (Fig. 5C). These results suggest that SfNVs promote osteoblast differentiation via the BMP-2/4-mediated activation of the MAPK/Smad signaling pathway, leading to Runx2-driven osteogenesis.

Fig. 5.

Sargassum fulvellum-derived exosome-like nanovesicles (SfNVs) upregulate osteoblast differentiation by activating the BMP-2/4 and MAPK signaling pathways. (A) mRNA expression of BMP-2/4, Smad1/5/8, and MAPK pathway components (ERK, JNK, and p38) were analyzed by quantitative real-time polymerase chain reaction after 7 days of SfNV treatment. (B) Protein expression patterns were evaluated after 7 days of SfNV treatment. (C) Regulatory mechanisms of osteoblast differentiation via the BMP-2/4 signaling pathway were visualized using immunofluorescence staining. Experimental results are presented as the mean±standard error, and statistical significance was analyzed using the independent sample t-test. *P<0.05, **P<0.01, and ***P<0.001 indicate statistically significant differences between the SfNV-treated and control (CON) groups.

DISCUSSION

Exosomes are nano-sized vesicles that contain various proteins, nucleic acids, bioactive lipids, and secondary metabolites, which play pivotal roles in intercellular communication (Koniusz et al., 2016; O’Brien et al., 2020). Plant-derived exosomes or exosome-like NVs share similar molecular compositions and have garnered increasing attention as bioactive carriers because of their biocompatibility and biodegradability (Garaeva et al., 2021; Eom et al., 2022). Recent studies have shown that plant-derived NVs exert physiological regulatory functions in osteoporosis, inflammation, and oxidative stress-related diseases, which positions them as promising natural therapeutic agents. For instance, Pueraria lobata-derived NVs have demonstrated antiosteoporotic potential, ginseng-derived NVs inhibit osteoclast differentiation, and hemp-derived NVs mitigate inflammation and tissue damage through gut-liver axis modulation (Seo et al., 2023; Zhan et al., 2023).

Marine algae-derived extracts (MAEs) are rich in bioactive components and have been reported to prevent osteoporosis in in vitro and in vivo studies. Furthermore, MAEs rich in calcium and minerals have been shown to enhance osteocalcin and osteoprotegerin (OPG) secretion and improve the OPG/RANKL ratio in ovariectomized mouse models, thereby mitigating bone loss (Deng et al., 2018). Fucoidan, a marine polysaccharide that is abundant in brown algae, has been shown to promote osteoblast differentiation and inhibit osteoclastogenesis because of its anti-inflammatory and antiviral properties (Fitton, 2011; Carson and Clarke, 2018). In addition, a marine polysaccharide SFP-1 derived from Hizikia fusiforme suppressed adipogenesis and promoted bone formation by upregulating the expression of Runx2, OCN, and SPARC, which correlated with the increased bone mass in zebrafish models (Deng et al., 2024). Osteoblast differentiation is regulated by various signaling pathways, including BMP, Wnt/β-catenin, TGF-β, and MAPK. Among these, BMP-2 is a key inducer of Runx2 expression and promotes differentiation and mineralization via the ERK, p38, JNK, and Smad1/5/8 signaling pathways (Ogasawara et al., 2004; Guo and Wang, 2009). The present findings support the hypothesis that bioactive compounds in MAEs can activate the BMP- and MAPK-related signaling cascades to enhance osteogenic differentiation.

However, to comprehensively validate these in vitro results, further in vivo studies, particularly in ovariectomized mouse models, are warranted. Such studies would confirm the bone-protective effects of MAEs and elucidate their potential to modulate the systemic metabolic processes. This could promote the development of MAE-based interventions aimed at preventing or treating osteoporosis as well as a broad spectrum of metabolic disorders, including obesity, dyslipidemia, and insulin resistance, thereby expanding the therapeutic applicability of marine algae-derived bioactives.

In this study, super-resolution imaging using stochastic optical reconstruction microscopy revealed the presence of exosome markers, including CD9, CD81, and CD63, on SfNVs. These markers are widely used to characterize mammalian exosomes, and their detection suggests that SfNVs may share certain structural or antigenic features with extracellular vesicles of mammalian origin. However, given that SfNVs are derived from marine algae, the identification of these markers based on antibody binding should be interpreted with caution. The observed signals may reflect epitope similarity or antibody cross-reactivity rather than definitive evidence of mammalian-type exosomal biogenesis. Therefore, while these findings support the classification of SfNVs as extracellular vesicle-like nanovesicles, further studies employing comprehensive proteomic profiling, lipidomic analysis, and investigations into vesicle biogenesis pathways are required to clarify their precise identity and biological relevance.

Building upon previous research, this study investigated the effect of S. fulvellum-derived SfNVs on osteoblast differentiation. The results demonstrated that the SfNVs increased ALP activity in MC3T3-E1 cells, significantly enhanced the expression of osteogenic markers, including ALP and OPN, and upregulated the production of transcription factors, such as Runx2 and ProCOL I. Moreover, SfNV treatment upregulated BMP-2/4 expression at the mRNA and protein levels and enhanced the phosphorylation of p38, ERK, JNK, and Smad1/5/8.

In this study, the results supported that SfNVs effectively induced the osteogenic differentiation of MC3T3-E1 preosteoblasts. The beneficial effects were demonstrated via the increased ALP activity, formation of mineralized nodules, and upregulation of osteogenic markers, including Runx2 and OPN. Furthermore, the molecular mechanism of SfNVs was elucidated through the upregulation of BMP-2/4 expression and activation of the Smad1/5/8 and MAPK signaling pathways. These findings are consistent with previous reports on the osteogenic induction mechanisms from various natural compounds. As this study is based on in vitro experiments, further validation using in vivo models is required to assess bone regeneration efficacy, in vivo stability, and bioavailability. Nevertheless, SfNVs derived from seaweed reveal potential as functional materials for improving bone quality and suggest future applications in the fields of tissue engineering and functional foods.

These findings suggest that SfNVs promote osteoblast differentiation and mineralization via the activation of the BMP-2/4/MAPK/Smad-1/5/8-dependent Runx2 pathway. This aligns with previous studies demonstrating that plant- or seaweed-derived bioactive compounds, such as NVs and fucoidan, contribute to bone formation via similar mechanisms.

When discussing the potential translational applications of SfNVs, it is worth noting that recent studies have highlighted the role of plant-derived extracellular vesicle-like nanovesicles (PELVs) in regulating bone metabolism. PELVs, which are structurally and functionally comparable to those of mammalian exosomes, exert broad modulatory effects on osteoclasts, osteoblasts, and bone marrow mesenchymal stem cells, while also promoting the regeneration of chondrocytes and exhibiting anti-inflammatory activity. These properties suggest their potential use in alleviating osteoporosis and delaying the progression of osteoarthritis, which aligns with mechanisms proposed in traditional Chinese medicine (Xia et al., 2025). Moreover, when compared to mammalian-derived exosomes, PELVs offer the advantages of broader biological origins and scalable production. However, despite these promising attributes, evidence from human studies remains scarce, and further research is required to clarify their precise effects and establish their translational feasibility.

In conclusion, this study highlights the potential of SfNVs as a functional material to enhance osteogenic differentiation and bone quality, which suggests their applicability in the fields of functional foods and tissue engineering-based bone regenerative therapies.