Small Extracellular Vesicle Treatment of Trabecular Meshwork Fibrosis: 2D/3D In Vitro and In Vivo Analyses

Abstract

Purpose

Fibrosis of the trabecular meshwork (TM) is a key pathological mechanism in POAG. Small extracellular vesicles (sEVs), a type of extracellular secretion, have various functions, such as antifibrotic effects and injury repair. This study investigated the antifibrotic effects of human bone marrow mesenchymal stem cell-derived (hBMMSC) sEVs on TM cells in two-dimensional (2D)/three-dimensional (3D) cultures in vitro and in vivo.

Methods

Primary human TM cells were isolated from corneal rings and characterized. We generated 3D TM cultures via scaffoldless 3D culture. sEVs were extracted from hBMMSC supernatants via gradient ultracentrifugation and characterized via electron microscopy and nanometer flow analysis. A TGFβ2-induced fibrosis model was established in 2D/3D TM cell cultures, and the effects of sEVs treatment were assessed via Western blot, immunofluorescence, and morphological analyses. A chronic ocular hypertension mouse model was constructed by injecting the TGFβ2-overexpressing adenovirus Ad-TGFβ2^C226/228S. The hBMMSC sEVs were injected into the anterior chamber 2 weeks later. The intraocular pressure (IOP) and changes in fibronectin (FN) and α-smooth muscle actin (α-SMA) in the iridocorneal angle were determined.

Results

The hBMMSC sEVs significantly reduced FN and α-SMA expression in both the 2D and 3D TM fibrosis models. sEVs also mitigated the TGFβ2-induced reductions in 3D cultured TM volume, porosity, and pore density. In vivo, sEVs injection effectively reduced TGFβ2-induced IOP elevation and decreased FN and α-SMA expression in the iridocorneal angle.

Conclusions

hBMMSC sEVs significantly attenuate TGFβ2-induced TM fibrosis via both protein expression and morphological changes. In addition, hBMMSC sEVs have therapeutic potential in alleviating the TGFβ2-induced increase in IOP linked to TM fibrosis.

Glaucoma is a leading cause of irreversible blindness resulting in optic nerve damage and visual field loss,^1,2 the main risk factor for which is increased intraocular pressure (IOP).^3,4 The outflow of aqueous humor is the decisive factor of IOP, and the main outflow is via the trabecular meshwork (TM)/Schlemm's canal route.⁵ Studies have shown that fibrosis of the TM causes the obstruction of the outflow of aqueous humor.⁶ Previous studies in our group have shown that the abnormality of TM is an important pathological mechanism of POAG.^7,8 The team's preliminary research found that the TGFβ2-overexpressing adenovirus Ad-TGFβ2^C226/228S successfully induced sustained IOP elevation in mice. The antifibrotic effect of the TM may become a new target for glaucoma treatment.

Small extracellular vesicles (sEVs) are external vesicle-like structures secreted by cells and range in diameter of less than 200 nm.⁹ sEVs contain various mRNA and protein components, which are important for material transport and information exchange between cells.¹⁰ Studies have shown that sEVs secreted by mesenchymal stem cells (MSCs) can repair cell damage to a certain extent.^11,12 Human bone marrow MSC (hBMMSC) sEVs have been shown to have antifibrotic effects on various tissues, such as the liver¹³ and kidney.¹⁴ Studies have shown that sEVs regulate gene expression in target cells through their internal microRNAs.^15,16 Compared with conventional drug therapy, sEVs treatment tends to last longer. Compared with stem cell therapy, sEVs have a substantially reduced risk of cancer and immune rejection.¹⁷

Three-dimensional (3D) culture technology can be used to construct a cell growth model with spatial structure in vitro in two ways: scaffold culture and scaffoldless culture.¹⁸ Compared with traditional two-dimensional (2D) cultures, 3D cultured cells are more similar to the growth microenvironment in the body and can better reflect the physiological and pathological changes in cells; thus, these systems are often used in drug screening and pathological mechanistic research on various human tissues.^19,20 Moreover, 3D culture can provide some information about morphological changes. Some studies have shown that, compared with control cells, fibrotic TM cells cultured without scaffolds have notable morphological changes and increased overall hardness.²¹ Therefore, we used in vitro scaffoldless 3D culture as one of the research methods to study the antifibrosis effect of hBMMSC sEVs.

The aim of this study was to investigate the antifibrotic effect of hBMMSC sEVs on TM cells by establishing 2D and 3D cell culture models via in vitro and in vivo experiments and to explore the potential mechanism.

Methods

Extraction and Identification of hBMMSC sEVs

The hBMMSCs (CP-H166, Procell Life Science & Technology, Wuhan, China) were cultured with DMEM/Nutrient Mixture F-12 (DMEM/F12; Thermo Fisher Scientific, Waltham MA, USA), 10% fetal bovine serum (16000044, Gibco, Grand Island, NY, USA) and 1% penicillin-streptomycin (SP; 15140122, Gibco) (complete medium, hereafter). When hBMMSCs reached 80% confluence, the medium was replaced with sEV-depleted serum (C3801, Vivacell, Shanghai XP Biomed, Shanghai, China) medium for an additional 48-hour incubation, and then, the cell supernatant was collected and centrifuged at 300×g for 10 minutes, 2000×g for 10 minutes and 10,000×g for 30 minutes in an SW 32 Ti rotor (Optima XE-100, Beckman Coulter, Brea, CA, USA). The precipitated supernatant was discarded after each centrifugation. The collected supernatant was then centrifuged at 100,000×g for 70 minutes. After the precipitate was collected and washed twice with PBS (GNM14190-5, GENOM Bio, Hangzhou City, China), it was resuspended in PBS. The sEVs suspension was filtered with a 0.22-micron filter (DS-ZSGLQ-02, GREEN MALL, Jiangsu Province, China) and stored at −80°C.²² sEVs were identified via transmission electron microscopy (Tecnai Spirit TEM T12, FEI, Morristown, NJ, USA). The size, concentration, and surface marker expression of the sEVs were analyzed via nanoflow analysis. Nanoflow analysis is a highly sensitive analytical technique specifically designed for characterizing nanoparticles (e.g., sEVs, viruses, and nanomedicines with a size range of 30–1000 nm). Building upon conventional flow cytometry, this technology achieves high-resolution detection of submicron particles through optimized optical systems, advanced signal processing algorithms, and precision fluidic control. The sEVs were diluted to 1 × 10⁶ particles/mL with PBS and incubated with fluorescent antibodies against CD9 (555371, BD Biosciences, New Jersey, USA) and CD63 (561983, BD Biosciences,New Jersey, USA) at 37°C for 30 minutes to label the transmembrane proteins. The solution was centrifuged again at 100,000×g for 70 minutes, the supernatant was discarded, and the free dye was removed by washing and precipitation twice with PBS. The precipitates were resuspended in PBS to 1 × 10⁸ particles/mL and then analyzed via nanoflow analysis (U30E, NanoFCM, Xiamen, China).

sEVs Phagocytosis Experiment

Three to five generations of human TM cells (hTMCs) were cultured on 14-mm diameter cell coverslips (801010, NEST, Jiangsu, China) to 80% confluence. The sEVs were diluted to 1 × 10⁸ particles/mL, incubated with PKH67 (D0031, Solarbio, Beijing, China) stain at 37°C in the dark for 5 minutes, and then incubated at 4°C for 15 minutes. The solution was centrifuged at 100,000×g for 90 minutes, and the supernatant was discarded. The precipitate was washed twice with PBS to remove free dye.²³ In the control group, the sEVs suspension was replaced with PBS, and the other steps were the same. The stained sEVs were diluted to 1 × 10⁸ particles/mL with PBS and added to the TM cell culture medium for 12 and 24 hours. To stain the localization of cytoskeleton and nucleus, hTMCs were stained successively with rhodamine-labeled phalloidin (CA1610, Solarbio) and DAPI with antiquenching agent (8961S, Cell Signaling Technology, Danvers, MA, USA) and observed on the slide with a confocal laser scanning microscope (LSM980, Carl Zeiss, Oberkochen, Germany).

Identification and Culture of Human Trabecular Cells

HTMCs were extracted from the corneoscleral ring after corneal transplantation. The corneoscleral materials were provided by the Gu Jianjun team of Zhongshan Ophthalmology Center of Sun Yat-Sen University. This study was approved by the hospital Ethics Committee. Marginal corneoscleral tissue (2 × 2 mm) with TM attached to the bottom of a six-well plate (3516, Corning, Corning, NY, USA) and cultured with complete medium. The culture temperature was 37°C, and the concentration of CO₂ was 5%. The cells were treated with dexamethasone (D4902-25MG, Sigma-Aldrich, St. Louis, MO, USA) at 100 nM in the medium for 7 days to detect the expression of myocilin, a reliable marker of hTMCs, which was evaluated by Western blotting.⁸

Cell Proliferation Assay

To investigate the effects of fibrosis and sEVs on TM cell activity, we conducted cell proliferation assays. The suspensions of hTMCs were uniformly seeded in 96-well plates (CLS3599-100EA, Corning) and divided into two large groups: the fibrotic group and the sEVs coculture group. The fibrotic cells were treated with 5 ng/mL, 10 ng/mL, or 20 ng/mL TGFβ2 (HY-P7119, MedChemExpress, Monmouth Junction, NJ, USA). For the sEVs coculture group, 20 ng/mL TGβ2 and 10⁶ particles/mL, 10⁸ particles/mL, and 10¹⁰ particles/mL hBMMSC sEVs were added. After treatment, cell viability was quantified at 24 hours via a Cell Counting Kit-8 (CCK-8) assay (CK04-500T, DOJINDO, Rockville, MD, USA). The absorbance at 450 nm was measured via a spectrophotometer (Bio-Rad, Hercules, CA, USA) and normalized to that of the control.

Western Blotting

To examine amounts of the proteins present, we performed Western blotting. The cells were divided into three groups. The control group was cultured with complete medium (DMEM/F12, 10% fetal bovine serum, and 1% SP), the fibrotic group was cultured with medium containing 20 ng/mL TGβ2, and the antifibrotic group was cultured with 20 ng/mL TGβ2 and 10⁸ particles/mL hBMMSC sEVs. Subsequent 2D cell immunofluorescence experiments, qPCR experiments and 3D trabecular cell experiments were grouped via this method. The protein samples were separated on a 4% to 20% SurePAGE Bis-Tris gel (GenScript, Piscataway, NJ, USA) and then transferred onto membranes. The membranes were blocked with rapid blocking solution (P30500, NCM Biotech, Jiangsu, China) for 1 hour at room temperature. The membranes were subsequently probed with primary antibodies against fibronectin (FN), α-smooth muscle actin (α-SMA), and β-tubulin overnight at 4°C. After washes with Tris-HCl buffered saline containing 0.1% Tween-20, the membranes were incubated with secondary antibodies for 1 hour at room temperature. The immunoblots were visualized via an imaging system (ChemiDoc MP, BIO-RAD), and densitometric analysis was performed via ImageJ software. The protein levels were normalized to those of the loading control β-tubulin. The concentrations of the antibodies used in the experiments are shown in Table 1.

Table 1.

Primary and Secondary Antibodies Used for Western Blotting and Immunocytochemistry

Antibodies	Western Blot	Immunocytochemistry	Supplier, Catalog No
Primary antibodies
Anti-myocilin	1:2000	—	Abcam, ab318197
Anti–α-SMA	1:10,000	1:500	Abcam, ab124964
Anti-FN	1:1000	1:500	Abcam, ab2413
Secondary antibodies
Anti-rabbit IgG HRP linked	1:2000	—	Cell Signaling Technology, 7074S
Anti-rabbit 488	—	1:500	Cell Signaling Technology, 4412S
Anti-rabbit 647	—	1:500	Cell Signaling Technology, 4414S

Immunofluorescence Analysis

To detect the expression of fibrosis-associated proteins, we performed immunofluorescence staining assays. For the cell experiments, the cell suspension was plated on a 14mm diameter cell coverslips (801010, NEST) and left overnight. Different groups of cells were treated under these conditions for 24 hours. For the animal experiments, frozen eyeball sections were taken from the corresponding groups. The cells were fixed with 4% paraformaldehyde solution at room temperature for 20 minutes. After three washes with PBS, the samples were permeabilized with 0.5% Triton X-100 (9036-19-5, MP Biomedicals, Irvine, CA, USA) solution for 30 minutes. The samples were subsequently washed three times with PBS and incubated with 10% goat serum at room temperature for 1 hour. Afterward, primary antibodies against FN and α-SMA were added, and the samples were incubated overnight at 4°C. The samples were again washed five times with PBS and incubated with fluorescent secondary antibodies for 1 hour at room temperature in the dark. After three washes with PBS, the samples were sealed with DAPI with antiquenching agent and observed under a confocal laser scanning microscope. The concentrations of the antibodies used in the experiments are shown in Table 1.

Quantitative Real-time PCR (qPCR)

To detect the gene expression of fibrosis-related proteins and matrix metalloproteinase (MMP)-related proteins, we conducted qPCR assays. Total RNA was extracted from hTMCs via the TaKaRa MiniBEST Universal RNA Extraction Kit (#9767, TaKaRa BIO, Shiga, Japan). The RNA was reverse transcribed via the PrimeScript RT reagent Kit with gDNA Eraser (#RR047A, TaKaRa BIO). mRNA was amplified via TB Green Premix Ex Taq II (#RR420A, TaKaRa BIO). The forward and reverse sequences of primers used in this study are listed in Table 2. The thermal cycling conditions were 95°C for 1 minutes, 40 cycles of 20 seconds at 95°C, 30 seconds at 58°C, and 40 seconds at 72°C. β-Tubulin was used as an internal control. qPCR was performed on a QuantStudio 3 Real-Time PCR system (Thermo Fisher Scientific, Rockford, IL, USA), and the data were exported for analysis via the 2−ΔΔct calculation method.

Table 2.

Forward and Reverse Sequences of Primers Used in qPCR

Primers	Forward Sequence (5′-3′)	Reverse Sequence (5′-3′)
β-tubulin	TGGACTCTGTTCGCTCAGGT	TGCCTCCTTCCGTACCACAT
FN1	GCAGGCTCAGCAAATGGTTC	GTCCGCTCCCACTGTTGATTTA
ACTA2	CAGAAGGAGATCACGGCCCTA	CGGCTTCATCGTATTCCTGTTTG
MMP3	GGACAAAGGATACAACAGGGAC	GCTTCAGTGTTGGCTGAGTG
MMP9	ACGCACGACGTCTTCCAGTA	CCACCTGGTTCAACTCACTCC
TIMP1	TCTGGCATCCTGTTGTTGCT	CACGAACTTGGCCCTGATGA

Construction of 3D Cultured hTMCs

The cultured third-generation hTMCs were digested with 0.25% pancreatic enzymes (25200056, Thermo Fisher Scientific, Waltham, MA, USA). The cell suspension was centrifuged, and the supernatant was subsequently removed. The cell precipitate suspension was diluted to 1 × 10⁶ cells/mL with complete medium containing 0.25% methylcellulose and seeded uniformly into each well of a 96-well sphere microplate (4515, Corning) (30 µL/well).²¹ For cell growth, 15 µL of medium containing methylcellulose was changed each time. After 6 days of culture, complete medium, TGFβ2, and sEVs were added to each group, and the culture was continued for another 24 hours.

3D TM Cell Immunofluorescence, Maximum Cross-sectional Area, Internal Pore Density, and Porosity Detection

To assess protein expression levels in 3D-cultured TM cells and conduct morphological analyses, we performed immunofluorescence assays. We fixed 3D TM cells at room temperature for 30 minutes with 4% paraformaldehyde solution. After three washes with PBS, the cells were permeabilized with 0.5% Triton X-100 solution for 30 minutes. The samples were also washed three times with PBS and incubated with 10% goat serum at room temperature for 1 hour. Afterward, anti-FN and anti–α-SMA antibodies were added, and the samples were incubated at 4°C overnight. After being washed with PBS three times, the cells were incubated with a fluorescent secondary antibody in the dark at room temperature for 1 hour. After three washes with PBS, the samples were observed with High-Content Imaging Microscopy (ImageXpress Micro 4, PerkinElmer, Waltham, MA, USA) at 20×. High-content imaging microscopy is an automated microscopy system that combines high-throughput imaging with quantitative multiparameter analysis for cell biology research and drug discovery. The concentrations of the antibodies used in the experiments are shown in Table 1.

Z-stack mode was used to scan 3D hTMCs, and the scanning interval was set to 30 µm. The maximum cross-sectional area was selected from the section with the clearest boundaries of the sphere, and the area was measured according to the system's own program. A representative cross-section with a clear internal structure was selected for each sphere, the fluorescence threshold was set to capture the pore area, the pore area and number were obtained through the system's own program, and then the porosity and pore density were calculated. Porosity equals the pore area divided by the total cross-sectional area of the sphere, and pore density equals the number of pores divided by the total cross-sectional area of the sphere.

TM Cell Morphology Detection by Atomic Force Microscopy (AFM)

To characterize the surface morphology of 3D-cultured TM cells, we performed AFM analysis. AFM is a high-resolution scanning probe microscope that measures atomic-scale forces (e.g., van der Waals, electrostatic, or magnetic interactions) between a sharp tip and the sample surface to generate 3D morphology and physical properties. It operates in air, liquid, or vacuum, enabling nanoscale characterization of conductors, insulators, and even biological specimens. We selected a 60 µm × 60 µm area at the apex of each 3D-cultured cell spheroid for morphologic examination. AFM experiments were performed by an Asylum Research MFP-3D Infinity AFM system. To reduce the effects of porous structures, the force mapping mode was used for the morphology detection of TM cells by using BL-AC40TS-C2 cantilever (Olympus, Tokyo, Japan) with 0.11 nN/nm nominal spring constant.^24,25 The spring constant of the cantilever was obtained by force constant calibration. To ensure the physiological conditions of cells, the cantilever was completely immersed in the liquid of culture medium during the AFM experiments. During the morphology mapping, the cantilever tip was initially approached toward the surface of TM cells at a constant velocity at 15 µm/s. Upon the contact between the AFM tip and the cell surface, a loading force was increased until reaching a predefined value of 2 nN and measuring the surface morphology.

Animals

Wild-type C57BL/6J male mice, aged 6 to 8 weeks and weighing 20 to 26 g, were housed on a 12-hour light‒dark cycle and supplied with standard food and water ad libitum (room temperature: 23 ± 2°C; air humidity: 50%‒60%). The experimental animals were supplied by Jiangsu Jicui Yaokang Biotechnology Co., Ltd. (Nanjing, Jiangsu Province, China) (license number: SCXK(Su)2023-0009). All experimental procedures adhered to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Institutional Animal Care and Use Committee of Zhongshan Ophthalmic Center (Z2022047).

Ocular Hypertension Animal Model and Drug Treatments

To investigate the effects of sEVs on IOP and fibrotic protein expression in the anterior chamber angle, we conducted in vivo animal experiments. The mice were divided into a control group, a fibrosis group and an antifibrosis group. All the mice were intraperitoneally anesthetized with avertin (250 mg/kg; Sigma-Aldrich, Darmstadt, Germany). Proparacaine hydrochloride eye drops (Alcon Couvreur NV, Puurs, Belgium) were used for ocular surface anesthesia. The mice in the fibrotic group and antifibrotic group were injected with 2.5 µL of 1 × 10⁸ pfu/mL Ad-TGFβ2^C226/228S carrying red fluorescent labels (Dongze Bio Co., Guangzhou, China), whereas the control mice were injected with 2.5 µL of Ad-Null in the right eye. The virus carries both a red fluorescent gene and a TGFβ2 overexpression gene. The red fluorescent gene serves as a marker, and the presence of red fluorescence in the TM region confirms successful viral transfection of the cells. The day of virus injection was defined as day 0. IOP was first measured on day 8 and then every 4 days until the end of week 4. On day 14, mice in the antifibrotic group were injected with 2.5 µL of 10⁸/mL hBMMSC sEVs suspension in the anterior chamber of the right eye, whereas the mice in the fibrosis group were injected with 2.5 µL of PBS in the right eye. Tobramycin ointment was applied to the eye surface of the mice after eye surgery to prevent infection.

IOP Measurement

All the mice were anesthetized intraperitoneally. The IOP was measured via an iCare TONOLAB tonometer (iCare Finland Oy, Vantaa, Finland). The IOP was measured six times in each eye, and the mean value was taken as the final data.

Image Analysis

Images of the cells and immunocytochemistry sections were obtained via a confocal laser scanning microscope (LSM980, Carl Zeiss, Oberkochen, Germany) with a 20× objective. ImageJ 1.53f51 was used to quantify the positively stained cells and the TM area, as indicated by the blue color due to the DAPI and red signals from the secondary antibodies.

Statistical Analysis

Individual sample sizes are specified in each figure caption. Data analysis was conducted via SPSS Statistics 22.0 (IBM, Chicago, IL, USA) and Prism 9.0 (GraphPad, San Diego, CA, USA). All data are presented as means ± SDs. For data among multiple groups, one-way or two-way ANOVA with Tukey's multiple comparison test was used. Statistical significance was set at a P value of less than 0.05.

Results

Extraction and Identification of hBMMSC sEVs

Under scanning electron microscopy, hBMMSC sEVs exhibited a characteristic ellipsoidal shape with central concavities approximately 100 nm in diameter, which is consistent with classic sEVs morphology (Fig. 1A). To establish a standard curve for nanoparticle size detection, we employed silica nanospheres with known diameters (53–120 nm) as size standards in nano-flow cytometry. The particle size–signal intensity relationship was fitted based on scattering signals (Figs. 1B, 1C). Nanoflow cytometric analysis revealed that the particle size predominantly ranged from 40 to 160 nm (Fig. 1C), which aligns with the microscopic findings. Additionally, fluorescence-based nanoparticle analysis revealed that approximately 70% of the particles expressed at least one of the classic sEVs markers, CD9 or CD63, with more than 20% coexpressing both markers (Fig. 1D). These results confirmed the morphological and surface marker characteristics of the sEVs.

Figure 1.

Extraction and Characterization of hBMMSC-derived sEVs. (A) Morphology of hBMMSC sEVs observed under transmission electron microscopy. (B) Selection of size calibration standards. (C) Establishment of the standard curve for nanoparticle size detection (R² = 0.99944). (D) Particle size distribution of sEVs measured by nanoscale flow cytometry. (E) Fluorescence analysis results from nanoscale flow cytometry. The FITC-A channel is used to detect CD9-positive particles, while the PC5-A channel detected CD63-positive particles. FITC-H, fluorescence channel; SS-H, scattering signals.

sEVs Uptake Experiments in TM Cells

Compared with that in the control group, the fluorescence intensity in the sEVs coculture groups at 12 hours and 24 hours was significantly greater (P < 0.05) (Fig. 2C). Confocal microscopy revealed that green fluorescence colocalized with the cytoskeletal boundaries, indicating that the hBMMSC sEVs were successfully internalized into the cytoplasm of the TM cells within 24 hours. Furthermore, there was no significant difference in the average fluorescence intensity between the 12-hour and 24-hour groups (P > 0.05) (Fig. 2C), suggesting that most of the sEVs were internalized by the TM cells within the first 12 hours.

Figure 2.

(A) Growth morphology of hTMCs under light microscopy. (B) TM-specific identification: Myocilin expression significantly increased after 7 days of dexamethasone (DEX) treatment (n = 3; P < 0.05). (C) Confocal microscopy showing PKH67 and F-actin expression in control and sEVs co-culture groups at 12 and 24 hours. (D) Comparison of cell viability between fibrosis and treatment groups (n = 4). ***P < 0.001; ****P < 0.0001; ns, not significant.

Effects of TGFβ2 Treatment and sEVs Coculture on Trabecular Cell Proliferative Activity

The cells were divided into two main groups: the fibrosis group and the treatment group. The fibrosis group was subdivided into three groups, and the cells were treated with 5 ng/mL, 10 ng/mL, or 20 ng/mL TGFβ2 for 24 hours. The results of the CCK-8 assay revealed that cell viability decreased with increasing TGFβ2 concentration, with significant differences between the groups (P < 0.05). For the treatment group, the cells were cocultured with 20 ng/mL TGFβ2 or 10⁶/mL, 10⁸/mL, or 10¹⁰/mL hBMMSC sEVs for 24 hours. The results revealed that cell viability was significantly greater in the sEVs-treated groups than in the 20 ng/mL TGFβ2 group (P < 0.05) (Fig. 2D). Furthermore, we found significant differences in cell viability between the groups with 10⁶/mL and higher concentrations (P < 0.05), whereas no significant difference was observed between the 10⁸/mL and 10¹⁰/mL groups (P > 0.05) (Fig. 2D). Therefore, 20 ng/mL TGFβ2 was used to establish the fibrosis model, and 10⁸/mL sEVs were used for antifibrotic treatment in subsequent experiments.

Coculture of hBMMSC sEVs With TM Cells Reduces TGFβ2-induced Fibrosis In Vitro

The antifibrotic effect of hBMMSC sEVs was assessed via Western blotting, immunofluorescence, and qPCR. The cell samples were divided into the control, fibrosis, and antifibrosis groups. Western blot results revealed a significant increase in the fibrosis markers α-SMA and FN in the fibrosis group (P < 0.05) (Figs. 3D, 3E). Compared with the fibrosis group, the antifibrosis groups presented significantly lower levels of these markers, but the levels remained higher than those in the control group (P < 0.05) (Figs. 3D, 3E). The immunofluorescence results were consistent with the Western blot findings (Figs. 3A–C), except that there was no significant difference between the antifibrosis and control groups (P > 0.05). Furthermore, qPCR analysis revealed a significant increase in the expression of FN1 and ACTA2 in the fibrosis group, with a significant reduction in the antifibrosis group compared with the fibrosis group, although the expression was still greater than that in the control group, which is consistent with the Western blot results (Fig. 3F). These three experiments strongly indicate that hBMMSC sEVs can mitigate TGFβ2-induced fibrosis in 2D cultured TM cells. In addition, qPCR was carried out to measure the mRNA expression of MMP-related genes: MMP3, MMP9, and the MMP inhibitor TIMP1. These results demonstrated that, consistent with previous literature, TGFβ2 treatment for 24 hours significantly upregulated the expression of MMP3, MMP9, and TIMP1 in TM cells.²⁶ In comparison with the TGFβ2-treated group, sEVs treatment for 24 hours markedly increased MMP3 expression while decreasing TIMP1 levels, with no significant effect on MMP9 expression.

Figure 3.

Antifibrotic effects of hBMMSC sEVs on 2D cultured TM. (A–C) Confocal microscopy images showing FN and α-SMA expression (n = 6). The y axis represents the mean fluorescence integrated density per unit area (IntDen/Area). (D, E) Western blot analysis of FN and α-SMA protein expression (n = 6). A, antifibrosis group; C, control group; F, fibrosis group. (F) qPCR analysis of FN1 and ACTA2 mRNA levels (n = 6). (G) qPCR analysis of MMP3, MMP9, and TIMP1 mRNA levels (n = 4). Statistical significance: *P < 0.05; ***P < 0.001; ****P < 0.0001; ns, not significant.

Construction and Identification of Scaffold-free 3D Cultured TM Cells

Primary hTMCs were used for scaffold-free 3D culture, and AQP1, a marker of TM and Schlemm's canal cells, was positively stained by immunofluorescence (Fig. 4B), confirming the obtained spheroid structures were composed of TM cells. Via high-content imaging microscopy, we found that trabecular cells spontaneously form spherical structures free in the culture medium under gravity, displaying a 3D structure (Fig. 4A). After adjustment of the focal plane, different Z-axis sections revealed varying morphological characteristics of the 3D trabecular cells, confirming their 3D spatial configuration (Fig. 4A).

Figure 4.

Antifibrotic effects of hBMMSC sEVs on 3D-cultured TM. (A) High-content imaging microscopy of different z axis sections of 3D-cultured TM. (B) Immunofluorescence of AQP1 protein in 3D TM using high-content imaging microscopy. (C) Maximum cross-sectional area and immunofluorescence of FN and α-SMA across groups (n = 8). (D) Measurements of internal porosity and pore density in 3D TM (n = 3). (E) Surface morphology of 3D TM using AFM (lighter areas in the results indicate relatively higher points, while darker areas represent lower points). Statistical significance: *P < 0.05; ***P < 0.001; ****P < 0.0001; ns, not significant.

Coculture of hBMMSC sEVs Reduces TGFβ2-induced Fibrosis in 3D Cultured Trabecular Cells

Scaffold-free 3D cultured trabecular cells were divided into the control, fibrosis, and antifibrosis groups. Similar to those in the 2D culture group, FN and α-SMA immunofluorescence in the fibrosis group significantly increased (P < 0.05), whereas those in the antifibrotic group significantly decreased compared with those in the fibrosis group (P < 0.05) (Fig. 4C). Morphologically, the volume of 3D-cultured trabecular cells in the fibrosis group was significantly smaller than that in the control group, with a reduced maximum cross-sectional area, and this change was reversed by sEVs coculture (Fig. 4C). In the fibrosis group, the cells and extracellular matrix (ECM) were densely packed, whereas in the antifibrotic group, they were relatively loose. The porosity and pore density of the cross-sections in the antifibrotic group were significantly greater than those in the fibrosis group (P < 0.05) (Fig. 4D), suggesting that the sEVs alleviated the collapse of the trabecular cell structure caused by fibrosis. Figure 4E shows the morphology of 3D TM by AFM. The 3D TM in the control group displayed a uniform "hill-like" surface. This surface became compact and irregular after fibrosis induction. Importantly, in the antifibrosis group, although other metrics showed improvement (Figs. 4C, 4D), AFM revealed that the surface morphology was not fully restored to the control state and remained irregular. This finding suggests that the therapeutic effect of sEVs may not completely reverse all aspects of fibrotic structural damage, especially at the fine-structural level of the cell surface.

Coculture of hBMMSCs Reduces TGFβ2-induced Fibrosis in Mouse TM Cells In Vivo

Immunofluorescence confirmed specific expression of the Ad-TGFβ2^C226/228S virus carrying red fluorescent labels in the TM of the mice (Fig. 5B). The baseline IOP did not significantly differ among the control, fibrosis, and antifibrosis groups (P > 0.05) (Fig. 5A). Eight days after Ad-TGFβ2^C226/228S injection, the IOP in the fibrosis group were significantly increased compared with that in the control group (P < 0.05) and had stabilized by day 12 (Fig. 5A). Compared with that in the fibrosis group, the IOP in the antifibrotic group decreased significantly from day 20 to day 28 (P < 0.05), but remained greater than that in the control group (P < 0.05) (Fig. 5A). Immunofluorescence of frozen eye sections revealed notable fibrin deposition in the TM of the fibrosis group, with significantly increased FN and α-SMA expression (P < 0.05) (Figs. 5C, 5D). FN and α-SMA expression in the antifibrosis group was significantly lower than that in the fibrosis group, but still higher than that in the control group (Figs. 5C, 5D).

Figure 5.

Effects of anterior chamber injection of hBMMSC sEVs on IOP reduction and antifibrotic activity in mice with ocular hypertension. (A) IOP trends over 28 days among three groups of mice (n = 8). *Significant differences in IOP between the antifibrotic group and the fibrotic group. #Significant differences between the fibrotic and control groups. (B) Autofluorescence of Ad-TGFβ2^C226/228S in the anterior chamber angle observed in frozen sections using confocal microscopy. (The TM region is demarcated by yellow dashed circles.) (C–E) Immunofluorescence analysis of FN and α-SMA expression in the anterior chamber angle (n = 8). The y axis represents the mean fluorescence integrated density per unit area (IntDen/Area) in the TM region (demarcated by white dashed circles). Statistical significance: *P < 0.05; ***P < 0.001; ****P < 0.0001; ###P < 0.001; ns, not significant.

Discussion

The increase in TM outflow resistance in the aqueous humor is a primary factor contributing to elevated IOP in glaucoma.⁵ In POAG, TM tissue exhibits substantial increases in ECM components such as FN, collagen IA1, and collagen IV,²⁷ suggesting that TM fibrosis is a critical pathological process in POAG progression.^28,29 TGFβ, particularly TGFβ2, is significantly elevated in the aqueous humor of glaucoma patients, especially in POAG patients. Previous studies from our group have shown that TGFβ2 overexpression can cause TM fibrosis, leading to aqueous humor outflow disorder, which can be inhibited by pirfenidone.⁸ The hBMMSC sEVs have been widely demonstrated to exert antifibrotic and tissue repair effects across various tissues.¹³ This study evaluated the antifibrotic effects of these sEVs in 2D adherent TM cell cultures, 3D scaffold-free TM cell cultures, and TM cells from mice in vivo. These sEVs effectively and sustainably reduced the fibrosis-induced IOP elevation in mice.

The sEVs used in this study were obtained from hBMMSCs through the conventional ultracentrifugation gradient method and characterized by scanning electron microscopy and nanoparticle flow analysis. Compared with traditional Western blot methods for detecting surface markers, nanoparticle flow analysis provides more comprehensive information, such as diameter, size, and purity, using minimal samples, ensuring experimental reliability.

Our research findings indicate that hBMMSC sEVs can reduce the expression of fibrosis-related proteins in TGFβ2-induced fibrotic TM cells in both 2D/3D cultures in vitro and in the TM of mice in vivo. Additionally, hBMMSC sEVs mitigate the morphological impact of fibrosis on the TM. However, the specific molecular mechanisms underlying the antifibrotic effects of sEVs on the TM remain unclear. Studies have shown that hBMMSC sEVs inhibited renal fibrosis partially by regulating the Smurf 2/Smad 7 axis, which might be one of the mechanisms of antifibrotic effect on TM cells. Studies have shown that hBMMSC sEVs upregulate MMPs in TM cells, zinc-dependent neutral proteases that promote ECM degradation.²² The miR-451a and miR-125b were shown to inhibit MMP-2 expression, whereas these genes were downregulated in TM cells pretreated with sEVs, suggesting a potential antifibrotic mechanism.²² Our findings regarding MMP-3 and TIMP-1 regulation were consistent with the reported effects of MSC sEVs in TM and other experimental systems.^22,30 Notably, MMP-3 plays a crucial role in ECM dissolution and homeostasis, and its elevated expression may represent a key mechanism through which sEVs alleviate ECM deposition. Several studies have demonstrated that inhibition of the Smad pathway can suppress TGFβ-induced upregulation of MMP3, suggesting that TGFβ may promote MMP3 expression through the TGFβ/Smad signaling pathway.^31,32 TGFβ not only upregulates the expression of MMP3, but also increases the levels of the protease inhibitor plasminogen activator inhibitor and TIMP1, ultimately leading to decreased MMP3 activity and subsequent ECM accumulation.³³ Studies have shown that MSC sEVs can further enhance MMP3 upregulation through the ERK/MAPK pathway, thereby increasing the MMP3/TIMP1 ratio and restoring MMP3 activity, which contributes to their antifibrotic effects.³⁴ The differential regulation of these matrix-remodeling factors suggests a potential therapeutic role for sEVs in modulating ECM dynamics. Our CCK-8 assays revealed that these sEVs also increase TM cell viability, supporting tissue repair. Additionally, sEV-pretreated TM cells exhibited increased oxidative stress resistance.²² In contrast with drugs, sEVs have complex compositions with diverse functions and mechanisms. Further research is needed to determine the precise components and mechanisms underlying their antifibrotic effects.

The 3D culture technique can generate TM tissues with spatial structures that closely mimic the in vivo environment of TM cells, providing more accurate responses to drug stimulation.³⁵ This approach also offers a unique opportunity to study structural and pathological changes in the TM. This study innovatively investigated both the internal structure of 3D trabecular spheroids and the surface morphology of living 3D trabecular spheroids under culture environments, offering new insights into the relationship between TM morphology and pathology in POAG patients.

The TM is a triangular, porous meshwork located in the anterior chamber angle. Under physiological conditions, its structure is relatively loose, with abundant pores that allow aqueous humor to flow through into Schlemm's canal.⁵ In pathological states, increased ECM secretion by TM cells leads to a denser structure, thereby impeding aqueous humor outflow.³⁶ Our 3D-cultured TM cell experiments confirmed that fibrosis indeed results in a more compact TM architecture. AFM revealed that the surface of control 3D TM cells exhibited uniformly sized, evenly distributed, and directionally aligned protrusions. In contrast, TGFβ2-treated 3D TM cells displayed large, irregular protrusions. In humans, the TM consists of three layers: the uveal meshwork, corneoscleral meshwork, and juxtacanalicular tissue.³⁷ The first two layers are composed of regularly arranged sheet-like connective tissue covered with evenly distributed TM cells, whereas the juxtacanalicular tissue appears to be irregular under electron microscopy.⁵ Notably, the juxtacanalicular tissue is the primary source of aqueous humor outflow resistance. Thus, we hypothesize that the orderly arrangement of the TM is functionally linked to aqueous humor drainage. TGFβ-induced disruption of this arrangement may be one mechanism by which it impairs outflow. Although hBMMC sEVs failed to fully restore the TM’s regular surface morphology, they partially alleviated fibrosis-induced TM densification, thereby improving aqueous humor outflow function.

In our animal experiments, we constructed a mouse model of TM fibrosis and chronic IOP elevation by intravitreal injection of the TGFβ2-overexpressing virus Ad-TGFβ2^C226/228S. Previous study from our group has shown that this virus specifically targets the mouse TM and induces sustained elevation of IOP in mice.⁸ The results show that a single anterior chamber injection of hBMMSC sEVs effectively reduced IOP and maintained this effect for at least 12 days and can reduce the expression of fibrosis-related proteins α-SMA and FN in the TM region—a promising duration for therapeutic application. Other studies have shown that hBMMSCs could enhance the activity of RGCS for up to 1 month.^38,39 However, the function of sEVs depends on being engulfed by cells.⁴⁰ Because anterior chamber injection is invasive, future research must focus on safer, more effective methods to deliver sEVs directly into the aqueous humor.

This study still has certain limitations. First, this study used a TM fibrosis model, but TM pathology in POAG encompasses more than fibrosis. The therapeutic mechanisms of sEVs in POAG remain unclear. Although previous studies have shown increased stiffness in fibrotic 3D TM cell models, the effect of sEVs on restoring stiffness has not been addressed.²¹ Additionally, sEVs may act beyond TM-targeting mechanisms to lower IOP.

In conclusion, this study demonstrated that hBMMSC sEVs were internalized by TM cells and showed antifibrotic effects in 2D/3D TM cultures and mouse models, providing new insights into POAG pathology.