CEFFE was provided by SEME Cell Technology Co., Ltd (Shanghai, China), and it was extracted from fresh adipose tissues as previously described (24). Briefly, the fat tissue was washed with saline several times to discard the blood components. Next, the tissue was emulsified mechanically, centrifuged several times to obtain the lowest fluid layer, and stored at –80°C. The protein concentration of CEFFE was measured using a BCA protein kit (Vazyme, China), and the solution was diluted to 3 mg/mL before use.

Specific pathogen free 10-month-old female mice and 8-week-old female and male mice of the C5BL/6J strain were purchased from and housed in Shanghai Branch of Beijing Vital River Laboratory Animal Technologies Co. Ltd., China. All animal experiments were conducted according to the guidelines of Animal Ethics Committee (P2022002). The Control group comprised 8-week-old female mice, and the 10-month-old female mice were equally divided into two groups: the Aged and Aged +CF groups. The mice from the Aged +CF group were treated with 200 μL CEFFE (protein concentration: 3 μg/μL) via the tail vein every two days for 2 weeks. At the same time, mice from the Control and Aged groups were injected with the same volume of saline.

Blood samples from mice were collected and stored at room temperature for 30 min, and they were then centrifuged at 2000 rpm for 10 min. The upper supernatant was harvested to detect the serum concentration of AMH, E₂, and FSH via ELISA (Mlbio, China) as previously described (27).

Ovaries were harvested and fixed with 4% paraformaldehyde for more than 24 h, and dehydration, wax leaching, and embedding were performed. Ovaries were sliced into 4 μm sections. Subsequently, the sections were dewaxed with ethanol and stained with Hematoxylin and Eosin dye. The maximum longitudinal section of each ovary was selected to count all follicular stages based on the morphological characteristics (28). Finally, the number of primordial, primary, secondary, antral, and atretic follicles were recorded and analyzed.

To evaluate the function of CEFFE on embryonic development in aged mice, mice were treated with ovulation induction after two weeks from the last treatment. Briefly, the mice were intraperitoneally injected with 10 IU of pregnant mare serum gonadotropin (Sansheng, China), and after 48 h, 10 IU of human chorionic gonadotropin (Sansheng, China) was injected. Subsequently, female mice were mated with sexually matured male mice in a 2:1 ratio. After 18 h, female mice were sacrificed, and oocytes were collected from the ampulla. Embryo culture was performed as previously described (29). The numbers of total oocytes, 2-cell embryo rate (number of 2-cell embryos/number of total oocytes retrieved), 4-8 cell embryo rate (4-8 cell embryo number/2-cell oocytes retrieved number), morula rate (morula number/2-cell oocytes retrieved number), and blastocyst rate (blastocyst number/2-cell oocytes retrieved number) were recorded.

To assess the efficiency of CEFFE involved in the improvement of fertility, reproductive tests were conducted. Two weeks after the last treatment, female mice from different groups were mated with sexually mature male mice in a 2:1 ratio for 2 months, and the fertility information was recorded. Following this, male and female mice were separated for 21 days to confirm the final pregnancy status. During the process, the interval time (the time from mating to birth) and number of offsprings were recorded (30).

Ovarian sections from mice were dewaxed, retrieved for antigen, blocked with serum, and incubated in the presence of primary antibodies at 4°C overnight. Subsequently, the slides were incubated with secondary antibodies at room temperature for 1 h. Finally, the target was colored with DAB chromogenic reaction, and the nucleus was counterstained with hematoxylin (Servicebio, China). Images were captured using an optical microscope, and they were analyzed using Image J software. The detail information of antibodies is listed in Supplementary Table 1 .

The ovaries were collected and fixed in the fixative for TEM (Servicebio, China) in the dark at 4 °C. Following this, the ovaries were dehydrated at room temperature with gradient concentrations of ethanol. Next, the ovaries were penetrated and embedded with resin, and the samples were moved into a preheated 65 °C oven for more than 48 h for polymerization. The resin blocks were cut into 80 nm-thin sections and fished out using 150 meshes. Following this, 2% uranium acetate saturated alcohol solution and 2.6% lead citrate (Servicebio, China) were used to stain the section. Finally, the ovary section was observed under TEM (Hitachi, Japan) and images were captured.

Cells were fixed with 4% PFA (Servicebio, China) at room temperature for 15 min, permeabilized with 0.1% Triton X-100 for 10 min, and blocked with 5% BSA in PBS for 30 min. Next, the cells were incubated with different primary antibodies at 4 °C overnight. On the second day, the cells were washed with PBST (PBS+0.1% Tween-20) and incubated with Alexa Fluor 488-conjugated rabbit secondary antibody. Subsequently, the nucleus was labeled with Hoechst 33342 for 15 min. Finally, the samples were photographed using a Zeiss-Axio Vert.A1 fluorescence microscope. The detail information of antibodies is listed in Supplementary Table 1 .

The ovaries were collected from mice, washed with PBS, immediately frozen in liquid nitrogen, and stored at -80 °C for RNA-seq and RT-PCR. RNA-seq was supported by the Shanghai Silver Crown Biomedical Technology Co., Ltd. Briefly, the total RNA of ovaries was extracted using miRNeasy Mini Kit (QIAGEN, USA), and the concentration and integrity of RNA were analyzed via NanoDrop and Agilent 4200 TapeStation (Agilent Technologies, USA), respectively. Only the RNAs with RNA integrity values between 8 and 10 and purity between 1.8 and 2.1 were included in this study. The RNA-Seq library preparation and sequencing were performed using Illumina NovaSeq 6000 (Illumina, USA). Three biological replicates were used for RNA-Seq. The cutoffs of Fold Change ≥ 1.5(|log2 ratio| ≥ 0.58) and P adjust < 0.05 were applied to identify the differentially expressed genes (DEGs) between every two groups. The DEGs were further analyzed via Venn analysis and Gene Ontology (GO) & Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses.

Total RNA was extracted from different ovaries or cells using Total RNA Isolation kit (Vazyme, China). The conversion of RNA to complementary DNA was performed using Hiscript II Reverse Transcriptase (Vazyme, China) according to manufacturer’s protocol, and the relative quantitative analysis was performed via QuantStudio TM 6 Flex (Applied Biosystems, USA). The synthesized cDNA was amplified and detected using ChamQ SYBR qPCR Master Mix (Vazyme, China). A comparative threshold cycle was normalized using GAPDH, and it was analyzed using the 2^-ΔΔCT method to compare the mRNA expression of different samples. The primer sequences are listed in Supplementary Table 2 .

The procedure of primary human granulosa cell collection was approved by the Institutional Review Board of Reproductive Medicine of Ruijin Hospital. The follicular fluids were collected during oocyte retrieval from women with the age less than 35 who sought infertility treatment due to tubal obstruction or male factors with informed consent, approved by the ethics committee of Ruijin Hospital (2020104a). And our study conformed the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network guidelines. The patients with abnormal chromosome, endometriosis, polycystic ovarian syndrome and other diseases which could affect folliculogenesis were excluded. The granulosa cells were collected from follicular fluids using density gradient centrifugation with lymphocyte separation medium (Solarbio, China). Firstly, the follicular fluid was centrifuged to collect the sediment, and it was then washed with PBS once. The cells were resuspended with PBS, transferred to the lymphocyte separation medium in an equal volume ratio, and then centrifuged at 1000 g for 20 min. Next, the medium layer cells were carefully collected and washed with PBS. Subsequently, red blood cells were discarded with Red Cell Lysis Buffer (Solarbio, China), and they were digested with 1 mg/mL hyaluronidase (Yeasen, China). Finally, the cells were resuspended in complete medium, and they were cultured in a dish precoated with 0.2% gelatin at 37 °C for 1 h. KGN cell line was purchased from Feiya Biotechnology Co., Ltd., China. Both human primary ovarian granulosa cells (hGCs) and KGN cells were cultured in DMEM/F12 medium containing 10% FBS (Gibco, USA), 100 U/mL penicillin, and 100 µg/mL of streptomycin (Solarbio, China). Besides, during the culture of hGCs, recombinant human follitropin (Switzerland) was added, and both cells were maintained under 5% CO₂ at 37°C. To induce DNA damage models of KGN and hGCs in vitro. 10 μM camptothecin (CPT, MedChemExpress, U.S) was added to the cell culture medium and incubate for 2 h. Then the cell culture medium was changed and cells were further treated with/without CEFFE.

The cells were fixed with fixed liquid at room temperature for 15 min. After washing with PBS thrice, the cells were incubated with β-galactosidase staining solution at 37 °C without CO₂ overnight. On the second day, cells were observed and pictured under optical microscope, and the senescence cells showed blue-green color.

The CCK-8 assay kit (Yeasen, China) was used to observe cell growth. KGN cells were plated in 96-well plates at 24 h before treatment. At the end of the treatment period, 10 µL of CCK-8 reagent was added to each well, and it was incubated in a 5% CO₂ atmosphere at 37°C for 1 h. The optical density was then measured using a Multiskan GO instrument at 450 nm.

To analyze the effect of CEFFE on the proliferation of human primary granulosa cells, EdU assay was performed according to manufacturer’s instructions using fluorescent microscope. Green fluorescent labeled cells represented the cells that possessed proliferative activity and blue fluorescent labeled the nucleus of all cells. The number of cells labeled with different fluorescence markers was counted using Image J, and the ratio of Green/Blue was calculated.

Proteins were extracted from cells using RIPA cell lysis, and they were degenerated with 5× protein loading. The proteins were separated and transferred using sodium dodecyl sulfate polyacrylamide gel electrophoresis (Epizyme, China) and 0.2-μm polyvinylidene difluoride membranes (Millipore, USA). Subsequently, the membrane with proteins was incubated with primary and secondary antibodies as described in Supplementary Table 1 .

Every experiment was repeated at least three times. All experimental data were expressed as mean ± SD, and it was analyzed using GraphPad Prism 8.0. Two groups were compared using Student’s t-test, and the difference of more than two groups was analyzed via one-way analysis or nonparametric Kruskal-Wallis test according to the variances. P value of less than 0.05 was considered statistically significant.

Aged mice are characterized by obesity, ovarian atrophy, and decreased ovarian function. We monitored the change in body mass of mice during CEFFE treatment for two weeks, and the results showed that the body mass in the Aged+CF group gradually decreased compared to that in the Aged group ( Figure 1A ). Meanwhile, the diet, mental state, and activity of Aged+CF mice were not affected compared to those of Aged and Control mice. Besides, after 2 weeks of treatment, the weight of ovaries, sex hormones, and follicle count in ovaries were analyzed. Higher ovary weight (Aged+CF vs Aged: 95% CI, 0.02 to 2.28 mg; P<0.05) ( Figures 1B, C ), increased AMH and E₂ levels and decreased FSH levels were observed in the Aged+CF group than in the Aged group ( Table 1 ). Besides, the recovery effect of CEFFE on sex hormones lasted up to 2 months after treatment ( Figures 1D–F ). All these indices showed an opposite trend in the Aged group than that in the Control group. The above results suggested that CEFFE could promote the recovery of hormone synthesis function in aged mice, which is responsible for regulating the stability of ovarian microenvironment and stimulating the mature of oocytes. Furthermore, ovaries were stained with Hematoxylin and Eosin to assess the changes in ovarian morphology and follicular number. As expected, in the Aged group, the ovaries exhibited a significantly reduced number of follicles at all stages. After treatment with CEFFE for 2 weeks, many more growth follicles appeared in the Aged+CF group than in the Aged group, especially the primary (95% CI, 0.44 to 4.91; P<0.05) and secondary follicles (95% CI, 1.02 to 8.19; P<0.05) ( Figures 1G, H ). Besides, granulosa cells were more regularly arranged in Aged+CF group as compared to that in Aged group. In conclusion, CEFFE could improve the function of aged ovaries on the size of ovaries, hormone secretion, and number of follicles.

As indicated by the above findings, CEFFE could improve the ovarian function. To further investigate the effect of CEFFE on the fertility of mice, we collected oocytes from the ampulla of female mice after 18 h of intercourse, and we cultured them in vitro to observe and record the process of embryo development. The results showed that more oocytes, 2 cell embryos, 4-8 cell embryos, morulas, and blastocysts were observed in the Aged+CF group than in the Aged group, indicating that CEFFE could enhance the development of embryos ( Figures 2A, B ). Next, we performed natural fertility assays in each group to evaluate the litter size. The litter size was the highest and the time from mating to birth was the shortest in the Control group. Compared to the Aged group, mice from Aged+CF group had a greater litter size and shorter time from mating to birth ( Figures 2C, D ). The detail results were displayed in Table 2 . So far, we found that CEFFE significantly improved the number of retrieved oocytes, embryonic developmental ability, and final litter size, indicating that CEFFE significantly improved the fertility of aged mice.

The ovary was mainly composed of the medulla, rich in blood and lymphatic vessels, and cortex, rich in different stage follicles. In aged ovaries, the number of microvessels in the medulla decreased due to medullary fibrosis. Vessels could provide sufficient nutrients for follicle development, and they were essential for the maintenance of ovarian microenvironment. Therefore, expression of the vascular endothelial marker CD31 was detected via IHC, and the results showed that angiogenesis significantly decreased in the Aged group (Aged vs Control: 95% CI, -26.68 to -11.10; P<0.0001) whereas it significantly increased after treatment (Aged+CF vs Aged: 95% CI, 9.55 to 26.51; P<0.0001), suggesting that CEFFE could promote ovarian medulla angiogenesis in aged mice ( Figure 3A ). Next, proliferation and senescence were evaluated with Ki67 and P16 via IHC analysis. In the Aged group, expression of the proliferation marker Ki67 decreased (Aged vs Control: 95% CI, -0.51 to -0.16; P<0.001) and that of the senescence marker P16 increased in GCs (Aged vs Control: 95% CI, 0.07 to 0.41; P<0.01). After treatment with CEFFE, more proliferative (Aged+CF vs Aged: 95% CI, 0.11 to 0.43; P<0.01) and less senescent GCs (Aged+CF vs Aged: 95% CI, -0.42 to -0.01; P<0.05) were observed ( Figures 3B, C ). Additionally, to rule out the tumorigenic risk of CEFFE treatment, the expression of tumor marker P53 (oncogenic gene) and PTEN (suppressive gene) was analyzed, and the results displayed no difference between each group, indicating that CEFFE had no tumorigenic risk for ovarian function recovery ( Figures 3D, E ). It is well known that during aging, the morphology and number of mitochondria change, which are vital for maintaining cell function. To observe changes in the internal microstructure of cells, ovary TEM was performed. In zona pellucida, the microvilli between inner zona pellucida and oocytes, which are vital for material and message exchange between granulosa cells and oocytes in aged ovary, were seriously degenerated. However, the bidirectional granulosa cells-oocytes signaling is critical for creating a dynamic intrafollicular microenvironment (8).In granulosa cells, more vacuoles and swollen mitochondria with less cristae were observed in the aged ovary. Interestingly, in the Aged+CF group, the microvilli were integrated, and mitochondrial morphology was clear with more cristae. These results strongly indicated that CEFFE treatment effectively ameliorated the microstructure disorder of ovary ( Figure 3F ). Above all, we demonstrated that CEFFE could promote angiogenesis in medulla and alleviate the senescence of granulosa cells.

We performed transcriptome sequencing and differential gene expression analysis on ovaries from the Control, Aged, and Aged+CF groups to find the main signal pathways with changes. Firstly, by comparing the transcriptome data of each of the two groups (Aged vs Ctrl and Aged+CF vs Aged, respectively), only the genes with absolute fold change >1.5 and P value < 0.05 were defined as DEGs ( Figures 4A–C ). Next, the intersection of the two clusters was performed via Venn analysis, and a total of 252 genes that showed opposite trends were selected in the two clusters ( Figure 4D ). The GO pathway analysis showed that the 252 DEGs were most enriched in “transmembrane transport” of the top 20 pathways, which might be related to the process of active proteins in CEFFE interacting with cells ( Figure 4E ). The KEGG pathway analysis showed that the 252 DEGs were enriched in pathways related to cell senescence, including “Homologous recombination”, “Longevity regulating pathway”, and “Glutathione metabolism” ( Figure 4F ). Among these, bioinformatic analysis strongly suggested that homologous recombination (HR), an essential method for DNA double-strand break (DSB) repair, was significantly enriched, indicating that CEFFE could repair the ovarian function in aged mice by promoting DSB repair. To further confirm the change in DNA damage repair in vivo, γH2AX was detected in the ovaries via IHC, and more γH2AX-positive cells were observed in the Aged group than in the Control and Aged+CF groups (Aged vs Control: 95% CI, 0.16 to 0.50; P<0.001), (Aged+CF vs Aged: 95% CI, -0.69 to -0.35; P<0.0001), especially in granulosa cells (the red arrows indicate γH2AX positive cells) ( Figure 4G ). Collectively, the RNA-Seq and bioinformatic analyses indicated that DNA damage was most obviously changed in the ovaries with CEFFE treatment, which might play an essential role in rescuing the ovarian function.

Next, we established an in vitro model of senescent human granulosa cell line (KGN) through D-gal treatment. Senescence β-Galactosidase Staining Kit was used to examine the expression of β-Galactosidase, which increased during senescence, and the results showed that 10 mg/mL and 20 mg/mL D-gal intervention for seven days increased the number of SA-β-gal-positive cells ( Supplementary Figure 1A ). Additionally, the expression of senescence markers P16 and P21 significantly increased following D-gal treatment compared to that in the Control group ( Supplementary Figure 1B ). Finally, treatment with 10 mg/mL D-gal for 7 days was used to establish the senescent KGN model. Following this, senescent KGN was incubated with three different concentrations of CEFFE (CFL: 0.06 μg/μL, CFM: 0.15 μg/μL, CFH: 0.3 μg/μL) for 24 h, and EdU test was performed. The results showed that the Aged group was less proliferative than the Control group (95% CI, -26.37 to -1.59%; P<0.05), and only CFH could increase the proliferation of aged KGN cells at 24 h (95% CI, 6.03 to 32.15%; P<0.01), ( Figures 5A, B ). To further analyze whether CEFFE could affect the aged KGN proliferation, we prolonged the co-culture time to 72 h. The results showed that CEFFE could promote the proliferation of senescent KGN in a dose- and time-dependent manner ( Figure 5C ). Further, E₂ content in cellular supernatant was measured via ELISA test with CEFFE intervention for 48 h, and the data demonstrated that senescent KGN secreted less E₂ (Aged vs Control: 95% CI, -160.21 to -32.89 pg/mg; P<0.01) and CEFFE could significantly promote E₂ secretion (95% CI: Aged+CFL vs Aged, 5.80 to 116.07 pg/mg, P<0.05; Aged+CFM vs Aged, 0.31 to 117.26 pg/mg, P<0.05; Aged+CFH vs Aged, 16.33 to 126.60 pg/mg, P<0.01), ( Figure 5D ). Taken together, our findings suggested that CEFFE contributed to the proliferation and hormone secretion of senescent KGN cells in a dose-dependent manner.

To examine the effect of CEFFE on KGN senescence, CEFFE was incubated for 48 h, and it was used in the follow-up experiment. As shown in Figure 6A , CEFFE could significantly reduce the number of SA-β-gal-positive cells compared to those in the Aged group. Meanwhile, the results of Western Blotting also demonstrated that the expression levels of senescent markers P21 (Aged vs Control:95% CI, 0.59 to 3.91; P<0.05) and P16 (Aged vs Control: 95% CI, 0.12 to 0.93; P<0.05) and DNA damaged marker γH2AX (Aged vs Control: 95% CI, 1.05 to 3.67; P<0.01) increased in the Aged group while they decreased in the Aged+CF group (Aged+CF vs Aged: P21, 95% CI, -3.65 to -0.33, P<0.05; P16, 95% CI, -1.02 to -0.21, P<0.01; γH2AX, 95% CI, -3.22 to -0.60, P<0.01), ( Figures 6B, C ). To further verify the anti-senescent effect of CEFFE, human primary ovarian granulosa cells (hGCs) were cultured with/without CEFFE in vitro, and the mRNA expression of some senescence-related genes was detected via RT-PCR. Interestingly, CEFFE could significantly reduce the expression of senescence-related genes in hGCs, which was consistent with KGN cells ( Figure 6D ). Next, given the essential role of CEFFE in HR pathway upon DSBs indicated by the transcriptome data in aged ovary, we examined the effect of CEFFE on the repair of camptothecin-induced DNA damage, which could only be rescued by HR but not by non-homologous end joining pathway, another method for DSBs. Briefly, KGN was incubated with 10 μM camptothecin for 2 h to induce DSBs (31). Subsequently, CEFFE was added and co-cultured for 2 h. γH2AX, a sensitive marker of DSBs, was measured via Western Blotting, and it was found to be significantly decreased with CEFFE intervention (CPT+CF vs CPT: 95% CI, -5.72 to -1.78; P<0.01), ( Figures 6E, F ). Meanwhile, the hGCs were identified with FSHR ( Figure 6G ) and similar results were observed in hGCs (CPT+CF vs CPT: 95% CI, -9.23 to -1.80%; P<0.01), ( Figures 6H, I ). Taken together, our results indicated that CEFFE could promote the repair of DNA damage and reverse cellular senescence of KGN cells.