The hPDLFs were commercially obtained as primary cells from company (hPDLF, #2630, ScienCell Research Laboratories). This cell type was classified as a primary human cell tissue and was not immortalized cell lines. Detailed product information is available at the supplier's website: https://sciencellonline.com/en/human-periodontal-ligament-fibroblasts/. Although the cells were commercially sourced, the study was conducted in accordance with the Declaration of Helsinki and the research design was officially approved by the Institutional Review Board of Kyung Hee University (IRB No. KHSIRB-25-417). The hPDLF cells were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 µg/ml streptomycin in a humidified atmosphere of 5% CO₂ at 37°C. Osteogenic medium (OM) was prepared by supplementing culture media with 50 µg/ml L-ascorbic acid (Sigma-Aldrich; Merck KGaA) and 10 mM β-glycerophosphate (Sigma-Aldrich; Merck KGaA) (14). The hPDLF cells were sub-cultured at 80–90% confluence, and experiments were conducted using passages 4–7.

T-MSCs were obtained from human tonsil tissues, following approval by the Institutional Review Board (Department of Otorhinolaryngology, Yonsei University Wonju College of Medicine, IRB-CR320104), with written informed consent obtained from all donors. This cell type was classified as a primary human cell tissue and was not immortalized cell lines. T-MSCs were cultured in medium supplemented with EV-depleted FBS prepared by ultracentrifugation, as previously described. At 70–80% confluency, the cells were washed, switched to fresh EV-depleted medium, and conditioned for 24–48 h. Conditioned medium (CM) was collected and cleared of cells, apoptotic bodies, and debris by sequential low- to medium-speed centrifugation and 0.22 µm filtration. Small EVs were then isolated from the clarified CM by high-speed ultracentrifugation, washed by a second ultracentrifugation step, and resuspended in Dulbecco's PBS (DPBS). The resulting EVs were then aliquoted and stored at −80°C while minimizing freeze-thaw cycles. The T-MSC-EVs used in the experiments in the present study were manufactured and purified by Hyundai Meditech Co., Ltd. using internal standard operating procedures (SOPs). EV identity, particle size distribution, and concentration were confirmed by nanoparticle tracking analysis (NTA). Transmission electron microscopy (TEM) was performed using a Talos L120C microscope (FEI, USA) operated at 120 kV with a LaB6 electron gun. A carbon-film-coated copper grid (400 mesh, Ted Pella) was rendered hydrophilic using a PELCO easiGlow^™ glow discharge system (Ted Pella) under negative polarity for 20 sec before mounting the samples. For negative staining, a 2% uranyl acetate (UA) solution was freshly prepared. One drop of EV suspension was placed on a glow-discharged grid and incubated for approximately 1 min. Excess liquid was gently removed using filter paper. The grid was then floated on a 20 µl droplet of 2% UA solution for 20 sec for staining, and excess stain was blotted off. Finally, the grid was air-dried for 10 min at room temperature before imaging.

T-MSC-EVs were quantified by NTA and administered to hPDLFs at final concentrations of 1×108 or 5×108 particles/ml, depending on the experimental condition. Particle number-based dosing was used because the total EV protein content was insufficient for reliable quantification and did not necessarily reflect the number of vesicles or functional EV cargo. The selection of these specific doses (1×108 or 5×108 particles/ml) was based on previous studies demonstrating that MSC-derived EVs exert significant immunomodulatory effects within this concentration range in in-vitro human cell models (15).

Total RNA was isolated from cells using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Reverse transcription was performed using AccuPower RT PreMix (Bioneer). qPCR was performed on the cDNA samples using Axen^TM qPCR Master Mix (Macrogen) on a 7500 Real-Time PCR System (Thermo Fisher Scientific, Inc.). The relative mRNA levels of target genes were normalized to β-actin mRNA levels and analyzed using the comparative Ct method (ΔΔCt) (16). The primer sequences used in this study are listed in Table I.

Total cells were lysed in radioimmunoprecipitation assay (RIPA) buffer supplemented with protease and phosphatase inhibitor cocktails. Whole-cell lysates and nuclear extracts were prepared according to standard protocols. Protein concentrations were determined using a protein assay. Equal amounts of protein (25 µg per lane) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes. After transfer, the membranes were rinsed three times with Tris-buffered saline containing 0.1% Tween-20 (TBST) and blocked with a blocking solution for 1 h at room temperature. The membranes were then incubated overnight at 4°C with primary antibodies, followed by incubation with horseradish peroxidase-conjugated secondary antibodies for 1 h at 37°C. Primary antibodies against β-actin, p65, c-Jun, and c-Fos (Santa Cruz Biotechnology, Inc.), as well as phosphorylated or total extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) (Cell Signaling Technology) were used at a dilution of 1:1,000.

For characterization, EVs were lysed in RIPA buffer containing protease inhibitors, and protein concentrations were determined. Equal amounts of vesicle proteins were subjected to SDS-PAGE and transferred onto PVDF membranes, as described above. The membranes were probed with primary antibodies against the exosome markers cluster of differentiation (CD)63 (rabbit monoclonal, Cell Signaling Technology, #52090, 1:1,000) and CD9 (rabbit monoclonal, #13174, 1:1,000). Albumin (rabbit polyclonal, #4929; Cell Signaling Technology, 4929, 1:1,000) was used as a negative marker to evaluate contamination by soluble serum proteins. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system (Amersham), and images were captured using an Amersham^™ ImageQuant^™ 500 imaging system (Cytiva).

Cell proliferation was evaluated using EZ-Cytox (DoGenBio), according to the manufacturer's instructions. Briefly, cells were seeded in 96-well plates at a density of 5×10³ cells/well and cultured under the indicated conditions. At the designated time points, 100 µl of solution, diluted 1:10, was added to the culture was added to each well and incubated for 1 h at 37°C. The absorbance at 450 nm was measured using a microplate reader (Thermo Fisher Scientific, Inc.). Cell proliferation was expressed as a percentage relative to the control group. All experiments were performed in triplicate.

ALP activity was measured using a sensoLyte^® pNpp alkaline phosphatase assay kit (Anaspec). Cells were cultured in 96-well plates, and osteogenic differentiation was induced under the specified conditions. After the designated incubation period, the culture medium was removed, and the cells were gently washed twice with PBS to eliminate serum components that could interfere with the assay. Subsequently, 100 µl of pNPP substrate solution was directly added to each well. The plate was then incubated at room temperature for 1 h, protected from light, and quantified by measuring absorbance at 405 nm using a microplate reader. ALP staining was performed using a TRACP & ALP double staining kit (Takara) according to the manufacturer's instructions. The cells were fixed in a fixation solution for 20 min, washed twice with PBS, and stained with ALP staining solution. After incubation in the dark at room temperature for 1 h, the stained cells were photographed under a light microscope (ECLIPSE TS100; Nikon).

Calcium deposition was assessed by Alizarin Red S staining. Cells were cultured in 24-well plates, and osteogenic differentiation was induced under the specified conditions. The cells were then washed twice with PBS and fixed in 70% ethanol for 20 min at room temperature. After fixation, the cells were rinsed with distilled water and incubated with Alizarin Red S solution (Samchun Pure Chemical) for 2 h in the dark at room temperature. The stained cells were washed thoroughly with distilled water to remove excess dye and then air-dried. The stained calcium deposits were imaged using an ECLIPSE TS100 system (Nikon).

Statistical analyses were performed using GraphPad Prism, version 8.4.3 (GraphPad Software). For the mRNA expression of inflammatory markers (PCR data), ordinary one-way analysis of variance (ANOVA) was performed, followed by Bonferroni's post-hoc test. For all other experiments, two-way ANOVA with Tukey's multiple comparison test was used. All values of *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 were regarded as indicative of statistical significance.

NTA, TEM, and western blotting were performed to confirm the physical characteristics and molecular identity of T-MSC-derived EVs (Fig. 1). The NTA analysis considered only particles ≤200 nm as EV-sized populations; additionally, minor high-diameter shoulders (>250 nm) were excluded as potential aggregates. The size-concentration curve showed a unimodal distribution with a peak at approximately 140–150 nm. The particle concentration was approximately 1.5×108 particles/ml. Consistent with these measurements, TEM revealed round vesicle-like structures within the expected exosomal size range (50–150 nm). The isolated T-MSC-derived EVs showed a typical exosomal morphology without apparent aggregation. Western blotting to assess the expression of EV-associated markers and to further validate the exosomal nature and purity of the isolated vesicles revealed tetraspanins CD63 and CD9 in the T-MSC-derived EVs, confirming the enrichment of exosome-associated proteins. In contrast, albumin, which was used as a negative marker for soluble protein contamination, was not detected in the EV preparations.

Collectively, these results demonstrated that the isolated T-MSC-derived EVs exhibited the size distribution, morphology, and molecular marker profile characteristic of exosomes, indicating successful isolation and high purity (Fig. 1).

Fig. 2 shows the effects of LPS and EVs on the PDL. LPS inhibited cell proliferation in a time- and dose-dependent manner. Although no significant differences were observed on day 2, marked differences were observed among the groups on day 3. T-MSC-EVs effectively prevented the LPS-induced inhibition of cell proliferation at both concentrations, and higher concentrations promoted cell growth to a level exceeding that of the control.

The expression levels of inflammatory cytokines, including IL-1β, IL-6, IL-8, and IFN-γ, were markedly elevated following LPS stimulation in hPDLFs (Fig. 3). In the T-MSC-EVs-treated group, IL-6 and IL-1β levels were stable or slightly increased, with no statistically significant change in IL-1β, while T-MSC-EV treatment significantly reduced IL-8 and IFN-γ levels. These results demonstrated a cytokine response pattern characterized by selective reduction of IL-8 and IFN-γ in the experiment.

LPS stimulation activated classical MAPK pathways, including JNK and ERK, which mediate inflammatory responses and lead to the activation of transcription factors, including AP-1 (c-Jun, c-Fos) and NF-κB (Fig. 4). The EV-treated group showed markedly decreased phosphorylation of MAPK components (ERK and JNK), suggesting that T-MSC-EVs suppressed the upstream inflammatory signaling cascade (Fig. 4A). Furthermore, downstream transcription factors c-Jun, c-Fos, and NF-κB were also inhibited, supporting the anti-inflammatory effects of T-MSC-EVs (Fig. 4B).

The effects of T-MSC-EVs on bone formation were evaluated using ALP activity assay, ALP staining, and Alizarin Red staining over 21 days. LPS treatment markedly reduced ALP activity and mineralization capacity (Fig. 5). However, T-MSC-EV treatment restored or even enhanced ALP activity and mineralized nodule formation, especially at later time points (days 14 and 21), indicating a potential osteogenic effect of T-MSC-EVs despite inflammatory conditions.

Fig. 6 shows changes in gene expression induced by LPS and T-MSC-EVs. LPS treatment resulted in an overall decrease in the mRNA expression of key osteogenic genes, including ALP, BSP, OPN, and OCN, with BSP and OPN exhibiting a more pronounced suppression trend over time. In contrast, treatment with EVs resulted in recovered or increased gene expression, with a significant level of recovery observed for some indicators. In particular, OPN and OCN, which are mid- and late-stage osteogenic differentiation markers (17,18), showed a tendency for expression to increase rapidly after 14 days due to T-MSC-EVs. The levels of SOST, an osteogenic inhibitor (19), increased following LPS treatment but were suppressed again upon T-MSC-EV treatment, suggesting the positive effect of EVs on the normal regulation of osteogenic gene expression, even under inflammatory conditions.