Primary culture and identification of hUC-MSCs

The human umbilical cord (UC) was collected from healthy full-term fetuses that were born via cesarean delivery at Qilu Hospital of Shandong University. The use of the UC was approved by the Ethics Committee of the Qilu Hospital of Shandong University (KYLL-2021(KS)-086). Briefly, the UC was cut into small fragments (1–2 mm3) after the umbilical arteries and vein were dissected and placed in dishes with fresh MSC complete medium containing 90% α-MEM (Macgene Technology Ltd., Beijing, China), 10% fetal bovine serum (FBS; Bovogen Biologicals Pty Ltd, Keilor East, Victoria, Australia) and 1% penicillin and streptomycin (P/S; C125C5, New Cell & Molecular Biotech, Suzhou, China) in 5% CO2 at 37 ℃. When the cells reached 80%–90% confluence, they were trypsinized with 0.25% trypsin (C125C1, New Cell & Molecular Biotech) and passaged at a 1: 3 ratio. hUC-MSCs at passage three in the logarithmic growth period were collected to perform tri-lineage differentiation and phenotypic identification. The adipogenic, osteogenic, and chondrogenic abilities of the hUC-MSCs were detected after using the corresponding induction solution. After 4 weeks of induction, Oil Red O, Alizarin Red S, and Alcian Blue staining were used to detect adipogenic, osteogenic, and chondrogenic differentiation, respectively. In addition, hUC-MSCs were stained with antibodies against the human antigens CD29-PE (TS2/16, 303,004), CD31-PE (WM59, 303,106), CD44-PE (IM7, 25–0441-82), CD45-PE (2D1, 368,510), CD73-PE (AD2, 344,004), CD90-PE (eBio5E10,12–0909-42), CD105-PE (SN6, 12–1057-42), and CD271-PE (ME20.4, 12–9400-42). CD29, CD31, CD45, and CD73 antibodies were purchased from BioLegend (San Diego, CA, USA) and CD44, CD90, CD105, and CD271 antibodies were purchased from Invitrogen (Thermo Fisher Scientific, Inc.). The antibody dilution ratio was 1:20. The cells were then analyzed by flow cytometry (Guava easyCyte 6HT, EMD Millipore, Billerica, MA, USA), and the data were examined using FlowJo (version 10, Tree Star Inc., Ashland, OR, USA).

Three-dimensional culture of hUC-MSCs

hUC-MSCs (passages 3–6) were cultured in T75 cm2 flasks. For 3D culture, 5 × 106 hUC-MSCs (passages 3–6) were seeded in ultra-low attachment dishes (633,180, Greiner, Frickenhausen, Germany). 3D-cultured hUC-MSCs were stained with PKH26 (MIDI26, Sigma-Aldrich, USA) and 4′,6-diamidino-2-phenylindole (DAPI, ab104139, Abcam) and observed under a fluorescence microscope (BX53; Olympus, Tokyo, Japan).

Cell proliferation assay

Cell proliferation was determined using carboxyfluorescein diacetate succinimidyl ester for both 2D- and 3D-cultured cells (CFSE; BD Biosciences, San Jose, CA, USA). Cells were incubated for 5 min with 10 μM of CFSE in phosphate-buffered saline (PBS) at 37 ℃. All assays were performed using flow cytometry. Cell proliferation analyses were performed using ModFit LT software (version 3.2; Verity Software House, Topsham, ME, USA).

Quantitative real-time polymerase chain reaction (qRT-PCR) assay

Total RNA from 2D- and 3D-cultured cells was extracted using an RNA fast kit (Fastagen, Shanghai, China). Total RNA was reverse-transcribed using the ReverTra Ace qPCR RT Kit (Toyobo, Osaka, Japan). qRT-PCR was performed to detect the transcript levels of Oct4, Nanog, Igf, Sox2, Hgf, Tgfb1, Sdf, Vegf, Ido1, Il6 and Pge2 using an SYBR Green real-time PCR master mix (Toyobo, Osaka, Japan). Relative expression levels were normalized to those of GAPDH. The value for 2D-cultured cells was set to 1. Primer sequences used are listed in Supplementary Material 1.

Isolation and purification hUC-MSC-derived EVs

hUC-MSCs were cultured in flasks and ultra-low attachment dishes with EVs-depleted FBS. EVs-depleted FBS was prepared by separating EVs from FBS using ultracentrifugation at 100,000 × g for 18 h at 4 ℃ (Thermo Fisher Scientific, Sorvall Lynx 6000, Osterode, Germany) [33]. After 48 h, cell culture supernatants were collected to isolate the EVs. The 2D and 3D cell culture supernatants were centrifuged at 300 × g for 10 min, 2,000 × g for 20 min, and 10,000 × g for 30 min at 4 ℃ to eliminate the dead cells and cell debris. The supernatants were filtered using a 0.22-μm filter (Millipore). Then, the supernatants were centrifuged at 100,000 × g for 70 min, resuspended in PBS, and centrifuged again at 100,000 × g for 70 min at 4 ℃. Finally, EVs purified from the media of 2D- and 3D-cultured cells (2D-EVs and 3D-EVs, respectively) were resuspended in 100–200 μL of PBS and stored at -80 ℃.

Transmission electron microscopy (TEM)

The morphology of the EVs was assessed by TEM, which was conducted by Shanghai XP Biomed Co., Ltd. (Shanghai, China). Droplets were deposited onto a TEM copper grid for more than 1 min. The grid was then negatively stained with phosphotungstic acid for 5 min, blotted with filter paper, and dried at room temperature. Finally, the EVs were observed under a transmission electron microscope (JEM-1200EX, JEOL, Tokyo, Japan) and photographed.

Nanoparticle tracking analysis (NTA)

The size distribution and concentration of EVs were determined using NTA. The EV nanoparticle tracking analysis was conducted by Shanghai XP Biomed Co., Ltd. (Shanghai, China). Briefly, EVs were diluted in PBS and injected into the ZetaView particle tracker (ZetaView PMX 110, Particle Metrix, Meerbusch, Germany). NTA measurements were recorded and analyzed at 11 positions. ZetaView software (version 8.04.02 SP2) was used to analyze the data. We used the bicinchoninic acid assay (BCA) method to detect the protein concentration of EVs and combined it with NTA analysis to determine that there were 109 particles per μg of protein and 5 × 1011 particles per 0.5 mg of protein.

Preparation of ovarian ECM, lyophilized EVs, and gels

Five-week-old female Sprague–Dawley rats (Sibeifu Beijing Biotechnology Co. Ltd., Beijing, China) were euthanized for ovary collection. The ovaries were decellularized with 0.2% sodium dodecyl sulfate (SDS, Solarbio, Beijing, China) for 12 h at 18–20 ℃ on a horizontal shaker at 50 rpm. Then, the decellularized ovaries were washed in PBS eight times for 2 h. A Masson’s trichrome stain kit (G1340, Solarbio) was used to confirm that the ovaries were completely decellularized. The ovarian ECM was lyophilized using a Scientz − 12N (Ningbo, China). The lyophilized ovarian ECM was digested with pepsin (Solarbio) for 72 h at a final concentration of 10 mg/mL to obtain ovarian ECM solutions. A gel consisting of chitosan (419,419, Sigma-Aldrich, USA)/β-sodium glycerophosphate (Merck, Darmstadt, Germany)/gelatin (Amresco) loaded with ovarian ECM solutions was obtained as described by Xu et al. [31]. Then, 2D-EVs and 3D-EVs were lyophilized and mixed with the ECM gel at 0.5 mg/mL. The viscosity of the gel was measured using an NDJ-9S viscometer (Shanghai Performance Tai Electronic Technology Co., Ltd, Shanghai, China) at 60 rpm [34].

Isolation and characterization of rat ovarian granulosa cells (OGCs)

Rat OGCs were isolated from four-week-old female Sprague–Dawley rats as described previously [35]. The rat OGCs were cultured in DMEM/F12 (Macgene) medium with 10% FBS and 1% P/S at 37 ℃, 5% CO2. Immunofluorescence staining was used to identify the follicle-stimulating hormone receptor (FSHR, GB11275-1, Servicebio, Wuhan, China) specifically expressed in granulosa cells. The nuclei were stained with DAPI (ab104139, Abcam).

Labeling of EVs and cellular uptake in vitro

2D- and 3D-EVs were labeled with PKH26 (Sigma) according to the manufacturer’s instructions. These EVs were then co-cultured with OGCs for 24 h (the final concentration of EVs was 50 μg/mL). The nuclei of the OGCs were stained with DAPI (Abcam). The fluorescence signals of PKH26 and DAPI were observed under a fluorescence microscope.

Cell counting kit-8 assay (CCK-8) and cytocompatibility assay in vitro

Rat OGCs were cultured in 96-well plates to reach 70–80% confluence. Then, rat OGCs were divided into six groups: Control, ECM gel, CTX, CTX + ECM gel, CTX + 2D-MSC-EVs-ECM gel, and CTX + 3D-MSC-EVs-ECM gel groups (gel: media = 1:10; final concentration of CTX (PHR1404, Sigma-Aldrich, USA), 6 mg/mL; final concentration of EVs, 10 μg/well). After 48 h, cell viability was detected using a CCK-8 assay (C6005, New Cell & Molecular Biotech). Absorbance was measured with an enzyme-labeled analyzer (DNM-9602, Beijing Perlong New Technology Co., Ltd.) at a wavelength of 450 nm.

Establishment of COF rat model and therapeutic experiments

Five-week-old female Sprague–Dawley rats were obtained from Sibeifu Beijing Biotechnology Co., Ltd., Beijing, China. The use of rats was approved by the Ethics Committee of the Qilu Hospital of Shandong university (DWLL-2021–147). The rats were maintained at 22 ± 2 ℃ under a 12-h light/dark cycle, with free access to water and food. The rats were randomly divided into six groups (5–7 rats per group): Control, ECM gel, CTX, CTX + ECM gel, CTX + 2D-MSC-EVs-ECM gel, and CTX + 3D-MSC-EVs-ECM gel group. The Control group received no treatment. The first day of the CTX injection was designated as day 1. The CTX, CTX + ECM gel, CTX + 2D-MSC-EVs-ECM gel and CTX + 3D-MSC-EVs-ECM gel groups received intraperitoneal injections of CTX (50 mg/kg on day 1 and 8 mg/kg on days 2–14). The CTX group did not receive any treatment after 14 days. The ECM group received an intra-ovarian injection of ECM gel on day 15. The CTX + ECM gel, CTX + 2D-MSC-EVs-ECM gel, and CTX + 3D-MSC-EVs-ECM gel groups received intra-ovarian injections of ECM gel, 2D-MSC-EVs-ECM gel, and 3D-MSC-EVs-ECM gel, respectively, on day 15. Each ovary was injected with 25 μL of gel containing 0.5 mg/mL of EVs. All the rats were weighed daily. All vaginal smears were collected daily and stained with hematoxylin and eosin (H&E) to determine the estrous cycle phase. All rats were euthanized on day 30 for further analysis.

Enzyme-linked immunosorbent assay (ELISA)

Blood samples were obtained by cardiac puncture. Serum was collected by centrifuging at 1,200 × g for 15 min at 4 ℃, and it was stored at –20 ℃ for ELISA analysis. Serum levels of FSH (Cloud-clone Corp, CEA830Ra), estradiol (E2, Cloud-clone Corp, CEA461Ge), and anti-Mullerian hormone (AMH, Cloud-clone Corp, CEA228Ra) were detected using an ELISA kit according to the manufacturer’s instructions.

H&E staining

Rat ovaries were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned into 5-μm sections. Five discontinuous sections of each ovary were selected for follicle counting and morphological observation. The ovarian sections were dewaxed and rehydrated for H&E staining.

RNA sequencing

Ovary tissues (Control, CTX, CTX + 2D-MSC-EVs-ECM gel, and CTX + 3D-MSC-EVs-ECM gel groups) were chosen for RNA sequencing. Each group had one biological replicate. RNA sequencing and bioinformatics analyses were performed at Shandong Xiuyue Biotechnology Co. Ltd.

Western blotting

The protein lysates from exosomes or ovary tissues were prepared following standard protocols, and the protein content was determined using a BCA protein assay kit (P0010S, Beyotime, Shanghai, China). Equal amounts of proteins were electrophoresed on 7.5–15% SDS-PAGE gels and transferred onto polyvinylidene difluoride membranes (ISEQ00010, Merck Millipore, Darmstadt, Germany). Membranes were blocked for 2 h in blocking buffer composed of 5% non-fat dry milk in tris-buffered saline containing 0.1% Tween 20 at room temperature and were incubated with primary antibodies overnight at 4 ℃. The polyvinylidene difluoride membranes were then washed three times for 10 min each and incubated with secondary antibodies for 1 h at room temperature. A chemiluminescence detection system (Shanghai Jiapeng Technology Co., Ltd., Shanghai, China) was used to visualize the protein bands. Protein expression was expressed as relative abundance normalized to that of GAPDH (1:1000, 5174, Cell Signaling Technology). The primary antibodies used were anti-Alix (1:1000, 18269S, Cell Signaling Technology), anti-CD63 (1:1000, 25,682–1-AP, Proteintech), anti-CD9 (1:1000, 13174S, Cell Signaling Technology), anti-CD81 (1:1000, 52892S, Cell Signaling Technology), anti-TSG101 (1:1000, ab125011, Abcam, Cambridge, UK), anti-calnexin (1:1000, 2433S, Cell Signaling Technology), and anti-Bax (1:1000, 2772S, Cell Signaling Technology). The secondary antibody used was Goat Anti-Rabbit IgG (H + L) HRP (1:5000, AB0101, Abways).

Statistical analysis

All experiments were performed in triplicate. All data were analyzed using GraphPad Prism version 9 (GraphPad, San Diego, CA, USA). Comparisons between two groups or multiple groups were analyzed using Student’s t-test or one-way ANOVA. Quantitative data were expressed as mean ± standard deviation (SD). p < 0.05 was considered statistically significant.