Follicular fluid-derived exosomal HMOX1 promotes granulosa cell ferroptosis involved in follicular atresia in geese (Anser cygnoides).

After dissection, HSYFs and ASYFs with similar sizes were selected for comparison. The results showed that compared to HSYFs, ASYFs had a dark color, wrinkled surface, and multiple hemorrhagic spots. The surface of the atresia follicles was covered in red ( Figure 1A ). Observation and statistics of the follicles on the whole ovary showed that some small yellow follicles were in the process of atresia ( Figure 1B ). H&E staining results showed that the GC layer of the HSYF was uniform and intact, closely connected with theca cells. Compared with the HSYF, the GCs of the ASYF were loosely connected and with theca cells, with some areas showing inward shedding ( Figures 1C –1H). Figure 1 Appearances and morphologies analysis of HSYFs and ASYFs. (A) General appearances of the HSYF and ASYF. (B) Statistical analysis of number of HSYFs and ASYFs per ovary. (C–H) H&E staining of the GC layers of HSYFs and ASYFs. The red arrows represent the GC layers and the blue arrows represent the membrane layer. Scale bar, 400 μm (C and F), 200 μm (D and G), 100 μm (E and H). Data are expressed as the means ± SD. Figure 1 Appearances and morphologies analysis of HSYFs and ASYFs. (A) General appearances of the HSYF and ASYF. (B) Statistical analysis of number of HSYFs and ASYFs per ovary. (C–H) H&E staining of the GC layers of HSYFs and ASYFs. The red arrows represent the GC layers and the blue arrows represent the membrane layer. Scale bar, 400 μm (C and F), 200 μm (D and G), 100 μm (E and H). Data are expressed as the means ± SD. Tunnel staining showed that the mortality levels of the GC layers in ASYFs were significantly higher than that of HSYFs ( Figure 2A ). The relative contents of E2 and P4 of the GC layers in ASYFs were significantly lower than those in HSYFs ( P <0.05). The mRNA expression levels of CYP19A and HSD17B7 of the GC layers in ASYFs were significantly lower than those in HSYFs ( P < 0.05) ( Figure 2B ). Figure 2 GCs activity and sex steroid synthesis ability analysis of the GC layers of HSYFs and ASYFs. (A) GCs apoptosis in HSYFs and ASYFs were determined by TUNEL assay. The blue stain indicated the cell nucleus. Scale bar: 25 mm. (B) Relative content of sex steroid E2, P4 and mRNA expression levels of sex steroid synthesis-related genes CYP19A, HSD17B7, FSHR . Data are expressed as the means ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001. Figure 2 GCs activity and sex steroid synthesis ability analysis of the GC layers of HSYFs and ASYFs. (A) GCs apoptosis in HSYFs and ASYFs were determined by TUNEL assay. The blue stain indicated the cell nucleus. Scale bar: 25 mm. (B) Relative content of sex steroid E2, P4 and mRNA expression levels of sex steroid synthesis-related genes CYP19A, HSD17B7, FSHR . Data are expressed as the means ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001. The mRNA expression levels of GPX4, FPN1 , and FTH1 of the GC layers in ASYFs were significantly decreased ( P < 0.05), while COX2, NCOA4 , and VDAC3 mRNA expression levels were significantly higher than those of HSYFs ( P < 0.05) ( Figure 3A ). Compared with HSYFs, the contents of Fe 2+ and MDA were significantly increased in the GC layers of ASYFs ( P < 0.01) ( Figures 3B –3C). The expression level of GPX4 in GC layers of ASYFs was significantly lower than those of HSYFs ( P < 0.05) ( Figures 3D –3E). Electron microscope observation of mitochondrial status in GCs of HSYFs and ASYFs revealed partial mitochondrial shrinkage, wrinkling, disappearance of mitochondrial cristae, and membrane rupture in ASYFs ( Figure 3F ). Figure 3 The ferroptosis markers changes of the GC layers of HSYFs and ASYFs. (A) The relative mRNA expression levels of ferroptosis-related genes GPX4, COX2, NCOA4, FPN1, FTH1, VDAC3 of the GC layers between HSYFs and ASYFs. (B) The relative amount of Fe 2+ of the GC layers between HSYFs and ASYFs. (C) The relative content of MDA of the GC layers between HSYFs and ASYFs. (D–E) The expression level of GPX4 between HSYFs and ASYFs detected by western bolting and immunohistochemistry, respectively. Scale bar: 200 μm. (F) Ultrastructure of mitochondria of the GC layers between HSYFs and ASYFs. Data are expressed as the means ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001. Figure 3 The ferroptosis markers changes of the GC layers of HSYFs and ASYFs. (A) The relative mRNA expression levels of ferroptosis-related genes GPX4, COX2, NCOA4, FPN1, FTH1, VDAC3 of the GC layers between HSYFs and ASYFs. (B) The relative amount of Fe 2+ of the GC layers between HSYFs and ASYFs. (C) The relative content of MDA of the GC layers between HSYFs and ASYFs. (D–E) The expression level of GPX4 between HSYFs and ASYFs detected by western bolting and immunohistochemistry, respectively. Scale bar: 200 μm. (F) Ultrastructure of mitochondria of the GC layers between HSYFs and ASYFs. Data are expressed as the means ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001. Fluorescence probe detection of ROS level showed a significant increase in ASYFs compared with that of HSYFs ( P < 0.001) ( Figure 4A ). The relative contents of H 2 O 2 and MDA of the GC layers in ASYFs were significantly higher than those of HSYFs ( P < 0.01), while the relative contents of T-AOC, SOD, and GSH were significantly lower than those of HSYFs ( P < 0.01) ( Figures 4B –4C). Figure 4 ROS content and oxidative stress levels of the GC layers of HSYFs and ASYFs. (A)The content of ROS of the GC layers between HSYFs and ASYFs. (B) The oxidative stress levels of the GC layers between HSYFs and ASYFs. Data are expressed as the means ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001. Figure 4 ROS content and oxidative stress levels of the GC layers of HSYFs and ASYFs. (A)The content of ROS of the GC layers between HSYFs and ASYFs. (B) The oxidative stress levels of the GC layers between HSYFs and ASYFs. Data are expressed as the means ± SD. * P < 0.05, ** P < 0.01, *** P < 0.001. Exosomes extracted from goose FF were characterized by TEM, NTA, and Western blot. The results of TEM showed that the isolated exosomes had a typical cupped structure. The average particle size of HSYF-FF exosomes and ASYF-FF exosomes were 70.67 nm and 70.59 nm, respectively. The concentration of HSYF-FF exosomes and ASYF-FF exosomes were 4.86 × 10 11 particles/mL and 4.72 × 10 11 particles/mL, respectively. Exosome markers Tsg101, CD63, and CD9 tested positive for HSYF-FF exosomes and ASYF-FF exosomes, which were consistent with previously reported characteristics of exosomes ( Figure 5 ). Figure 5 Characterization of exosomes derived from goose HSYF-FF and ASYF-FF. (A) The transmission electron micrographs (TEM) and nanoparticle tracking analysis (NTA) profile of exosomes of HSYF-FF. (B) The TEM and NTA profile of ASYF-FF exosomes. Scale bar, 100 nm. (C) The positive exosomal marker (Tsg101, CD63, CD9) immunoblots of follicular fluid exosomes (HSYF-Exo and ASYF-Exo) and whole cell lysis of granulosa cells (HSYFs and ASYFs). Figure 5 Characterization of exosomes derived from goose HSYF-FF and ASYF-FF. (A) The transmission electron micrographs (TEM) and nanoparticle tracking analysis (NTA) profile of exosomes of HSYF-FF. (B) The TEM and NTA profile of ASYF-FF exosomes. Scale bar, 100 nm. (C) The positive exosomal marker (Tsg101, CD63, CD9) immunoblots of follicular fluid exosomes (HSYF-Exo and ASYF-Exo) and whole cell lysis of granulosa cells (HSYFs and ASYFs). GCs in HSYFs were successfully isolated, and the marker FSHR was identified to be widely present on the isolated GCs ( Figure 6A ). To determine if exosomes in ASYF-FF could be taken up by GCs, we labeled the extracellular vesicle membrane with Dil (red fluorescent dye) and the cell membrane of GCs with DiO (green fluorescent dye), the exosomes in ASYF-FF could be taken up by GCs observed under an inverted fluorescence microscope ( Figure 6B ). Figure 6 Effect of HSYF-FF and ASYF-FF exosomes on ferroptosis in GCs. (A) Isolation and identification of goose GCs. Diagram of GCs (Scale bar: 100 μm), DAPI staining of cell nucleus, and fluorescent image of FSHR. (B) ASYF-FF exosomes were uptaken by GCs. (C) Cell proliferation, relative content of MDA and the amount of Fe 2+ of GCs treated with HSYF-Exo, ASYF-Exo, and ASYF-Exo+Fer-1, respectively. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). Figure 6 Effect of HSYF-FF and ASYF-FF exosomes on ferroptosis in GCs. (A) Isolation and identification of goose GCs. Diagram of GCs (Scale bar: 100 μm), DAPI staining of cell nucleus, and fluorescent image of FSHR. (B) ASYF-FF exosomes were uptaken by GCs. (C) Cell proliferation, relative content of MDA and the amount of Fe 2+ of GCs treated with HSYF-Exo, ASYF-Exo, and ASYF-Exo+Fer-1, respectively. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). To investigate the effect of exosomes in ASYF-FF on the activity of isolated GCs in geese, the CCK8 kit was used to detect the changes in GCs activity caused by different concentration gradients of ASYF-FF exosomes. The results showed that with the exosome concentration in ASYF-FF increased, GCs activity gradually decreased ( P < 0.05), with a significant difference observed at 20 µM ( Figure 6C -a). When 20 µm of the ferroptosis inhibitor Fer-1 was added, it interfered with the death of GCs induced by exosomes in ASYF-FF ( Figure 6C -b). To detect the effects of different states of exosomes on the activity of GCs, we treated GCs with HSYF-FF exosomes (20µM) and ASYF-FF exosomes (20µM), respectively. The results showed that HSYF-FF exosomes did not reduce the cell activity of GCs, while ASYF-FF exosomes significantly decreased the cell activity of GCs. Moreover, the addition of the ferroptosis inhibitor Fer-1 (20 µM) in vitro mitigated the decrease in GCs activity induced by exosomes in ASYF-FF ( P < 0.05) ( Figure 6C -c). Additionally, we used an MDA kit to detect the effects of adding HSYFs exosomes and ASYFs exosomes on the lipid peroxide content and Fe 2+ content of GCs. The results showed that the addition of ASYF-FF exosomes significantly elevated the MDA content and Fe 2+ levels in GCs compared to the group treated with HSYF-FF exosomes. Moreover, Fer-1, a ferroptosis inhibitor, inhibited the increase of MDA and Fe 2+ in GCs caused by the addition of ASYF-FF exosomes ( P < 0.05) ( Figure 6C -d, e). Ferroptosis is closely related to mitochondrial function, and multiple studies have used changes in MMP as an important indicator to evaluate the occurrence of ferroptosis. We further examined the effects of different treatments on the MMP in GCs of geese. The results showed that the exosomes of ASYF-FF decreased the MMP of GCs compared to those of HSYF-FF. Additionally, Fer-1 inhibited the decrease in MMP of GCs induced by exosomes in ASYF-FF ( Figure 7 ). Figure 7 Mitochondrial Membrane Potential (MMP) changes of GCs treated with HSYF-Exo, ASYF-Exo and, and ASYF-Exo+Fer-1, respectively. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). Figure 7 Mitochondrial Membrane Potential (MMP) changes of GCs treated with HSYF-Exo, ASYF-Exo and, and ASYF-Exo+Fer-1, respectively. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). Label-free proteomic analysis was used to detect the proteomic profiles in exosomes extracted from HSYF-FF and ASYF-FF. The MaxQuant software was used for protein qualitative and quantitative analysis. The qualitative screening criteria for proteins were protein FDR ≤ 0.01 and peptide FDR ≤ 0.01. In total, 4082 proteins were identified, among which 3973 proteins were quantified in the HSYFs FF exosomes and 4046 proteins were quantified in the group ( Figures 8A –8B, Supplementary Tables S1 – S2 ). Differentially expressed proteins were identified with a cut-off of absolute fold change ≥ 2 and P ‐value < 0.05. The results showed that 1615 proteins were differentially expressed between ASYFs and HSYFs control. Among them, 905 exosomal proteins were significantly higher, and 710 proteins were lower in ASYF-FF exosomes than in HSYFs control ( Figures 8C –8E, Supplementary Table S3 ). To select the key proteins involved in the progression of follicular atresia, we focused on the proteins that were differentially expressed between the two groups. KEGG pathway analysis was conducted to classify the affected biological functions and processes. Significant enriched pathways included Protein export, Ferroptosis, Proteasome, etc., which were determined according to Fold Enrichment ( Figure 8F , Supplementary Table S4 ). GO analysis of differentially expressed exosomal proteins in FF from HSYFs and ASYFs revealed several important biological processes, including cell surface receptor signaling pathway, cell adhesion, and biological adhesion. Additionally, various molecular functions such as signaling receptor activity, molecular transducer activity, and transmembrane signaling receptor activity were identified. Furthermore, significant changes were observed in the localization of proteins, particularly in the cell periphery, plasma membrane, and intrinsic component of the membrane ( Supplementary Figure. S1 ). Figure 8 Proteomic analysis of FF exosomes derived from HSYFs and ASYFs. (A) Statistics of identification and quantitative results. (B) Venn plot of protein identified between HSYFs and ASYFs group. (C–E) Histogram, volcano plot and heat map of differentially expressed proteins in HSYF-FF exosomes (n = 3) and ASYF-FF exosomes (n = 3). (F) KEGG enrichment pathway analysis for differentially expressed proteins derived from HSYF-FF exosomes and ASYF-FF exosomes. (G) The protein protein interaction (PPI) network of DEPs Cluster (MCODE scores >4). (H) Linear regression fitted for Log2 fold change of selected proteins determined via 4D label-free quantitative proteomics and PRM. n = 3 per group. (I) Expression validation of the key protein HMOX1 in HSYF-FF exosomes and ASYF-FF exosomes. Figure 8 Proteomic analysis of FF exosomes derived from HSYFs and ASYFs. (A) Statistics of identification and quantitative results. (B) Venn plot of protein identified between HSYFs and ASYFs group. (C–E) Histogram, volcano plot and heat map of differentially expressed proteins in HSYF-FF exosomes (n = 3) and ASYF-FF exosomes (n = 3). (F) KEGG enrichment pathway analysis for differentially expressed proteins derived from HSYF-FF exosomes and ASYF-FF exosomes. (G) The protein protein interaction (PPI) network of DEPs Cluster (MCODE scores >4). (H) Linear regression fitted for Log2 fold change of selected proteins determined via 4D label-free quantitative proteomics and PRM. n = 3 per group. (I) Expression validation of the key protein HMOX1 in HSYF-FF exosomes and ASYF-FF exosomes. Proteins usually interact with other proteins in biological functions. Therefore, the key differential proteins that play a role in HSYF- and ASYF-FF exosomes were analyzed using PPI (the medium confidence score to > 0.4, the modules with MCODE scores > 4). The results revealed that HOXM1 is not only the differential protein of the top pathway ferroptosis but also the key node protein in the PPI network ( Figure 8G ). We selected 18 DEPs for PRM analysis, including ANXA2, PSMA7, COL12A1, MYO6, ATP6V1H, ATG3, ETF1, NCL, SLC4A11, ACTR3, MYH11, PCBP2, UBE2O, FTH1, APOB, COPA, TLN1, and HSPA5, which are important for cell growth and death. The results showed that the expression trend of all proteins was consistently detected by PRM and 4D-Label-Free proteomics analysis. In the exosomes of ASYF-FF, ANXA2, PSMA7, COL12A1, ATP6V1H, ATG3, ETF1, NCL, ACTR3, MYH11, UBE2O, COPA, and TLN1 levels were up-regulated, SLC4A11, PCBP2, FTH1, APOB, and HSPA5 levels were decreased compared to the normal HSYF-FF group, which were consisted with the results of proteomics analysis ( Figure 8H , Table S5). Meanwhile, the expression of exosomal HOXM1 from HSYF-FF and ASYF-FF was validated by WB. The result confirmed the presence of HMOX1 in exosomes derived from ASYF-FF, whereas its presence was barely detected in the HSYFs control ( Figure 8I ). To investigate whether the occurrence of ferroptosis in GCs of ASYFs was related to the overexpression of the key protein HMOX1, the recombinant plasmid pEGFP-N1-HMOX1 was successfully constructed. Changes in the proliferation level, GPX4 expression, MDA, and Fe 2+ contents of GCs were detected by transfecting the pEGFP-N1-HMOX1 recombinant plasmid into GCs and adding Znpp, an HMOX1 inhibitor, respectively. The results showed that both the mRNA and protein expression levels of HMOX1 in the overexpression group were significantly increased compared to the pEGFP-N1 group ( P < 0.05), while the HMOX1 expression level in the Znpp group and Znpp + HMOX1 group was significantly decreased compared to the HMOX1 overexpression group ( P < 0.05) ( Figure 9A ). The overexpression of HMOX1 significantly reduced the expression level of GPX4 in GCs ( P <0.05), and this situation was significantly inhibited after Znpp treatment ( Figures 9B –8F). Besides, the overexpression of HMOX1 significantly decreased the activity of GCs ( P < 0.05), and the cell activity of both the Znpp group and the Znpp+HMOX1 group was significantly higher than that of the HMOX1 group ( P < 0.05) ( Figure 9C ). Moreover, the overexpression of HMOX1 could cause an increase in Fe 2+ and MDA content in GCs, and Znpp could alleviate the increase in Fe 2+ and MDA content caused by HMOX1 overload ( Figures 9G –9H). Figure 9 The ferroptosis changes in GCs induced by FF-derived exosomal HMOX1. The expression of HMOX1 (A, E), GPX4 (B, F), GCs proliferation (C), relative content of MDA (G), and the relative amount of Fe 2+ (H) of GCs after overexpression and inhibition of HMOX1. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). Figure 9 The ferroptosis changes in GCs induced by FF-derived exosomal HMOX1. The expression of HMOX1 (A, E), GPX4 (B, F), GCs proliferation (C), relative content of MDA (G), and the relative amount of Fe 2+ (H) of GCs after overexpression and inhibition of HMOX1. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). To further deduce the mechanism of HMOX1 in the ferroptosis of GCs, we detected the expression changes of key proteins ATG5, FPN1, FTH1, PCBP2, and LC3II in the ferroptosis pathway in GCs transfected with pEGFP-N1-HMOX1 recombinant plasmid and supplemented with the HMOX1 inhibitor Znpp, respectively. After transfection with pEGFP-N1-HMOX1, the mRNA and the protein expression levels of NCOA4 , ATG5, and LC3II significantly increased, while the mRNA and the protein expression levels of FPN1, FTH, and PCBP2 significantly decreased. These changes could be suppressed by the addition of Znpp ( Figures 10A –10C). Figure 10 HMOX1 promotes ferroptosis via decreasing FTH1 and FPN1 expression in ferroptosis pathway. (A) The changes of mRNA expression of key proteins in ferroptosis pathways. (B, C) The changes of key proteins expression in ferroptosis pathways. (D) The deduced mechanism of HMOX1 promoting ferroptosis via decreasing FTH1 and FPN1 expression in ferroptosis pathway. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05). Figure 10 HMOX1 promotes ferroptosis via decreasing FTH1 and FPN1 expression in ferroptosis pathway. (A) The changes of mRNA expression of key proteins in ferroptosis pathways. (B, C) The changes of key proteins expression in ferroptosis pathways. (D) The deduced mechanism of HMOX1 promoting ferroptosis via decreasing FTH1 and FPN1 expression in ferroptosis pathway. Data are expressed as the means ± SD. Different lowercase letters indicate significant differences ( P < 0.05).

All animal procedures in this study were approved by the Institutional Animal Ethical Committee at Yangzhou University (Permit Number: YZUDWSY2022-018, Jiangsu Province, China). Ten-month-old, 30 laying Zhedong white geese (4.21 ± 0.26 Kg) were received from the National Waterfowl Germplasm Resource Pool (Taizhou, Jiangsu, China). These geese were fed standard commercial diets. After anesthesia with pentobarbital sodium, the geese were euthanized by neck bleeding. The geese's abdomen was disinfected with alcohol and dissected. The ovaries were collected under aseptic conditions and rinsed with phosphate-buffered saline ( PBS ) containing 2% penicillin-streptomycin solution to peel off the follicles. The ovaries were then cleaned thoroughly in a clean bench. The HSYFs and ASYFs were collected based on follicle size and morphology, as previously described, separately ( Kovacs, et al., 1992 ; Zhang et al., 2018 ; Yang, et al., 2019 ). Then randomly divided into three groups. And FF also was extracted from HSYFs and ASYFs (n=200, each group), respectively. Individual ovarian follicles are meticulously punctured using a fine needle or a similar instrument to aspirate the FF. The aspirated FF is then filtered through a sterile mesh or filter to eliminate any large cellular debris or clots, facilitating the acquisition of a clear and homogeneous fluid sample, then centrifuged at 4 ℃, 3000g for 10 min, and frozen at -80℃ for exosome extraction. Some HSYFs and ASYFs were fixed in 4% paraformaldehyde for subsequent paraffin and frozen sections. The remaining HSYFs and ASYFs were stripped and collected the GC layers, respectively, and then divided into two parts, one part was soaked in Gluta fixative for transmission electron microscopy ( TEM ) observation, and another part was placed in liquid nitrogen for RNA and total protein extraction. HSYFs and ASYFs were fixed for 24 h at 4°C in 4% neutral paraformaldehyde solution, dehydrated in grade ethanol, cleaned in xylene, and finally embedded in paraffin, respectively. The implanted samples were cut into 5 μm sections, slides were deparaffinized and hydrated using xylene and graded alcohol series. For antigen retrieval, slides were placed in the coupling jars containing 0.1 M sodium citrate (pH 6.0) and microwaved at 750 W for 1 min. After cooling, slides were incubated for 30 min at 25ºC in 0.1 M Tris-HCl (pH 7.5) containing 3% bovine serum albumin and 20% normal bovine serum. The slides were then washed with PBS for 3 times. Which were then utilized to perform H&E staining and TUNEL assay (Yeasen Biotechnology 40306ES50, Shanghai). Standard procedures were used to carry out the H&E staining. DAPI assays were performed on paraffin sections according to the instructions (Beyotime C1005, Shanghai). The remaining probes that did not enter the cells were washed with PBS three times for removal. Images were obtained using a fluorescence microscope (Zeiss 2.0, Germany), and ImageJ software (National Institutes of Health, MD, USA) was used for the quantification of TUNEL/DAPI positive nuclei cells. We stripped and collected the GC layers of HSYFs and ASYFs, respectively, homogenized them in PBS, and centrifuged them at 12,000 rpm for 15 min at 4°C. The concentrations of E2 and P4 were determined using an ELISA kit (MEIMIAN, Shanghai, China). The contents of GSH, T-AOC, SOD, and H 2 O 2 were measured according to the manufacturer's instructions (JianCheng, Nanjing, China). The results were quantified at 450 nm using a Multimode Microplate Reader (Varioskan LUX, Thermo). The tissues and the intracellular MDA levels were measured using the Lipid Peroxidation MDA Assay Kit (Beyotime, Jiangsu, China), which is based on the reaction of MDA and thiobarbituric acid ( TBA ). The absorptions were measured at 532 nm. MDA levels were calculated according to the established standard curve using the reference standards in the kit under the same condition. Fe 2+ levels in cells of each group were specifically determined using the iron assay kit (Abcam ab83366, USA) by the manufacturer's instructions. Free Fe 2+ , rather than Fe 3+ or Fe 2+ complexed in heme, reacts specifically with Ferene S to produce a stable-colored complex with absorbance at 593 nm. Briefly, the collected cells were homogenized following the addition of iron assay buffer. The mixture was centrifuged at 8,000 × g and 4°C for 15 min. The 50 μL supernatant was mixed with 100 μL of the iron probe and incubated for 1 h at 25°C in the dark. The absorbance of each treatment group was measured at 593 nm using a microplate reader. The HSYFs and ASYFs were rapidly harvested, fixed with 4% paraformaldehyde, and embedded in paraffin for immunohistochemical staining. Tissue sections (thickness of 5 μm) were incubated with primary antibodies against GPX4 (1:100, Abmart PHY7049) overnight at 4°C and then with secondary antibodies (anti-rabbit IgG, 1:5000, Thermo Pierce 31210) at 37°C for 1 h. The following specific antibodies were used, and the average integrated optical density ( IOD ) of the cells that positively expressed GPX4 in six randomly selected regions per sample was measured at a magnification of 200× using ImageJ software analysis. The HSYFs and ASYFs were taken for frozen sections, and incubated in 10 μmol/L DHE solution for 90 min under dark conditions at 37°C, washed with PBS, and photographed under a fluorescence microscope after sealing. Images were obtained using a fluorescence microscope, and ImageJ software was used for quantitative fluorescence analysis. HSYF-FF exosomes and ASYF-FF exosomes were separated via differential centrifugation as described previously ( Li et al., 2020 ). Approximately 10 mL of FF collected from the follicles was centrifuged at 2000g for 10min at 4℃ to remove follicular cells and cell fragments. Apoptotic bodies and microbubbles were removed via centrifugation at 12,000 g for 40 min at 4℃. The FF was filtered through a 0.22 mm filter. Exosomes were then centrifuged and purified at 120,000 g for 70 min at 4℃, resuspended in PBS, and centrifuged again at 120,000 g for 70 min at 4 ℃. Then, 500 mL of PBS was used to suspend the exosomes. After the concentration of exosomes was determined using the bicinchoninic acid ( BCA ) method (Beyotime, Jiangsu, China), at least about 60 μg of exosomes could be extracted from 1 mL FF. The freshly prepared exosomes were immediately used for the experiment or frozen at -80℃ for subsequent use. A 1 mm 3 piece of the GC layer was prefixed in 2.5% glutaraldehyde phosphate (0.1 M, pH 7.4) overnight at 4 ℃, and post fixed in 2% buffered osmium tetraoxide, then embedded in Epon812 (Merck) followed by dehydration. Ultrathin sections (approximately 70 nm thick) were cut, and stained with uranyl acetate and lead citrate. The sections were viewed with an H600 transmission electron microscope (Hitachi, Tokyo, Japan) at 80 kV. For observing mitochondria, the low magnification field (5000 ×) was first used to determine the viewing area, and then the high magnification field (20,000 ×) was used to observe ultrastructures of GCs. More than 3 sections per sample were photographed. Exosomes were analyzed by TEM as previously described ( Thery, et al., 2006 ; Lewandowska, et al., 2017 ). A total of 20 μL of exosome suspension (5 µg/µL) was fixed on a continuous grid and then negatively stained with 2% uranyl acetate solution for 1 min and air‐dried. The samples were observed by transmission electron microscope FEI Tecnai G2 Spirit by a digital camera Morada at an acceleration voltage of 120 kV. Exosome diameter was measured using nanoparticle tracking analysis ( NTA ). Nanoparticle tracking analysis measurements were performed using a VivaCellBiosceinces with ZetaView PMX 110 (Particle Metrix, Meerbusch, Germany) and corresponding software ZetaView 8.04.02. Isolated exosome samples were appropriately diluted using 1× PBS buffer (Biological Industries, Israel) to measure the particle size and concentration. NTA measurement was recorded and analyzed at 11 positions ( Peterson, et al., 2015 ). The ZetaView system was calibrated using 100 nm polystyrene particles. SDT (4% SDS, 100 mM Tris-HCl, 1 mM DTT, pH 7.6) buffer was used for sample lysis and protein extraction. The amount of protein was quantified with the BCA Protein Assay Kit (Bio-Rad, Hercules, CA). Protein digestion by trypsin was performed according to filter-aided sample preparation ( FASP ) procedure ( Wisniewski, et al., 2011 ). The digest peptides of each sample were desalted on C18 Cartridges (Empore SPE Cartridges C18 (standard density), bed I.D. 7 mm, volume 3 ml, Sigma), concentrated by vacuum centrifugation and reconstituted in 40 µl of 0.1% (v/v) formic acid (solvent A). The tryptic peptides were fractionated into fractions using high pH reverse-phase HPLC with an Agilent 300Extend C18 column (5 μm particles, 4.6 mm ID, 250 mm length) for LC‐MS/MS analysis. The gradient was comprised of an increase from 6% to 23% solvent B (0.1% formic acid in 98% acetonitrile) over 26 min, 23 to 35% in 8 min, and climbing to 80% in 3 min then holding at 80% for the last 3 min, all at a constant flow rate of 400 nL/min on an EASY‐nLC 1,000 UPLC system. The peptides were subjected to an NSI source followed by tandem mass spectrometry (MS/MS) in Q-Exactive HF (Thermo) coupled online to the UPLC. The resulting MS/MS data were processed using the MaxQuant search engine (v.1.5.3.17). The DAVID software ( https://david.ncifcrf.gov/ ) was used to test the statistical enrichment of the studied proteins in the Kyoto Encyclopedia of Genes and Genomes ( KEGG ) pathways. Pathways with P -values < 0.05 were considered significantly enriched. The STRING database ( http://www.string-db.org/ ) was employed for protein-protein interaction ( PPI ) analysis, and Cytoscape v3.7.1 Software was used for visualization ( Shannon, et al., 2003 ). To further validate the results of proteomic sequencing, we quantified the expression levels of the 18 selected proteins by PRM analysis using the remaining HSYF-Exo and ASYF-Exo samples (n = 3, each group). The protein extraction and trypsin digestion methods were consistent with those employed in the proteomic analysis experiment. Every target protein was monitored using several peptides from a target inclusion list. Subsequently, peptide samples were subjected to analysis using the TripleTOF 5600+ mass spectrometer in conjunction with the Eksigent microLC system (AB SCIEX, Framingham, MA). The resulting MS/MS data were processed using the Proteinpilot V4.5 search engine, followed by further analysis using the Skyline package( MacLean, et al., 2010 ). This comprehensive approach allowed us to obtain PRM spectrum files and quantitative information about the proteins. Separation and cultivation of GCs were performed as previously described ( Gilbert, et al., 1977 ). Briefly, 10 to 15 HSYFs, which have a thicker GC layer in egg-laying geese ( Yang et al., 2019 ), were collected under aseptic conditions and rinsed with Ca2 + - and Mg2 + -free PBS to remove the yolks and vitelline membrane as fully as possible. Thereafter, the tissue was cut into 1–2 mm 3 -sized blocks, digested with 1 mg/ml collagenase (Type II; Sigma Chemical Company, St. Louis, MO) at 37°C for 5 min, and filtered through a 200 µm nylon filter. Filtered suspensions were centrifuged twice at 67 × g for 5 min. Pellets were washed with DMEM media to remove the remaining collagenase and cell debris, resuspended in three mL of 50% Percoll, and centrifuged at 421 × g for 15 min. After centrifugation, the cell layer was aspirated. GCs suspensions were prepared by adding a pre-configured DMEM media (10% fetal bovine serum, 5 µg/mL transferrin, 2 mmol/L L-glutamine, 1.75 mM HEPES, 10 µg/mL insulin) and counted following staining with 0.1% trypan blue. Only samples with > 90% cell survival were used for subsequent experiments. Cells were allowed to adhere in tissue culture flasks for 24 h before experimental utilization. After that, purity was initially determined by an indirect immunofluorescence assay ( IFA ) to detect follicle GCs specific receptor, follicle-stimulating hormone receptor ( FSHR ) expression using anti-FSHR antibodies (1:500, Proteintech, Rosemont, IL, United States, L594-22665). For exosome uptake analysis, exosomes isolated from ASYFs and were labeled with the red membrane dye CM‐Dil (1, 1′-dioctadecyl-3, 3, 3′, 3′-tetramethylindocarbocyanine perchlorate) (Thermo C7000, USA) and then incubated 30 min in dark at 37°C. The working concentration of CM‐Dil was 1 mM. Approximately 2.5×10 6 cells/well GCs were seeded in six-well plates, the cells were starved for 12 h and then labeled with a green fluorescent dye CM-Dio (3, 3′-dioctadecyloxacarbocyanine perchlorate) (Thermo V22886) according to the manufacturer's protocol. Then, the cells were washed twice with PBS and incubated with labeled exosomes (100 µg/mL) for 24 h at 37°C, 5% CO 2 . The uptake of exosomes was observed by fluorescence microscopy (Leica, Weltzlar, Germany). Cell viability was measured using a CCK-8 assay kit (Vazyme A311-01, Nanjing, China), following the manufacturer's instructions. GCs were cultured in 100 mL of cell culture medium on a 96-well plate at a density of 10 5 cells per well. After 24 h of treatment with different concentrations of exosomes, 10 mL of CCK-8 solution was added to each well, and the plate was incubated at 37°C for 2 h without light. Optical density was measured at 450 nm using a microplate reader. GCs mitochondrial membrane potential ( MMP ) levels were detected using the JC-1 assay kit (Beyotime C2003S, China). Cells were incubated with 500 μL of JC-1 staining working solution for 20 min at 37°C. The fluorescent dye-labeled cells were washed with PBS twice and images were captured using a fluorescence microscope. The complete open reading frame ( ORF ) of goose HMOX1 was inserted into the corresponding pEGFP-N1 vector site using ClonExpress II One Step Cloning Kit according to the manufacturer's instructions (Vazyme, Nanjing, China). The PCR primers of HMOX1 are F: CTACTCGAGGCCACCATGGAAACTTCCCAGAAACACAGCT, R: CCGAAGCTTCCATAGCAAACAAGCCCACAGCG. The DNA sequencing of goose HMOX1 was verified by TSINGKE Biological Company (Beijing, China). GCs (2.5 × 10 6 cells/well) were seeded in 6-well plates and incubated until they were 80 to 90% confluent, were transfected with plasmid pEGFP-N1, pEGFP-N1-HMOX1 using TransIT-X2 (MirusBio MIR 6000) according to the manufacturer's instructions, respectively (Invitrogen, Carlsbad, CA). Then cells were treated with Znpp (2 μg/mL, MedChem Express HY-101193, USA) for 24 h for RT-qPCR and Western blotting analysis. RNA was isolated and purified from samples using an RNA Extraction Kit (Takara Bio, Tokyo, Japan) following the manufacturer's instructions. Then, approximately 600 to 800 ng of total RNA was reverse transcribed to cDNA using PrimeScript RT Reagent Kit (Takara Bio) with a gDNA Eraser. The reaction mixture consisted of 2× Super Real Pre-Mix Plus (Takara Bio), 2 mL of cDNA, 0.5 mL of sense primer (10 mmol/L), 0.5 mL of antisense primer (10 mmol/L), and 7 mL of RNAse-free water. The reaction was detected using a real-time PCR system (Mx3005P; Agilent, Santa Clara, CA), and the reaction conditions were as follows: 95 ℃ for 5 min, and 40 cycles of 95 ℃ for 15 s, 60 ℃ for 15 s, and 72 ℃ for 30 s. The results were analyzed using the 2 −△△Ct method with β-actin as the reference gene. All primers were synthesized by TSINGKE Biological Company (Beijing, China), and the sequences are listed in Table 1 Table 1 Quantitative polymerase chain reaction primer sequences. Table 1 Name of primer Sequence (5′-3′) Accession number Amplicon length (bp) CYP19A F: GGCAATTCTGGTGACTCTTCT XM_048060228.1 119 R: GCAGAGGCTGCCTTTCTATT HSD17B F: AAGGGGCAGGAATCATACA XM_013190199.2 117 R: GACGAGACCAGGACAAACAA FSHR F: TTAGGGCGGATCTTCGACAC XM_013192471.2 273 R: TGGTAAGGACAAATCTCAGTTCA GPX4 F: GCCATCTACGACTTCCACG XM_038169022.1 113 R: GGCGGTTTTCCCTCATTT COX2 F: CTGTAGACGCCCAAGAAGTAGA NC_011196.1 136 R: GATGGCTTTTAGGGTGAGGT NCOA4 F: GCCTCGCTTGCTGACTGGTTAC XM_013174872.2 119 R: TGGCTGGGTTCTGACACACATTTC FPN1 F: CACCTCTGCCAAGTTCCTCC XM_038182365.1 190 R: TGTCCACCCAGTCTCCAATAA FTH1 F: AATGACCCGCATCTGTGTGACTTC XM_048050004.1 135 R: GAGGTACTCTGCCATGCCATACTTG VDAC3 F: CTATGGATTTGGAATGGTC XM_013194405.2 165 R: CACTTCTGGAGGAATGTAA HMOX1 F: AGAGGACAAGCAGGAGCCAGAC XM_013181078.2 135 R: GGAAGGTGAAGGTCCAGCATGTTC ATG5 F: TACCCGATTGGTTTGCTCT XM_048079187.1 228 R: TACCCGATTGGTTTGCTCT LC3B F: CCTGGTGCCAGATCACGTCAAC XM_013197595.1 168 R: AAGCCGTCCTCGTCCTTCTCG PCBP2 F: GCCTATACCATTCAAGGACAGT XM 048056349.1 188 R: TTCGTTAGCGGAGGATTTT β-actin F: GAGAAATTGTCCGTGACATCA XM_013174886.1 152 R: CCTGAACCTCTCATTGCCA Quantitative polymerase chain reaction primer sequences. Tissues, exosomes, and GCs were lysed with Cold RIPA laced with proteinase inhibitors for 20min, and the protein concentration of tissues or exosomes or GCs was measured using the BCA method (Beyotime). Approximately 20 mg of total protein were subjected to 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis, and the separated proteins were transferred onto 0.45 mm polyvinylidene fluoride (PVDF) membranes, which were incubated in 5% bovine serum albumin for 1 h at room temperature. To detect exosome protein markers, the membranes were then incubated with primary antibodies ( Sung, et al., 2020 ) including rabbit anti‐Tsg101 (1:1000, Abmart T55985S), CD63 (1:1000, Abcam, ab134045), CD9 (1:1000, ABclonal A1703), β-actin (1:2000, Abclonal AC026). To analyze the expression of the key protein, the membranes were then incubated with primary antibodies including rabbit anti‐GPX4(1:100, Abmart PHY7049), HMOX1(1:1000, Bioss-bs-2075R), FTH1 (1:500, Abcam ab194947), ATG5 (1:500, Affinity DF6010), PCBP2 (1:500, Abclonal A2531), FPN1 (SLC40A1) (1:500, Abclonal A14885), LC3B (1:500, Abclonal A19665), β-actin (1:2000, Abclonal, AC026). After the excess primary antibodies on the PVDF membranes were cleaned, the membranes were incubated with secondary goat anti-rabbit antibodies which conjugated with horseradish peroxidase ( HRP ) (1:5000, Thermo Pierce 31210) at 4 ℃ overnight. After the excess secondary antibodies on the PVDF membranes were cleaned, the membranes were incubated with a chemiluminescence detection kit (Ncm-ECL Ultra, Soochow, China) following the manufacturer's instructions and were detected with a chemiluminescence detector (Tanon5200 Shanghai, China). ImageJ software was used to measure the band intensities, and the findings were normalized to β‐actin. At least three times each of the experiments were repeated. The SPSS 22.0 program (SPSS Inc., Chicago, IL) was used to analyze the data using one‐way ANOVA, post hoc Dunnett's test, and independent samples t‐test, and the results are reported as the mean ± standard deviation. P < 0.05 was considered a statistically significant difference. Graphs were constructed using the GraphPad Prism 6.0 software.

In this study, we found that the GC layers of ASYFs were shed, and a large number of GCs underwent ferroptosis compared to HSYFs. The FF closely surrounded GCs, a large number of exosomes were present in goose FF, and the exosomes of ASYF-FF could enter GCs through endocytosis and induce ferroptosis of GCs to participate in the regulation of follicular atresia compared to the HSYFs group. 1615 DEPs were obtained from FF-derived exosomes of HSYFs and ASYFs, which were mainly enrichment in protein transport and ferroptosis pathways. Among them, exosomal HMOX1 was found to initiate the ferroptosis pathway, promoting ferritinophagy to increase NCOA4 expression and Fe 2+ release. This led to the downregulation of GPX4, FTH1, and FPN1 expression, hindering the efflux of Fe 2+ and resulting in a large number of GCs undergoing ferroptosis in ASYFs. This study revealed a new mechanism of FF-derived exosomal HMOX1 mediating GCs ferroptosis and participating in follicular atresia, which will provide a new solution for improving the egg-laying performance of geese. At present, the mechanism of apoptosis and autophagy regulating follicular atresia in geese has been widely accepted ( Hou, et al., 2023 ). However, whether ferroptosis is involved in the process of follicular atresia in geese is still unclear. In this study, we found that ferroptosis was involved in the process of follicular atresia in female geese by comparing morphological and histological differences of HSYFs and ASYFs from the same ovary. Ferroptosis is a form of programmed cell death caused by iron-dependent lipid peroxidation and ROS accumulation, which is characterized by the reduction or disappearance of mitochondrial cristae, rupture of the mitochondrial outer membrane, and concentration of the mitochondrial membrane ( Cepelak, et al., 2020 ). Iron overload can lead to an imbalance of cellular processes, cellular dysfunction, apoptosis or necrosis, crude lipid peroxidation, protein, and DNA damage ( Imlay, et al., 1988 ; Papanikolaou and Pantopoulos, 2005 ). The investigation into the changes of trace elements during the atresia of ovarian follicles in goats has shown that the levels of iron ions in atresia follicles increase significantly ( Bhardwaj and Sharma, 2011 ). It was also found that an increase in iron ion content in FF could cause ferroptosis in follicular GCs in mice ( Ni, et al., 2022a ). In this study, compared with HSYFs, the expression of ferroptosis-related genes GPX4, FPN1 , and FTH1 were significantly decreased and the expression of COX2, HMOX1, PROM2, NCOA4, TXNIP , Fe 2+ level, and MDA content were significantly upregulated in ASYFs. The levels of ROS, MDA, and oxidation undergo a significant increase in ASYFs, suggesting an explosion in oxidative stress levels. At the ultrastructural level, mitochondria coagulated or swelled, membrane density increased, ridge decreased, and the outer membrane ruptured. Exosomes play an important role in intercellular communication ( Bischoff, et al., 2022 ). In mammals, FF-derived exosomes were found to participate widely in the development of follicles by regulating the signaling pathways associated with meiosis, ovulation, or follicular maturation, exosomes can affect GCs proliferation, gamete formation, oocyte maturation, embryonic development, and fertilization ( Mobarak, et al., 2019 ; Ranjbaran, et al., 2019 ; Esfandyari, et al., 2021 ). In this study, exosomes were first successfully extracted from goose FF and it was also found that FF-derived exosomes of HSYFs and ASYFs during the same egg-laying stage had different effects on GCs activity, ROS content, MMP, and MDA levels. These results suggest that differences in exosome composition could induce ferroptosis in GCs, a process critical for determining the ultimate fate of HSYFs, whether they proceed to hierarchical development or undergo atresia. Exosomes can transport lipids, nucleic acids, and other components to participate in the ferroptosis lipid peroxidation reaction of receptor cells ( Lu, et al., 2019 ; Zhang, et al., 2020 ; Lin, et al., 2022 ). Hence, it is postulated that the exosomes present in goose FF potentially transport bioactive compounds to GCs, thereby triggering ferroptosis in GCs and consequently exerting a pivotal regulatory function in follicular atresia. In this study, proteomic analysis revealed 1615 DEPs carried by the FF-derived exosomes of both HSYFs and ASYFs. Among them, HMOX1 was highly expressed in the ASYF-FF exosomes and enriched in the top ferroptosis pathway. HMOX, as an inducible enzyme, might exhibit cytoprotection by converting pro-oxidative hemoproteins and heme into antioxidants bilirubin and biliverdin, or conversely, exacerbate oxidative stress by releasing Fe 2+ and CO ( Schipper, et al., 2009 ; Tang, et al., 2021 ). Initially, HMOX1 was perceived with antioxidant and anti-inflammatory properties, safeguarding cells against ROS assaults. HMOX1, as an important antioxidant enzyme, could decompose heme into Fe 2+ , CO, and biliverdin to reduce the toxicity of heme in mammals and chickens ( Zhu, et al., 2008 ). HMOX1 has been demonstrated to primarily localize to the cytoplasm and shuttle between subcellular compartments, including caveolae, mitochondria, and the nucleus, which mainly play a key role in whole body iron recycling/homeostasis ( Dunn, et al., 2014 ). Increased the expression of HMOX1 in follicles was found could effectively improve oxidative stress in the ovaries of aging hens ( Liu, et al., 2018 ). However, recent investigations have revealed that the upregulation of HMOX1 can facilitate oxidation and trigger ferroptosis within cells, primarily through the augmentation of iron accumulation and lipid peroxidation ( Hassannia, et al., 2018 ; Han, et al., 2022 ; Zhang, et al., 2022 ). These imply that HMOX1 is a double-edged sword that can provide protection or enhance vulnerability ( Ahmad, et al., 2006 ; Yuan, et al., 2008 ). In this study, we found that FF-derived exosomal HMOX1 incited the ferroptosis pathway, promoted ferritinophagy-related ATG5, LC3II, and NCOA4 expression, and reduced the expression of FTH1, GPX4, PCPB2, and FPN1, leading to intracellular Fe 2+ accumulation, MDA surge, and GCs ferroptosis. It was also discovered that the expression of FPN1 was significantly reduced, PCBP was significantly increased, and abnormal metabolism of GPX4 in early atretic follicles of pigs, suggesting that iron accumulation initiates ferroptosis plays an important role in the regulation of follicular atresia ( Zhang et al., 2018 ). FTH1 is capable of storing high concentrations of Fe 2+ to prevent contact with substrates that produce reactive oxygen species, which is an important protein in iron-dependent ferroptosis ( Flórez and Alborzinia, 2021 ). FTH1 mRNA expression in the laying goose ovary was found 13 times greater than in the prelaying goose ovary and considered to mediate the reproductive processes of geese ( Kang, et al., 2015 ). Interference with the FTH1 would lead to the shrinkage of GCs in goose follicles, an increase in the apoptosis rate of GCs, a significant increase in the number of dead GCs, promotion of the accumulation of ROS, and harm to MMP ( Yuan, et al., 2008 ; Jiang, et al., 2023 ). In this study, the expression level of FTH1 was significantly decreased in ASYFs compared to HSYFs group. However, in a previous study, the expression levels of FTH1 in atretic follicles of geese were 2.1 times and 8 times higher than those of health SWFs and HSYFs, respectively. This difference may be related to the different stages of the selected atretic follicle or FTH1 was degraded in ferritinophagy during GCs ferroptosis ( Yu, et al., 2023 ). NCOA4 is a transport receptor for FTH1 degradation and is upregulated in ASYFs, in which C-terminal domain bound to FTH1 and delivered to the autophagosome ( Mancias, et al., 2014 ). Autophagic degradation of FTH1 occurred, with a significant increase in free Fe 2+ and ferroptosis in GCs ( Fang, et al., 2021 ). In conclusion, exosomes extracted from ASYF-FF in geese could transport excessive HMOX1 into GCs via endocytosis, thereby activating the ferroptosis pathway. This process decreased FTH1 expression and released Fe 2+ , as well as the reduction of FPN1 expression to impede Fe 2+ efflux. Consequently, an accumulation of Fe 2+ occurs within GCs, leading to the induction of ferroptosis and subsequent follicular atresia ( Figure 10D ).

In summary, the GCs of ASYFs exhibited a significant occurrence of ferroptosis in contrast to HSYFs in geese. The goose FF contained abundant exosomes, and the exosomes extracted from ASYF-FF and HSYF-FF displayed 1615 differentially expressed proteins, primarily enriched in the protein transport and ferroptosis signaling pathways. ASYFs FF-derived exosomes could enter GCs through endocytosis and transport excessive exosomal HMOX1, which in turn promoted the expression level of ATG5, LC3II, and NCOA4 and decreased the expression of GPX4, FTH1, PCBP2, and FPN1 in the ferroptosis pathway, resulting in an increase of Fe 2+ levels, MDA content, and mitochondria damage within the GCs. Our study contributes novel insights into the mechanisms underlying FF-derived exosomes mediated GCs ferroptosis-induced follicular atresia in geese.

The low egg production of geese, averaging 40–80 eggs per year, has seriously hindered the development of the goose industry ( Yao, et al., 2019 ). Therefore, improving egg production has become one of the urgent problems in the current goose industry. The development of poultry follicles strictly follows a hierarchical development system, and the egg-laying performance is mainly determined by the number of follicles that reach the hierarchical development stage ( Johnson, 2015 ). While in the ovaries of birds, only about 1‰ of follicles can eventually mature and ovulate, and most follicles stop growing during development and gradually degrade into atresia ( Onagbesan, et al., 2009 ). Among these different types of follicles, small yellow follicles (SYFs) are the key selection stage access to hierarchical follicles ( Lovell, et al., 2003 ). The number of atresia small yellow follicles ( ASYFs ) directly affects the egg production rate of poultry. An increase in the number of healthy small yellow follicles ( HSYFs ) entering graded development means a longer egg-laying period and an increase in egg production. Therefore, researching follicular atresia in geese holds important economic value and provides a medical reference for ovarian diseases. Poultry follicles are primarily composed of oocytes, a granulosa cell ( GC ) layer, and a thecal cell layer, which serve as the fundamental functional units of the ovary. Among these, GCs are the most abundant, specific, and functional cells within the follicle. Their proliferation, differentiation, and apoptosis play a crucial role in follicle development and determine follicular atresia. When the number of GCs undergoing apoptosis reaches 10%, it marks that the follicle is in atresia ( Sugimoto, et al., 1998 ; Matsuda, et al., 2012 ). Many studies indicated that oxidative stress is the main factor that induces mass mortality of GCs ( Freitas, et al., 2017 ; Saeed-Zidane, et al., 2017 ; Kunitomi, et al., 2020 ; Liu, et al., 2023 ). Ferroptosis is a recently discovered death form that primarily involves the accumulation of Fe 2+ , leading to lipid peroxidation of unsaturated fatty acids in cell membranes and excessive oxidative stress, which ultimately destroys selective permeability of the plasma membrane and subsequent cell death ( Stockwell, 2022 ). In humans, it has been observed that iron overload induced-oxidative stress in follicular fluid ( FF ) can trigger ferroptosis in GCs, impairing oocyte maturation and increasing the risk of infertility associated with endometriosis ( Ni, et al., 2022b ), and in mice, increased production of ROS in the ovaries, mediated by iron overload, might impair ovarian function and ultimately lead to gonadal failure ( Sze, et al., 2022 ). In pigs, it was found that ferroptosis may be involved earlier in the follicle atresia stage through the accumulation of iron molecules ( Zhang, et al., 2018 ). In chickens, the activation of ferroptosis signals increases oxidative damage in SWFs, and Rutin could alleviate the oxidative stress induced by ferroptosis and the damage to the GC layers in SWFs of aging chickens through the Nrf2/HO-1 pathway ( Wu, et al., 2023 ). Untargeted metabolomics uncovered that the ferroptosis pathway plays an important role in follicle selection during the development of chicken small yellow follicles ( Zhao, et al., 2023 ). Therefore, the anti-ferroptosis of GCs may contribute to follicles resisting oxidative stress and become a new target to improve follicular atresia. Finding key regulators that target to alleviate ferroptosis in GCs will aid in follicular development, relieve follicular atresia, and establish methods to improve egg production performance in geese. As the environment on which GCs depend for survival, FF contains abundant bioactive substances such as lipids, miRNA, mRNA, and proteins. The exosomes of FF, which carry miRNA, mRNA, and proteins, can communicate bidirectionally with GCs and oocytes to regulate GCs proliferation, steroid production, and oocyte maturation ( Collado-Fernandez, et al., 2012 ; Yuan, et al., 2021 ). For example, the exosomal miR-130b promotes oocyte maturation and the proliferation of cumulus and GCs during bovine oocyte development by targeting Smad family member 5 ( SMAD5 ) and mitogen and stress-activated kinase 1 ( MSK1 ) ( Sinha, et al., 2017 ). The exosomal miR-146b was significantly upregulated during follicular atresia and increased apoptosis of GCs by inhibiting CYP19A1 in pigs ( Li, et al., 2021 ). In human polycystic ovary syndrome, FF-derived exosomal S100-A9 protein significantly promotes inflammation and disrupts steroidogenesis via activation of the NF-κB pathway ( Li, et al., 2020 ). In more than 95% of the follicles that stop developing and occur atresia in in goose ovaries, the yolk is lost, the follicles gradually shrink, and the GCs apoptosis in large numbers of the atretic follicle ( Yang, et al., 2023 ). However, it remains uncertain whether the development of goose follicles is influenced by active substances mediated by FF-derived exosomes, which are implicated in the extensive mortality of GCs, ultimately leading to a high incidence of follicular atresia. In this study, laying geese were chosen as research subjects to investigate the differences in protein components of FF-derived exosomes between ASYFs and HSYFs, as well as the regulatory mechanism of key exosomal differentially expressed proteins in ferroptosis of GCs. The objective of this study was to uncover the role of the exosomes of ASYF-FF in the mass mortality of GCs. The results not only shed light on the physiological mechanism of follicular atresia in geese but also offer theoretical references for enhancing the egg-laying performance of geese.

The authors declare no conflicts of interest.