Adipose Mesenchymal Stem Cell-Derived Exosomes Promote the Regeneration of Corneal Endothelium Through Ameliorating Senescence

Abstract

Purpose: Human corneal endothelial cells (hCECs) have been considered unable to regenerate in vivo, resulting in corneal decompensation after significant loss of hCECs. adipose-derived mesenchymal stem cell (ASC)-derived exosomes can regenerate tissues and organs. In this study, we investigated whether ASC-derived exosomes could protect and regenerate CECs.

Methods: We performed cell viability and cell-cycle analyses to evaluate the effect of ASC-derived exosomes on the regeneration capacity of cultured hCECs. Transforming growth factor-β (TGF-β) and hydrogen peroxide (H₂O₂) were used to induce biological stress in CECs. The effect of ASC-derived exosomes on CECs was investigated in vivo. ASC-derived exosomes were introduced into rat CECs using electroporation, and rat corneas were injured using cryoinjury. Next-generation sequencing analysis was performed to compare the differentially expressed microRNAs (miRNAs) between ASC-derived and hCEC-derived exosomes.

Results: ASC-derived exosomes induced CEC proliferation and suppressed TGF-β- or H₂O₂-induced oxidative stress and senescence. ASC-derived exosomes protect hCECs against TGF-β- or H₂O₂-induced endothelial-mesenchymal transition and mitophagy. In an in vivo study, ASC-derived exosomes promoted wound healing of rat CECs and protected the corneal endothelium against cryoinjury-induced corneal endothelium damage. Next-generation sequencing analysis revealed differentially expressed miRNAs for ASC-derived and hCEC-derived exosomes. They are involved in lysine degradation, adherens junction, the TGF-β signaling pathway, the p53 signaling pathway, the Hippo signaling pathway, the forkhead box O (FoxO) signaling pathway, regulation of actin cytoskeleton, and RNA degradation based on Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis.

Conclusions: ASC-derived exosomes promoted wound healing and regeneration of endothelial cells by inducing a shift in the cell cycle and suppressing senescence and autophagy.

Human corneal endothelial cells (hCECs) do not regenerate in vivo, and their excessive loss results in corneal decompensation, which requires corneal transplantation.¹ Therefore, protecting and regenerating CECs should be the primary therapeutic goal for corneal endothelial diseases. However, such a treatment has not yet been developed.² The pathogenesis of CECs includes oxidative stress, endoplasmic reticulum (ER) stress, and RNA toxicity, which have been targeted for treatment.^3,4 One potential therapeutic strategy could be to inhibit senescence, because cell-cycle arrest in the G0/G1 phase, which is observed in CECs in vivo, resembles senescence.^5,6 Despite this, treatments that focus on CEC regeneration have not been developed. 

Mesenchymal stem cells (MSCs) are stromal cells that are capable of self-renewal and multilineage differentiation.⁷ The immunomodulatory and regenerative effects of MSCs are mediated by their paracrine activity.⁸ Specially, MSCs secrete biologically active molecules, including cytokines and growth factors, such as interleukin (IL)-1, IL-10, tumor necrosis factor–stimulated gene-6 (TSG-6), fibroblast growth factor (FGF), hepatocyte growth factor (HGF), and platelet-derived growth factor (PDGF), which activate multiple signal transduction pathways. These pathways include phosphoinositide 3-kinase (PI3K)/Akt, Janus kinase (JAK)/signal transducer and activator of transcription (STAT), and mitogen-activated protein (MAP) kinase pathways, which promote survival, proliferation, anti-apoptosis, and extracellular matrix remodeling.^8–10 

Exosomes are cell-derived vesicles present in many biological fluids that are reportedly involved in the paracrine effect of MSC and in cell–cell communication.¹¹ Adipose-derived mesenchymal stem cells (ASCs) secrete cytokines, growth factors, proteins, and extracellular vesicles and have great potential to regenerate tissues.^12,13 ASC-derived exosomes containing lipids, non-coding RNAs, microRNAs (miRNAs), DNA, heat shock proteins, and chaperones can alleviate ischemic injury and regenerate tissues and organs by modulating inflammation, senescence, ER stress, and oxidative stress.¹⁴ They have potential therapeutic applications; however, the effect on CECs and the mechanism of action of ASC-derived exosomes remain unclear. Thus, in this study, we investigated the effect of ASC-derived exosomes on CEC regeneration. We found that TGF-β or oxidative stress induces senescence that inhibits the regeneration of CECs. Furthermore, we revealed that ASC-derived exosomes promote CEC regeneration by inhibiting TGF-β/oxidative stress–induced senescence. We also identified miRNAs derived from ASC-derived exosomes and analyzed their cellular mechanisms. Our findings provide a new therapeutic strategy for CEC. 

Materials and Methods

Human CEC Culture

This study was performed in accordance with the tenets of the Declaration of Helsinki and was reviewed and approved by the institutional review board/ethics committee of the Hallym University Medical Center (NON2022-004). Cells were cultured according to previously published methods.^15–17 The corneas from three donors were used. A human CEC–Desçemet's membrane complex was incubated for 10 minutes in a solution of 0.25% trypsin/0.02% ethylenediaminetetraacetic acid (EDTA). The cells were then plated in six-well plates coated with a fibronectin–collagen (FNC) combination coating mix (Athena Environmental Sciences, Baltimore, MD, USA). Cells were cultured for 10 to 14 days until confluency and then passaged at a ratio of 1:3 using a 0.25% trypsin/0.02% EDTA solution. 

Isolating and Labeling Exosomes From the Conditioned Medium

ASC-derived exosomes were prepared as previously described¹⁸ and provided from Biosolution Co., Ltd. (Seoul, Korea).¹⁸ For labeling with DiO, purified exosomes were incubated with DiO green fluorescence membrane dye (Invitrogen, Waltham, MA, USA) at a final concentration of 2 µg/mL for 1 hour at room temperature. The sample was further purified using the columns provided, and the exosomes were eluted in 200 µL of PBS. The exosomes were stored at −80°C until use. For microscopic analysis, cells grown on cover glasses were incubated with DiO-labeled exosomes in Gibco Opti-MEM (Thermo Fisher Scientific, Waltham, MA, USA) at 37°C for 6 hours. After incubation, the cells were washed with PBS to remove the unbound labeled exosomes and subsequently imaged using a fluorescence microscope (DMi8; Leica, Wetzlar, Germany). 

Treating hCECs With ASC-Derived Exosomes

We seeded hCECs onto a 24-well culture plate at 4 × 10⁴ cells per well and cultured them for 24 hours in growth medium at 37°C under 5% CO₂. The cells were then washed with PBS and treated with recombinant human TGF-β1 (10 ng/mL, ab50036; Abcam, Cambridge, UK) or H₂O₂ (1-mM) with or without ASC-derived exosomes (2 × 10⁸ particles/mL, respectively) for 48 hours. 

Cell Viability and Lactate Dehydrogenase Cytotoxicity Assay

Cells (1 × 10⁴) were cultured in a 96-well plate and treated with ASC-derived exosomes for 48 to 72 hours. Cell viability was measured using a Cell Counting Kit-8 (CCK-8; Dojindo, Kumamoto, Japan) based on the water-soluble monosodium tetrazolium salt, WST-8 (2-[2-methoxy-4-nitrophenyl]-3-[4-nitrophenyl]-5-[2,4-disulfophenyl]-2H-tetrazolium). The plates were incubated with CCK-8 solution for 1 to 2 hours. Cell viability was determined by measuring absorbance at 450 nm using an Agilent BioTek Synergy HTX microplate reader (Agilent Technologies, Santa Clara, CA, USA) and expressed as the mean ± SD as a percentage of the control (100%). The experiments were repeated three times, and a representative experiment is shown (Fig. 2B). 

The lactate dehydrogenase (LDH) cytotoxicity assay was performed using a commercial kit (MK401; Takara Bio, Otsu, Japan) according to the manufacturer's instructions. Cells (1 × 10⁴ cells/well) were placed in 96-well plates and incubated for 48 hours in a humidified atmosphere containing 5% CO₂. The supernatant was transferred to new 96-well plates, mixed with reaction substrates, and incubated at 37°C for 30 minutes. Optical density was measured at 490 nm using the BioTek Synergy HTX microplate reader. Cytotoxicity was expressed as the fold change in the controls after subtracting the corresponding blanks. 

Cell-Cycle Analysis

Cell-cycle analysis was performed using CytoFLEX flow cytometry (Beckman Coulter, Brea, CA, USA) with propidium iodide (PI) staining according to the manufacturer's protocol. The cells were harvested using trypsinization, washed with cold PBS, and fixed with cold 70% ethanol. The samples were stored at −20°C. Cells were treated with 50 µg/mL PI and 100 µg/mL RNase A. 

Immunofluorescent Staining

hCECs were cultured on cover glasses in 12-well plates, washed with PBS, and fixed for 20 minutes in a 4% paraformaldehyde solution. The cells were permeabilized for 10 minutes with 0.5% Triton X-100 and blocked for 1 hour with 1% bovine serum albumin (BSA) at room temperature. After washing, the cells were incubated overnight with either mouse anti-mouse Ki67 (sc-23900; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or goat anti-goat Nestin antibody (sc-21248; Santa Cruz Biotechnology) at 4°C and then washed with PBS. The cells were incubated with either fluorescein isothiocyanate–conjugated goat anti-rabbit IgG antibody or rabbit anti-goat antibody (1:100) for 1 hour at 37°C in the dark and then counterstained with Hoechst 33342 nuclear staining dye (1:2000; Molecular Probes, Eugene, OR, USA). After extensive washing with PBS, the slides were mounted with a drop of mounting medium to reduce photobleaching. Negative control staining was conducted in parallel, omitting the primary antibodies. The slides were observed under the Leica DMi8 fluorescence microscope. 

Quantitative Reverse-Transcription PCR

The miRNA was extracted using the mirPremier microRNA Isolation Kit (Sigma-Aldrich, St. Louis, MO, USA). First-strand cDNA was synthesized using a Mir-X miRNA First-Strand Synthesis Kit (Takara Bio). The real-time quantification of transcripts was performed on a LightCycler 96 (Roche Life Science, Mannheim, Germany) using the AccuPower 2X GreenStar qPCR Master Mix (Bioneer, Oakland, CA, USA). Quantitative reverse-transcription PCR (RT-qPCR) results were presented as fold changes relative to the control. The relative quantification was calculated using the ΔΔCT method. The relative expression was normalized to U6 small nuclear RNA (snRNA). The primers are described in Supplementary Table S1. 

Mitochondrial Oxidative Stress Evaluation

Invitrogen MitoSOX Red was used to measure mitochondrial superoxide production.¹⁹ The cells were incubated with 5-µM MitoSOX reagent for 10 minutes at 37°C in the dark. The fluorescence intensity in each well was measured using a CytoFLEX flow cytometer. 

Senescence-Associated β-Galactosidase Assay

Senescence-associated β-galactosidase (SA-β-gal) staining was performed using a SA-β-gal staining kit (BioVision, Milpitas CA, USA).²⁰ Briefly, after the growth medium was removed from the cells, the cells were rinsed once with PBS. The fixative solution was added to each well, and the cells were fixed for 10 to 15 minutes at 25°C. After washing the cells with PBS, the cells were incubated in the β-galactosidase staining solution at 37°C overnight in a dry incubator. 

Mitochondrial Membrane Potential Assay and Autophagosome Staining

JC-1 dye was used to obtain images of the mitochondrial membrane potential. We used the Muse MitoPotential assay (Merck Millipore, Burlington, MA, USA), which uses MitoPotential dye to detect changes in the mitochondrial membrane potential. The data were analyzed using a CytoFLEX flow cytometer. The cells were seeded in cover glass–bottom dishes, stained with the CYTO-ID Autophagy Detection Kit (Enzo Life Science, NY, USA) for 30 minutes, and then washed with PBS. The cells were observed under a Leica DMi8 fluorescence microscope. 

Western Blot

Radioimmunoprecipitation assay buffer (Biosesang, Seoul, Korea) containing protease (Sigma-Aldrich) and phosphatase (PhosSTOP; Roche, Basel, Switzerland) inhibitor cocktails was used to isolate total cellular proteins. Western blotting was performed using standard protocols. We used 5% skim milk or gelatin to block non-specific binding for 1 hour. The primary antibodies were as follows: mouse anti-human N-cadherin antibody (1∶500 dilution, sc-59987; Santa Cruz Biotechnology); mouse anti-human YAP antibody (1∶500 dilution, sc-376830; Santa Cruz Biotechnology); rabbit anti-phospho-YAP antibody (1∶500 dilution, PA5-17481; Invitrogen); mouse anti-LC3 antibody (1∶1000 dilution, M186-3; MBL International Corporation, Woburn, MA, USA); or rabbit anti-GAPDH antibody (1∶5000 dilution, LF-PA0212; Abfrontier, Seoul, Korea). Horseradish peroxidase (HRP)-conjugated secondary antibody and a WEST-Queen Western Blot Detection Kit (iNtRON Biotechnology, Seongnam, Korea) were used to detect the immunoreactive bands. Data were quantified using video image analysis (Luminograph II; Atto, Tokyo, Japan). Protein bands were measured using densitometry. 

Animal Study

This study was approved by the Institutional Animal Care and Use Committee of Hallym University Medical Center (2021-3-0610-30). All procedures were performed according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Six-week-old female Sprague Dawley rats (Raonbio, Yongin, Korea) were used for this procedure, and six Sprague Dawley rats were included in each group. 

In Vivo Treatment of Rat CECs With Exosomes

ASC-derived exosomes (4.5 × 10⁸/10 µL or 2.25 × 10⁸/10 µL) or RNase-treated ASC-derived exosomes (4.5 × 10⁸/10 µL) or hCEC-derived exosomes (4.5 × 10⁸/10 µL) were injected into the anterior chamber of Sprague Dawley rats. A 7-mm-diameter tweezer electrode (BTX Harvard Apparatus, San Diego, CA, USA) was placed on both corneas, and electroporation was performed using the ECM 830 Square Wave Electroporation System (BTX, Cambridge, UK). The parameters for electroporation were as follows: five times, 100-ms duration each at 100 V with a 950-ms interval. Then, cryoinjury was employed to induce the corneal endothelial injury, which has been described previously,^21–23 in contact with a metal rod of 3-mm diameter for 15 seconds that had been frozen in liquid nitrogen for 10 minutes. Then, corneas were irrigated with normal saline. 

Clinical Evaluation

Corneal opacity was evaluated using photographs obtained on days 2, 4, 7, 10, and 14. Corneal opacity was graded as follows: grade 0, no opacity; grade 1, mild opacity with visible iris detail; grade 2, moderate opacity with partial masking of the iris; and grade 3, severe opacity without a view of iris detail. 

Alizarin Red S Staining

Vital staining was performed with 0.2% Alizarin Red S in 0.9% NaCl (pH 4.2) for 90 seconds, and the corneas were fixed in 2.5% glutaraldehyde. Corneal buttons were removed, placed on slides, and mounted under a coverslip with a drop of 0.9% NaCl. Corneal endothelium was observed under a Leica DMi8 light microscope. The cell density was then calculated. 

miRNA Isolation and Sequencing

All procedures for next-generation sequencing (NGS) analysis were performed by Macrogen (Seoul, Korea). Exosomal miRNAs were extracted using the miRNeasy Serum/Plasma Kit (217184; QIAGEN, Hilden, Germany) after exosome extraction from the conditioned medium of hCECs using Exo2D-for RNA (EP-RU10; Exosomeplus, Seoul, Korea). 

The samples were processed to produce exosomal RNA (10 ng) as an input for each library. Small RNA libraries were constructed using a SMARTer smRNA-Seq Kit for Illumina (Takara Bio), according to the manufacturer's guidelines. Sequencing libraries were constructed using polyadenylation, cDNA synthesis, and PCR amplification. cDNA fragments were sequenced by read length using a sequence-by-synthesis method on the Illumina platform. 

Analysis of miRNA Expression Levels

After sequencing, raw sequence reads were filtered based on their quality. The adapter sequences were trimmed from the raw sequence reads. Ribosomal RNA (rRNA)-removed reads were sequentially aligned to the reference genome, miRBase 22.1, and non-coding RNA database, RNAcentral 14.0, to classify known miRNAs and other types of RNA such as transfer RNA (tRNA), snRNA, and small nucleolar RNA (snoRNA). Novel miRNA prediction was performed using miRDeep2. The raw data were normalized using the trimmed mean of the M-values (TMM) in edgeR. One was added to the normalized read count of the filtered miRNAs to facilitate log2 transformation and draw a correlation plot. For each miRNA, logCPM and log fold changes were calculated between the groups. We conducted a statistical hypothesis test to compare the two groups using the exactTest function in EdgeR. miRNAs differentially expressed between the two groups were defined as those with a fold change ≥ 2 and a raw P value of <0.05. We also performed hierarchical clustering analysis using complete linkage and Euclidean distance as measures of similarity to display the expression patterns of the differentially expressed miRNAs that satisfied the criteria of a fold change ≥ 2 and raw P value of <0.05. All data analysis and visualization of the differentially expressed genes were conducted using the R 3.6.2 program (R Foundation for Statistical Computing, Vienna, Austria). 

Prediction of miRNA Targets and Functional Analysis of Differentially Expressed miRNAs

To investigate the biological role of the 12 dysregulated miRNAs, potential target genes were predicted using TargetScan 7.0 or microT-CDS 5.0 and DIANA Tools software.²⁴ Enrichment analysis, including pathway and functional annotation of differentially expressed miRNA target genes (MTGs) and dysregulated genes, was performed using the ShinyGo and DIANA-miRPath 3.0 web server.²⁴ To investigate the potential functional mechanism of the dysregulated miRNAs, we employed Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis^25,26 using the DIANA-miRPath 3.0 web server²⁴ with a false-discovery rate cutoff of P = 0.05. 

Prediction Statistics

Data are expressed as mean ± SD. An independent t-test and one-way and two-way analyses of variance (ANOVAs) were used. All statistical analyses were performed using Prism 9.3.1 GraphPad, San Diego, CA, USA). Statistical significance was set at P < 0.05. 

Results

Cellular Uptake of ASC-Derived Exosomes and Cell Proliferation

Initially, to investigate the uptake of exosomes, ASC-derived exosomes were stained with DiO for visualization and delivered using Opti-MEM I, a reduced serum medium. Transfected DiO-labeled exosomes were visualized with green fluorescence in hCECs using fluorescence microscopy (Fig. 1A). CCK-8 assays and cell-cycle analyses were performed to investigate whether ASC-derived exosomes affect cell viability or proliferation. Cell viability and cell-cycle analysis showed that ASC-derived exosomes induced the proliferation of hCECs and a shift in the cell cycle from the G0/G1 phase to the S phase in a dose-dependent manner (Figs. 1B–1F). RT-qPCR showed that miR-23a-3p and miR-302-3p were increased in exosome-treated cells (Fig. 1G). 

Figure 1.

 

ASC-derived exosomes promote the proliferation of cultured CECs. (A) Uptake of DiO-labeled exosomes in cultured CECs is shown in green. Scale bar: 100 µm. (B) Cell viability was evaluated using the CCK-8 assay. (C–F) Cell cycles were analyzed using propidium iodide staining and flow cytometry. One-half exosome = 1 × 10⁸ particles/mL. (G) RT-qPCR showed that miR-23a-3p and miR-302-3p were increased in exosome-treated cells. Data represent mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

ASC-Derived Exosomes Protect hCECs Against TGF-β- or H₂O₂-Induced Senescence

To investigate the effect of ASC-derived exosomes on cell viability under genotoxic stress, TGF-β or H₂O₂ was used to accelerate senescence,^27,28 and CCK-8 cell viability and LDH cytotoxicity were observed. TGF-β and H₂O₂ reduced cell viability and increased cytotoxicity that was alleviated by ASC-derived exosomes (Figs. 2A, 2B). We evaluated mitochondrial oxidative stress levels using the MitoSOX probe²⁹ and senescence using SA-β-gal staining (Figs. 2C–2F).²⁰ Mitochondrial oxidative stress levels and the number of SA-β-gal–positive cells increased after exposure to TGF-β or H₂O₂, demonstrating senescence and its associated reactions. ASC-derived exosomes alleviated the TGF-β- or H₂O₂-induced senescence in hCECs. 

Figure 2.

 

ASC-derived exosomes on TGF-β- or H₂O₂-induced senescence. (A) Cell viability was measured by CCK-8. (B) Cytotoxicity was assessed by lactate dehydrogenase (LDH) assay. (C, D) Mitochondrial oxidative stress level was measured using MitoSOX Red (500 nM of final concentration). The fluorescence intensity of MitoSOX was measured using a CytoFLEX flow cytometer. (E, F) Cellular senescence was assessed using senescence-associated β-galactosidase. Blue indicates SA-β-gal–positive senescent cells. Scale bar: 50 µm. Data represent mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

ASC-Derived Exosomes Protect hCECs Against TGF-β- or H₂O₂-Induced Endothelial–Mesenchymal Transition

We assessed the N-cadherin levels to investigate the effect of ASC-derived exosomes on endothelial–mesenchymal transition (EMT).³⁰ TGF-β or H₂O₂ upregulated N-cadherin levels, considered to be EMT markers, and ASC-derived exosomes alleviated this effect (Figs. 3A, 3B). We assessed the Hippo signaling pathway, which is responsible for the EMT and consists of pYAP and YAP.^31,32 pYAP was upregulated by ASC-derived exosomes (Figs. 3C, 3D), indicating activation of the Hippo signaling pathway. Furthermore, ASC-derived exosome–treated cells resulted in a decrease in the number of cells with YAP in the nucleus, although the exposure to TGF-β or H₂O₂ induced the nuclear translocation of YAP (Fig. 3F). To investigate whether inhibition of EMT by ASC-derived exosomes promoted proliferation, we performed immunofluorescence staining of Ki67, a proliferation marker. The ratio of Ki67-positive cells was increased by ASC-derived exosomes against TGF-β- or H₂O₂-induced EMT (Figs. 3G, 3H). 

Figure 3.

 

ASC-derived exosomes on TGF-β- or H₂O₂-induced endothelial-to-mesenchymal transition (EMT). (A, B) N-cadherin expression was evaluated by western blot analyses. (C, D) YAP and pYAP expression was evaluated by western blot analyses. (E) Immunofluorescence staining was used to evaluate the nuclear location of YAP. Scale bar: 50 µm. (F, G) Immunofluorescent staining of Ki67 was performed as a proliferation marker. Scale bar: 100 µm. Data represent mean ± SD. *P < 0.05.

ASC-Derived Exosomes Protect hCECs Against TGF-β- or H₂O₂-Induced Mitophagy

To assess the effect of ASC-derived exosomes on cell death, we evaluated the mitochondrial membrane potential and autophagy. Mitochondrial membrane potential was visualized using JC-1 and assessed using the Muse MitoPotential Kit (EMD Millipore Corporation, Hayward, CA, USA) and flow cytometry (Figs. 4A–4C). TGF-β or H₂O₂ reduced the mitochondrial membrane potential, and this effect was alleviated by ASC-derived exosomes. Autophagy was investigated using an autophagy detection kit and western blotting of light chain 3 (LC3).³³ TGF-β and H₂O₂ significantly promote autophagosomes and the conversion of LC3I to LC3II that occurs during autophagy. ASC-derived exosomes attenuated TGF-β- or H₂O₂-induced autophagic responses. 

Figure 4.

 

ASC-derived exosomes on TGF-β- or H₂O₂-induced mitophagy. (A–C) Mitochondrial membrane potential was imaged using a JC-1 probe (A) and measured using the Muse MitoPotential assay (B, C). Scale bar: 50 µm. (D) An autophagosome detection assay was used to assess autophagy. Green indicates autophagosome staining and blue indicates Hoechst 33342 nuclear staining. Scale bar: 50 µm. (E, F) Autophagy was assessed by the conversion of LC3I to LC3II using western blot analysis. Data represent mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

Rat Corneal Endothelium Takes Up ASC-Derived Exosomes That Promote the Wound Healing In Vivo and Protect the Corneal Endothelium Against Cryoinjury-Induced Corneal Endothelium Damages

An animal experiment was performed to confirm the effects of ASC-derived exosomes. First, we determined whether ASC-derived exosomes could enter the rat corneal endothelium by electroporation. Exosomes stained with DiO—which stains the cell membrane—showed green fluorescence in the corneal endothelium of the exosome-injected group (Fig. 5A). Cryoinjury led to apoptosis, which was detected by TUNEL staining (Fig. 5B). After cryoinjury with liquid nitrogen, the corneas became opaque because of corneal endothelial damage. Corneal opacity decreased faster in the exosome-injected group (Figs. 5C, 5D). Corneal endothelial density, evaluated using Alizarin Red S staining, was increased in the exosome-injected group (Figs. 5E, 5F). 

Figure 5.

 

Protective effect of ASC-derived exosomes on cryoinjured corneal endothelium of rats. (A) Uptake of DiO-labeled exosome in rat corneal endothelium is shown in green. Scale bar: 50 µm. (B) TUNEL staining showed the apoptotic cells on the corneal endothelium after cryoinjury. (C, D) Corneal opacity was evaluated as time went on. (E, F) Corneal endothelial density was measured using Alizarin Red S staining. Scale bar: 100 µm. Data represent mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

To confirm corneal endothelial regeneration, we performed immunofluorescence staining for Ki67, a proliferation marker.³⁴ Ki67-positive cells were more frequent in the exosome-injected group compared with the untreated group (Figs. 6A, 6B). To investigate stem cell activity, we performed immunofluorescence staining for Nestin, a neuroepithelial stem cell protein.³⁵ Nestin-positive cells were higher in numbers in the exosome-injected group compared with the untreated group (Figs. 6C, 6D). To investigate the mitochondrial membrane potential in vivo, JC-1 was used to stain mitochondria in the corneal endothelium of Sprague Dawley rats. The red fluorescence intensity, which indicates healthy and polarized mitochondria, was higher in the exosome group than in the control group (Fig. 6E). The MitoSOX red probe was used to assess mitochondrial oxidative stress levels in the corneal endothelium of Sprague Dawley rats. MitoSOX red intensity in the rat corneal endothelium was lower in the exosome group than in the control group, indicating lower oxidative stress in the former (Fig. 6F). 

Figure 6.

 

Regenerative effect of ASC-derived exosomes on cryoinjured corneal endothelium of rats. (A, B) Cells were stained with Ki67 as a proliferation marker. Scale bar: 50 µm. (C, D) Nestin expression was evaluated using immunofluorescence staining. Scale bar: 50 µm. (E) Mitochondrial membrane potential was assessed using JC-1 in rat corneal endothelium. Scale bar: 50 µm. One-half exosome = half dose of exosome group. (F) MitoSOX Red was used to evaluate the mitochondrial oxidation in rat corneal endothelium. Scale bar: 50 µm. Data represent mean ± SD. *P < 0.05; **P < 0.01; ***P < 0.001.

miRNAs in ASC-Derived Exosomes

We investigated the differences in miRNAs between ASC-derived- and hCEC-derived exosomes. The volcano plot (Fig. 7) shows the differences in miRNA expression levels between ASC-derived exosomes and hCEC-derived exosomes. The ASC-derived exosomes showed higher expression levels of hsa-miR-23a-5p, hsa-miR-196a-5p, hsa-miR-4301, hsa-miR-489-3p, hsa-miR-1538, hsa-miR-6126, hsa-miR-762, hsa-miR-1246, hsa-miR-1228-5p, hsa-miR-193b-5p, hsa-miR-4647, hsa-miR-486-3p, hsa-miR-619-5p, hsa-miR-6511b-3p, hsa-miR-8485, hsa-miR-1306-5p, and hsa-miR-1290 (Table 1) and lower expression levels of hsa-miR-423-3p, hsa-miR-222-3p, hsa-miR-25-3p, hsa-miR-29b-3p, hsa-miR-143-3p, hsa-miR-21-5p, hsa-miR-23b-3p, hsa-miR-29a-3p, hsa-miR-221-3p, hsa-miR-186-5p, hsa-let-7f-5p, hsa-miR-130a-3p, hsa-miR-125a-5p, hsa-miR-34a-5p, hsa-miR-125b-5p, hsa-miR-7i-5p, hsa-miR-7e-5p, hsa-miR-199a-5p, hsa-miR-23a-3p, hsa-let-7a-5p, hsa-let-7a-3p, hsa-let-7g-5p, hsa-miR-191-5p, hsa-miR-199a-3p, hsa-miR-199b-3p, hsa-miR-574-3p, hsa-miR-24-3p, hsa-let-7d-5p, and hsa-miR-7b-5p than the hCEC-derived exosomes (Table 1). 

Figure 7.

 

(A) Differentially expressed miRNA between ASC-derived exosomes and hCEC-derived exosomes. Volcano plot shows ASC-derived exosomes versus hCEC-derived exosomes. (B) Schematic diagram.

Table 1.

 

Identification of Differentially Expressed miRNAs in ASC-Derived Exosomes Via RNA Sequencing

 

We predicted the potential target genes of these differentially expressed miRNAs to further explore their function using miRbase and miRDB. The significant functions and signaling pathways in the KEGG pathway analysis were analyzed using miRPath 3.0. Pathway analysis of miRNAs overexpressed in ASC-derived exosomes showed lysine degradation, adherens junction, the TGF-β signaling pathway, the p53 signaling pathway, the Hippo signaling pathway, non-small cell lung cancer, the forkhead box O (FoxO) signaling pathway, regulation of actin cytoskeleton, and RNA degradation (Table 2). Pathway analysis of miRNAs underexpressed in ASC-derived exosomes showed lysine degradation, cell cycle, adherent junction, the Hippo signaling pathway, the p53 signaling pathway, the TGF-β signaling pathway, regulation of actin cytoskeleton, the Wnt signaling pathway, the FoxO signaling pathway, the mitogen-activated protein kinase (MAPK) signaling pathway, lysosomes, DNA replication, and apoptosis (Table 2). 

Table 2.

 

Enriched KEGG Pathways Regulated by miRNAs Overexpressed or Underexpressed in ASC-Derived Exosomes

 

Discussion

MSC-derived exosomes are known for their effectiveness in regenerating multiple organs.^36,37 Among MSCs, ASCs are readily available, safe, and effective for tissue regeneration and inflammation control.³⁸ ASC-derived exosomes have several advantages over ASCs, including smaller size, better tissue penetration, lack of cellular reactions, ease of storage, low immunogenicity, non-cytotoxicity, and non-mutagenic properties.³⁹ In this study, we investigated the effects and mechanisms of action of ASC-derived exosomes on promoting the wound healing and migration of CECs. 

We first demonstrated that ASC-derived exosomes promoted wound healing and the migration of CECs both in vitro and in vivo. To create a CEC disease model through direct damage and secondary inflammatory responses, cryoinjury has been used to destroy CECs.^23,40,41 The activation of Nestin-stained stem cells by ASC-derived exosomes results in the promotion of wound healing of CECs, thereby improving the transparency of the rat cornea. ASC-derived exosomes have multiple biological activities due to their packed cell-type–specific combinations of proteins (cytoskeletal proteins, transmembrane proteins, and heat shock proteins), nucleic acids (DNA, mRNA, miRNA, and long and short non-coding RNA), lipids, and enzymes (GAPDH, ATPase, and pgk1), which shuttle these active cargoes between different cells that are involved in a complex intercellular communication system.^42,43 The miRNAs in ASC-derived exosomes, such as let-7a-5p, miR-23a-3p, miR-29b-3p, miR-302-3p, and miR-1246, regulate pluripotency, inhibit the EMT, and promote wound healing.⁴⁴ Using RT-qPCR, we showed that miR-23a-3p and miR-302-3p were increased in exosome-treated cells. TNF-stimulated gene 6 (TSG-6) and cytokines, such as IL-1 and IL-10, modulate the immune response, and growth factors, such as fibroblast growth factor (FGF), hepatocyte growth factor (HGF), and platelet-derived growth factor (PDGF), promote wound healing.^45–48 These factors are believed to have regenerative effects on the corneal endothelium. 

In this study, we investigated the mechanism by which ASC-derived exosomes promoted the wound healing and migration of the corneal endothelium. We demonstrated that ASC-derived exosomes inhibit the EMT, which is a major mechanism that induces corneal endothelial dysfunction and opacity.^27,49 The EMT is induced by TGF-β and oxidative stress, which suppress tissue regeneration and promote corneal endothelial cell (CEC) dysfunction.^49,50 Hippo signaling, which is composed of yes-associated protein (YAP) and transcriptional coactivator with PDZ-binding motif (TAZ), is one of the signaling pathways of the EMT and induces the EMT in a transcriptional enhanced associate domain (TEAD)-dependent manner.⁵¹ YAP in the nucleus increases the expression of EMT genes such as SNAI, CTGF, BIRC5, and CYR61.⁵¹ YAP is degraded by ubiquitin-specific peptidase 9 X-linked (USP9X) after being phosphorylated to pYAP when it emerges from the cytoplasm.⁵² Thus, upregulation of pYAP results in suppression of the EMT.³¹ 

In addition to their inhibiting the EMT, we also showed that ASC-derived exosomes promote wound healing of the corneal endothelium by inhibiting senescence through the inhibition of reactive oxygen species (ROS). TGF-β and oxidative stress induce ROS, which in turn induces senescence and inhibits CEC regeneration by inducing cell hypertrophy and G0/G1 cell-cycle arrest.²⁷ ASC-derived exosomes downregulated senescence^53,54 and attenuated senescence-associated secretory phenotype-related activity through p16INK4a and IL-6.⁵⁴ Therefore, ASC-derived exosomes have the potential to regenerate the corneal endothelium by inhibiting the EMT and senescence. 

Our study demonstrated that ASC-derived exosomes can improve senescence both in vitro and in vivo by improving mitochondrial function and inhibiting autophagy. Depolarization of the mitochondrial membrane potential indicates mitochondrial dysfunction, also observed in early cell death.⁵⁵ Mitochondria play a crucial role in the survival and function of CECs, which require a significant amount of energy to maintain dehydration of the corneal stroma.⁵⁵ Oxidative stress and mitochondrial dysfunction are known to promote senescence, and we observed that ASC-derived exosomes improved mitochondrial integrity and reduced ROS levels in CECs. Specifically, ASC-derived exosomes restored mitochondrial function by transporting mitochondrial proteins, lipids, and mtDNA to and from the organelle. 

Autophagy is a type of cell death that can contribute to cell survival through recycling biomolecules. However, autophagy can also inhibit cell proliferation and regeneration.⁵⁶ In this study, we found that ASC-derived exosomes inhibited autophagy, which contributed to CEC regeneration. The EMT is associated with mitochondrial dysfunction, increased ROS levels, and decreased mitochondrial membrane potential.^57,58 Autophagy promotes the EMT through TGF-β–dependent signaling.^59,60 Therefore, inhibiting autophagy inhibits the EMT and restores mitochondrial function, leading to CEC regeneration. 

Cryoinjury has been used to cause injury in CECs in the in vivo model, which stimulates the Stk11-p53 signaling pathway as well as apoptosis.^21,40 In this study, apoptosis was elevated in response to cryoinjury-induced cellular damage and CEC density was reduced in cryoinjured cornea. Cryoinjury induces desquamation of CECs into the anterior chamber, and apoptosis helps to clear the damaged cells.^23,61 

Previously, the effects of MSC-derived extracellular vesicles (EV) and blood serum–derived EVs were compared.⁶² The study showed that EVs inhibited tunicamycin-induced ER stress and apoptosis, improving the survival of CECs.^62,63 Different from that study, our study revealed that ASC EVs promote the proliferation of CECs and suppress TGF-β- and H₂O₂-induced senescence through regulation of Hippo signaling, mitochondrial function, and autophagy. A schematic diagram is shown in Figure 7B. 

In this study, we investigated the differentially expressed miRNAs between ASC-derived and hCEC-derived exosomes and summarized the related pathways in Table 2. The differentially expressed miRNAs are associated with lysine degradation, proliferation, the TGF-β signaling pathway, the p53 signaling pathway, the Hippo signaling pathway, the FoxO signaling pathway, and regulation of the actin cytoskeleton. Lysine degradation occurs in the mitochondria and regulates mitochondrial energy production.^64,65 In this study, we showed that ASC-derived exosomes activate mitochondrial function. Furthermore, ASC-derived exosomes were shown to inhibit the TGF-β and Hippo signaling pathways, as well as senescence, which is associated with the p53 signaling pathway. 

The functions of several miRNAs overexpressed in ASC-derived exosomes have been previously described. miR-8485 activates the Wnt/β-catenin pathway⁶⁶ and promotes cell proliferation and migration.⁶⁷ miR-1298 attenuates myocardial ischemia–reperfusion injury by targeting protein phosphatase 2,⁶⁸ protecting cells against damage caused by chronic oxidative stress by inhibiting the TGF-β₂/Smad4 pathway, activating the canonical Wnt pathway, and inhibiting autophagy.⁶⁹ miR-619-5p promotes proliferation and suppresses the PTEN/AKT/p21 pathway.⁷⁰ miR-486-3p is regulated by the TGF-β/SMAD pathway and exerts an antifibrotic effect.⁷¹ miR-193b is downregulated by TGF-β1 and regulates TGF-β signaling by targeting the TGF-β receptor type 3.^71–73 miR-1228 prevents cellular apoptosis⁷⁴ and suppresses the EMT.⁷⁵ miR-762 is predominantly translocated to the mitochondria and participates in regulation of mitochondrial function.⁷⁶ miR-489-3p regulates the oxidative stress response,⁷⁷ inhibits Toll-like receptor 4/NF-κB signaling to prevent inflammation,⁷⁸ and reduces the expression of profibrotic markers induced by TGF-β1.⁷⁹ miR-196a-5p modulates gastric cancer stem cell–like characteristics by targeting SMAD4 and mitigating renal fibrosis by targeting TGF-β receptor 2.^80,81 miRNA-23a regulates the EMT in endometrial endometrioid adenocarcinoma by targeting SMAD3.⁸² miR-1306-5p targets TGF-β receptor 2 and inhibits the transcription factor SMAD4.⁸³ miR-1290 regulates stemness by targeting the retinoic acid receptor-related orphan receptor A (RORA).⁸⁴ 

Furthermore, miRNAs downregulated in ASC-derived exosomes compared with hCEC-derived exosomes have been described. Most of them, including hsa-let-7d-5p, hsa-let-7e-5p, hsa-let-7f-5p, hsa-let-7i-5p, hsa-let-7a-5p, hsa-miR-23a-3p, hsa-miR-34a-5p, hsa-miR-122-5p, hsa-miR-125a-3p, hsa-miR-125a-5p, hsa-miR-125b-5p, hsa-miR-181a-5p, hsa-miR-221-3p, hsa-miR-222-3p, hsa-miR-574-3p, and hsa-miR-574-5p, have been reported to be upregulated in senescent cells.^85–87 miR-423-3p drives the EMT by targeting CYBRD1.⁸⁸ miR-222-3p promotes the EMT by targeting E-cadherin and Notch signaling.^89,90 miR-25-3p regulates oxidative stress by upregulating NOX4.⁹¹ miR-29 induces cellular senescence through multiple signaling pathways⁹² and promotes particulate matter–induced inflammatory responses by regulating the AMP-activated protein kinase (AMPK) pathway.⁹³ miR-143-3p is a TGF-β-inducible miRNA ^94,95 that indirectly and negatively regulates ER stress.⁹⁶ miR-130a-3p inhibits the TGF-β–induced EMT via SMAD4.^97,98 miR-21 is involved in the TGF-β1 signaling pathway⁹⁹ and suppresses the Hippo signaling pathway.¹⁰⁰ miR-23b-3p activates TGF-β signaling by increasing SMAD2 phosphorylation¹⁰¹ and regulates autophagy by suppressing SIRT1.¹⁰² miR-29, miR-143, and miR-21 inhibit mitophagy.^3,103,104 miR-221-3p facilitates the EMT and promotes cell migration.¹⁰⁵ miR-186-5p inhibits cell proliferation and induces apoptosis.¹⁰⁶ miR-138a-3p induces senescence in renal carcinoma cells by targeting enhancer of zeste homolog 2 (EZH2)¹⁰⁷ and reinforces two separate senescence-regulating branches in progerin-expressing endothelial cells: the p53- and p16-associated pathways.¹⁰⁸ miR-199a-3p/5p participates in TGF-β- and epidermal growth factor (EGF)-induced EMT by targeting DUSP5/MAP3K11.¹⁰⁹ miR-191-5p is induced by hypoxia and regulates TGF-β signaling.^110,111 miR-199b-3p suppresses the growth and progression of ovarian cancer via the EMT signaling pathway by targeting ZEB1.¹¹² miR-574-3p is induced by TGF-β1 and induces apoptosis.^72,113 miR-24-3p inactivates the Hippo signaling pathway.¹¹⁴ 

A gene ontology is shown in Table 3. TGF-β and HIPPO signaling were investigated in this study to assess the function of selected miRNAs in exosomes. The functions of individual miRNAs will be elucidated in future studies. 

Table 3.

 

Top 10 Gene Ontology (GO) Differentially Expressed miRNA Target Genes

 

In conclusion, our findings demonstrate that ASC-derived exosomes have the potential to improve the wound healing and migration of CECs by inducing a shift in the cell cycle and inhibiting senescence and autophagy pathways. Hence, these exosomes may serve as a promising therapeutic approach for the treatment of CEC diseases. 

Acknowledgments

The authors thank Junho Kim, PhD, and Ga Eun You, MS (Research and Development Institute, Biosolution Co., Ltd.) for providing the ASC-derived exosomes. 

Supported by Hallym University Medical Center Research Fund and by a grant from the National Research Foundation (NRF-2023R1A2C2002674) funded by the Korean government. 

Disclosure: Y. Ryu, None; J.S. Hwang, None; K. Bo Noh, None; S.H. Park, None; J.H. Seo, None; Y.J. Shin, None