Proteomic Profiling of Exosomes Derived from Endometrial Stem Cells and Adipose-Derived Stem Cells.

Stem cells have differentiation and self-renewal abilities. The two main types are embryonic stem cells, which are pluripotent, and adult stem cells, which are multipotent. Mesenchymal stem cells (MSCs) are a type of adult stem cell that can differentiate into multiple cell types and produce growth factors and cytokines. MSCs are found in various tissues, including bone marrow, fat, and liver, and play a role in tissue repair and regeneration. Adipose-derived stem cells (ASCs) are a type of MSC that can be obtained from liposuction tissue and have therapeutic effects on wound healing and immune regulation ( 1 ). Human endometrial tissue is another excellent source of MSCs. The endometrium repeatedly proliferates and sheds, providing stem cells. In 2004, stromal stem cells were first identified in the human endometrium ( 2 ), and endometrial regenerative cells were characterized as a novel stem cell source due to their ease of extraction and pluripotency ( 3 ). Endometrial stem cells (EnSCs) show high viability even after aging, and their isolation from menstrual blood is non-invasive and avoids ethical issues. MSCs are used in regenerative medicine for their trophic effects and potential for structural integration in damaged tissues. Investigating the secretion profiles of growth factors and cytokines from MSCs and exosomes are important ( 4 ), as MSC-derived exosomes are key mediators of paracrine signaling ( 5 ). Studies have shown that MSC-derived exosomes can enhance wound healing and have therapeutic potential ( 6 - 8 ). A recent study investigated the use of EnSC-derived exosomes isolated from healthy menstrual blood to treat endometriosis by reducing inflammation, proliferation, migration, and angiogenesis in endometriotic cells. After 72 hours of exosome treatment, various assays demonstrated decreased expression of markers associated with inflammation, migration, and angiogenesis, accompanied by increased apoptosis in endometriotic cells ( 9 ). This study aimed to compare the benefits of EnSCs and ASCs for clinical therapy in regenerative medicine. We demonstrated that EnSCs, easily obtainable from adult tissue, have potential advantages over ASCs. Our results highlight the treatment potential of both cell types through exosome protein profiling.

Isolated EnSCs and ASCs formed colonies with fibroblast-like morphology upon adhesion ( Fig. 1A , 1B ). The isolated EnSCs and ASCs had stem cells characteristics. To confirm one of the characteristics of stem cells, their multi-differentiation ability was evaluated. Both EnSCs and ASCs were effectively differentiated into adipocytes, osteocytes, and chondrocytes ( Fig. 1C ). Phenotypic analyses of EnSCs and ASCs were performed using flow cytometry. After the isolated cells were confirmed to have the characteristics of MSCs, these MSCs were negative for CD34/CD45, and >95% positive for CD29, CD73, CD90, CD105, CD140b ( Fig. 1D ). In particular, EnSCs were positive for CD146 and SUSD2 ( Fig. 1D ). The EnSCs and ASCs were cultured under identical conditions. Initial proliferation rates were similar, but EnSCs significantly increased (>1.5-fold, p<0.05) from day 5 ( Fig. 1E ). Karyotype stability was maintained on continuous EnSC and ACSs cultures until passage 11 ( Fig. 1F ). For proteomic analysis, exosomal proteins from EnSCs and ASCs were isolated from each MSC culture medium ( Fig. 2A ). Nanoparticle tracking analysis (NTA) was performed to determine the size and quantify the isolated exosomes. The NTA results showed that the mode sizes of EnSC-ex and ASC-ex were 164.1±6.6 nm and 188.2±9.6 nm, respectively ( Fig. 2B , left). To verify whether the isolated particles were exosomes, western blotting was performed to detect tetraspanin proteins CD9 and CD81, which are well-known exosomal markers. Both CD9 and CD81 were expressed in EnSC-ex and ASC-ex ( Fig. 2B , right). To elucidate significant differences in protein composition between EnSC-ex and ASC-ex, we employed a high-throughput quantitative proteomic approach using label-free differential analysis ( Fig. 2C ). The detected protein list was generated by scoring analyzed peptide results aligned with the search engine. Peptide identification was performed using a developed program, where a positive score indicated the presence of a detected protein in the sample (see Materials and Methods section). The relative quantification of the identified proteins was expressed as an abundance value, calculated based on the area of distribution of the specific protein peptide ionization value in each sample. A total of 1,218 proteins were identified in EnSC- and ASC-derived exosome protein samples ( Fig. 3A ). Among these, 320 proteins showed significant differences between EnSC-ex and ASC-ex, as determined by an adjusted p-value of <0.01 ( Supplementary Table S1 ). Specifically, 229 proteins were uniquely detected in EnSC-ex, while 51 proteins were uniquely detected in ASC-ex ( Fig. 3A ). Additionally, 40 proteins found in both EnSC-ex and ASC-ex exhibited significant differences (p<0.001), with 7 showing higher abundance in EnSC-ex and 33 showing higher abundance in ASC-ex ( Fig. 3A ). Comparative GO analysis was performed at the proteomic level to analyze the enriched proteins in EnSC-ex and ASC-ex (p<0.001, FDR<0.05). For EnSC-ex, 236 enriched proteins were identified, and their top ten enriched GO terms are displayed in a bar chart ( Fig. 3B , left). For ASC-ex, 84 enriched proteins were identified, and their top ten enriched GO terms are displayed in a bar chart ( Fig. 3B , right). KEGG pathway analysis of proteins with higher abundances in EnSC-ex compared to ASC-ex (indicated by red bars) revealed significant associations with the following pathways: African trypanosomiasis (hsa05143), malaria (hsa05144), axon guidance (hsa04360), proteoglycans in cancer (hsa05205), and human papillomavirus infection (hsa05165) ( Fig. 3C ). Similarly, KEGG pathway analysis of proteins with higher abundances in ASC-ex compared to EnSC-ex (indicated by blue bars) showed significant associations with the following pathways: systemic lupus erythematosus (hsa05322), alcoholism (hsa05034), neutrophil extracellular trap formation (hsa04613), and viral carcinogenesis (hsa05203) ( Fig. 3C ). Table 1 provides a summary of these KEGG pathways along with the corresponding protein lists. Specifically, proteins that were more abundant in EnSC-ex and related to axon guidance were selected for protein-protein interaction network analysis using STRING ( Fig. 3D ).

In this study, we characterized EnSCs and compared them with ASCs, which are relatively easy to obtain and commonly used in clinical practice. Our findings demonstrate, for the first time, the superior properties of EnSCs in several aspects. EnSCs exhibit a higher proliferation rate than ASCs and possess enhanced osteogenic differentiation potential. Additionally, the CD146 marker protein, primarily expressed in vascular endothelial cells and pericytes, suggests the potential of EnSCs in treating fertility-related diseases, promoting bone formation, and supporting vascular regeneration. EnSCs, first identified and studied in 2007, are MSCs derived from menstrual blood ( 3 ). Liu et al. ( 11 ) reported they exhibit low tumorigenicity and minimal adverse effects in mice. To further investigate EnSCs, we isolated exosomes from EnSCs and ASCs and analyzed their protein composition using LC-MS/MS. Notably, stromal-derived factor 1 (SDF-1), exclusively detected in EnSC-derived exosomes, mediates MSC recruitment and tissue repair via the SDF-1/CXCR4 axis in organs including liver, brain, and skeletal muscles ( 12 ). Interestingly, our analysis also revealed the presence of both angiogenesis-inhibiting and angiogenesis-promoting proteins within EnSC-derived exosomes. Specifically, insulin-like growth factor-binding protein 5 (IGFBP5), a known IGF-1 pathway inhibitor, is linked to reduced VEGF activity, angiogenesis inhibition, and tumor suppression, particularly in ovarian cancer ( 13 ). Conversely, IGFBP7, another member of the IGFBP family, promotes angiogenesis and plays critical roles in decidualization and embryo implantation. Its dysregulation is associated with impaired uterine receptivity and abnormal angiogenic conditions, such as endometriosis and pathological pregnancies ( 14 ). This duality reflects adaptation of EnSCs to cyclical endometrial regeneration. We hypothesize that the balanced secretion of these opposing factors enables tight modulation of angiogenesis, thus ensuring efficient regeneration while preventing excessive or pathological angiogenesis. Furthermore, GO-based functional annotation demonstrated that EnSC-ex and ASC-ex differ significantly in their exosomal protein composition, suggesting distinct functional specializations. EnSC-derived exosomes showed enrichment of TAP binding protein (TAPBP, also known as tapasin), a crucial component involved in antigen processing and MHC class I presentation. This indicates that EnSC-derived exosomes might enhance immune surveillance and promote cytotoxic T cell activation, facilitating immunomodulation and effective tissue repair in inflammatory or hormonally responsive contexts ( 15 ). In contrast, ASC-derived exosomes exhibited substantial enrichment of proteins involved in chromatin remodeling, reflecting a potential role in regulating adipogenesis and modulating inflammation within adipose tissue ( 16 ). This enrichment suggests ASC-exosomes specifically support adipose tissue homeostasis and inflammation management. Recent studies demonstrated EnSC-ex therapeutic benefits in neural regeneration and ovarian function ( 17 , 18 ). Our proteomic analysis identified several specific proteins enriched in EnSC-ex that potentially contribute to these therapeutic outcomes. Notably, proteins involved in axon guidance pathways, such as NRAS, EFNB1, RAC2, WNT5B, and SRGAP2, were significantly elevated, suggesting possible roles in neurite outgrowth, cytoskeletal remodeling, and neural connectivity via signaling pathways such as MAPK/ERK, ephrin-Eph, Rho GTPase, and non-canonical Wnt ( 19 ). These molecular profiles align closely with functional studies demonstrating that EnSC-ex enhance neurite outgrowth, proliferation, and migration through activation of the PI3K/AKT pathway ( 17 ). Additionally, the enrichment of TAPBP (tapasin) in EnSC-ex indicates their immunomodulatory potential, which could provide advantages in managing inflammatory and immune-mediated conditions compared to ASC-ex. Collectively, the distinct protein signatures identified in EnSC-ex, coupled with their non-invasive sourcing, rapid proliferation, and multipotency, highlight their significant translational potential for specific clinical applications in regenerative medicine, notably in neurological and gynecological therapies. Although further clinical research is required, EnSCs hold valuable potential for regenerative medicine, comparable to existing cell therapy approaches ( 20 ). There is also potential in using EnSCs in combination with other cell therapies or EnSC-derived exosomes to enhance therapeutic effects and support tissue repair and regeneration. This study presents promising in vitro findings on EnSCs. However, their therapeutic potential for regeneration and immunomodulation requires further validation in animal models and clinical trials to confirm their efficacy and safety. Additionally, expanding the comparison to include other MSC sources, such as bone marrow or umbilical cord-derived MSCs, would provide a broader perspective on the unique advantages of EnSCs in regenerative therapy. Addressing these limitations will validate the therapeutic potential of EnSCs and clarify clinical applications.

Human endometrial and adipose tissues were collected with informed consent. This study was approved by the Public Institutional Bioethics Committee of the Korea National Institute for Bioethics Policy (P01-202011-31-004). Endometrial samples were obtained from menstrual blood collected during the early proliferative phase (days 2∼3 following the onset of menstruation). Samples included donors negative for endometriosis, autoimmune diseases, infections, mycoplasma, and common viruses. Women with recent vaginal infection, contraceptive, or endometrial-affecting medication were excluded. Adipose tissue was harvested from abdominal subcutaneous fat using a liposuction cannula. Human MSCs were isolated from the desquamated endometrium of menstrual blood from healthy donors. Endometrial tissue was minced and digested in DMEM/F12 (HyClone) supplemented with 10% fetal bovine serum (FBS), 1% penicillin-streptomycin (Sigma-Aldrich), and 1 mg/mL collagenase A (Gibco) for 60 minutes at 37℃ with agitation. Cells were filtered (70 μm sieve) to remove glandular epithelial components, centrifuged at 200 g for 10 minutes. The supernatant was resuspended in complete medium (DMEM/F12 with 10% FBS and 1% penicillin-streptomycin) and plated in plastic flasks at 37℃ in a humidified chamber (5% CO 2 ) for adhesion. After 24 hours, the medium was refreshed every other day for two weeks. The morphology of EnSC cultures was observed using an inverted light microscope (Olympus). ASCs were isolated using an ultrasonication device (Ultrastem; MediFutures). Cell proliferation was determined using CCK-8 (Dojindo Biotech). EnSCs and ASCs were seeded at 2,000 cells/well in 96-well plates. After 1 day, 10 μL of CCK-8 solution was added to each well and incubated for 2 hours at 37℃. Absorbance was measured at 450 nm using a universal microplate reader (Multiskan SkyHigh Microplate Spectrophotometer; Thermo Fisher Scientific). The CCK-8 assay quantifies cell viability based on the production of a formazan dye, which is proportional to the number of living cells. Cell surface markers specific for stem cells, including CD29, CD34, CD45, CD73, CD90, CD105, CD140b, CD146, and SUSD2, were assessed using flow cytometry. Cells (passages 3∼5) were resuspended in phosphate buffered saline (PBS) (2×10 5 cells/tube) The cells were incubated with the following antibodies for 20 minutes at room temperature in the dark: mouse IgG 1 kappa isotype-PE, human CD73-PE, CD90-PE, CD105-PE, CD45-PE, CD34-PE, CD29-PE, CD140b-PE, CD146-PE, and SUSD2-APC (Thermo Fisher Scientific). After incubation, cells were washed with PBS, centrifuged at 400 g for 5 minutes, and resuspended for analysis on a BD Canto II flow cytometer. Non-specific binding was identified using IgG 1 isotype-PE control. In this analysis, key gating strategies were applied to define specific cell populations based on target markers, with manual gating used to eliminate debris and doublets. Data were analyzed using FlowJo software (BD Biosciences). Adipogenesis differentiation and AdipoRed assay: EnSCs were seeded at 1×10 4 cells/well in a 48-well plate with 400 μL media per well. Upon reaching 90% confluence, the medium was replaced with adipogenic medium (A1007001; Gibco), and cells were cultured for 11 days with medium changes every 3∼4 days. Adipogenic capacity was evaluated using AdipoRed staining. Cells were fixed with 4% paraformaldehyde for 30 minutes, washed with PBS three times, and incubated with AdipoRed (PT7009; Lonza) stain for 10 minutes. After PBS washes, cells were observed by fluorescence microscopy. AdipoRed-positive cells contained cytoplasmic lipid. Osteogenic differentiation and Alizarin Red staining: EnSCs were seeded at 1×10 4 /cm 2 and cultured for 21 days in osteogenic differentiation medium (A1007201; Gibco). Cells in complete medium served as control. The medium was changed every three days. After 21 days, cells were stained with Alizarin Red to assess osteogenesis. Cells were PBS-washed and fixed with 4% paraformaldehyde for 30 minutes. After washing with distilled water, cells were stained with 2% Alizarin Red S (pH 4.2) for 2 minutes, then washed. Stained cells were observed under an inverted microscope (CKX53; Olympus). Chondrogenesis differentiation and Alcian blue staining: To induce chondrogenic differentiation, cell suspensions containing 3×10 5 cells per 15 mL polypropylene tube were centrifuged for 6 minutes at 350 g to form a pellet. Cell pellets were cultured 21 days in chondrogenic medium (37℃, 5% CO 2 ) (DMEM/F12 with 10% FBS, 1% ITS＋Premix [345352; Corning], 50 μg/mL ascorbic acid 2-phosphate [A8960; Sigma-Aldrich], 40 μg/mL L-proline [P5607; Sigma-Aldrich], 100 μg/mL sodium pyruvate [S8636; Sigma-Aldrich], 100 nM dexamethasone [D4902; Sigma-Aldrich], and 10 ng/mL TGF-b3 [243-B3-200; R&D system]). Medium was replaced every three days. Pellets were cryosectioned and evaluated by Alcian blue staining. For EnSC-exosome (EnSC-ex) and ASC-exosome (ASC-ex) isolation, the culture media were replaced with DMEM/F12 containing 1% penicillin-streptomycin after rinsing twice with PBS. After 67 hours, supernatants were collected, and exosomes were isolated using a previously optimized protocol ( 10 ). Supernatants were centrifuged at 300 g for 10 minutes and 2,000 g for 20 minutes at 4℃ to remove apoptotic bodies and cell debris, followed by 10,000 g for 30 minutes to remove microvesicles. Finally, supernatants were ultracentrifuged at 100,000 g for 60 minutes twice, and the pellets were collected and dissolved in PBS. All procedures were performed at 4℃. Exosome size/concentration were analyzed by NanoSight NS300 (Malvern Panalytical). For exosome western blotting, samples were normalized by seeded cell number and verified using CD9 (ab223052, Abcam) and CD81 (56039, Cell Signaling Technology). Fractionated peptide samples were analyzed using an LC-MS system consisting of an Easy-nLC 3000 coupled to an EASY-spray ion source on an Orbitrap Exploris 240 mass spectrometer (Thermo Fisher Scientific). Peptides were separated on a two-column setup with a trap column (100 μm×2 cm, nanoViper C18, 5 μm, 100 A Acclaim PepMap 100) and an analytical column (75 μm I.D.×50 cm, 3 μm C18 beads). Separation was achieved with a 180 minutes linear gradient from 3% to 30% solvent B (100% ACN and 0.1% formic acid). The spray voltage was 2.2 kV in positive ion mode, with a capillary temperature of 275℃. Mass spectra were acquired using the top 20 method on the Orbitrap Exploris 240, with data collected via Xcaliber software version 4.4. The Orbitrap analyzer scanned precursor ions (350∼1,800 m/z) at a resolution of 60,000 at m/z 200. The automatic gain control (AGC) target value was 3×10 6 , with an isolation window of 1.4 m/z. Higher-energy collisional dissociation scans were performed at a resolution of 15,000 and a normalized collision energy of 30. The AGC target value for MS/MS was 7.5×10 5 , with a maximum ion injection time of 22 ms. Dynamic exclusion was set for 15 seconds. Mass data were acquired using Proteome Discoverer 2.5 (Thermo Fisher Scientific). Protein identification was performed with Sequest-HT within the Proteome Discoverer platform using label-free differential analysis. To minimize false positives, the false discovery rate (FDR) for peptide-spectrum matches was set to 0.5%, and only peptides with at least seven amino acids found in three biological replicates were retained. Relative abundance changes were calculated for each peptide by comparing peak areas from the extracted ion chromatograms, expressed as protein quantification relative to EnSC-ex and ASC-ex. Gene ontology (GO) annotation was performed using UniProt, covering biological processes, molecular functions, and cellular components. Enrichment analysis was conducted using the DAVID bioinformatics tool with Benjamini-Hochberg FDR for multiple test corrections (FDR<0.05). Significant GO terms identified by Fisher’s exact test (Benjamini-Hochberg corrected, p<0.05). Gene symbols were analyzed using STRING networks in Cytoscape (version 3.9.1). The results are presented as mean and standard deviation. Statistical significance was tested via Student’s t-test (SPSS v22.0; IBM Corp.).The Benjamini–Hochberg statistic was used to identify statistically significant values to avoid Type 1 errors (false positives). Differences were analyzed by ANOVA using SPSS, and statistical significance was set at p<0.05.

Supplementary data including one table can be found with this article online at https://doi.org/10.15283/ijsc25031