Abstract

   Ovarian aging is mainly characterized by a progressive decline in
   oocyte quantity and quality, which ultimately leads to female
   infertility. Various therapies have been established to cope with
   ovarian aging, among which exosome-based therapy is considered a
   promising strategy that can benefit ovarian functions via multiple
   pathways. Here, we isolated and characterized exosomes derived from
   ovarian follicular fluid and profiled the differential expression
   patterns of noncoding exosomal RNAs in young and aged women. Treatment
   with young mouse-derived exosomes efficiently rescued ovarian function
   in aged mice. The follicular fluid exosomes from young mice and
   miR-320-3p can also promote the proliferation of ovarian granulosa
   cells and improve mitochondrial function from old mice in vitro. The
   mechanism may be involve that exosomes transfer miR-320-3p to granulosa
   cells, and inhibit the expression of FOXQ1. Exosomes also can increase
   the number of primordial and growing follicles, and improve the
   developmental ability of oocytes in the old mice in vivo. And hnRNPA2B1
   controls miR-320-3p entry into exosomes. This work provides insights
   into the antiaging potential of follicular fluid-derived exosomes and
   the underlying molecular mechanisms, which may facilitate prevention of
   ovarian aging and an improvement in female fertility.

   Keywords: exosome, granulosa cells, miRNAs, ovarian aging, follicular
   fluid

Introduction

   The ovary enters the malfunctional aging phase earlier and faster than
   other organs and systems, and the aging process has attracted extensive
   attention in the field of reproduction with three-child policy
   implemented in 2021 in China. Ovarian aging is mainly characterized by
   a progressive decline in oocyte quantity and quality, which leads to
   impaired ovarian reserve, a reduced pregnancy rate and live birth rate,
   and ultimately to the loss of female fertility [[46]1, [47]2].
   According to a new report in Lancet, it is estimated that by 2050, 151
   countries, including China, will have an extremely declining fertility
   rate and a clear trend of advanced reproductive age due to improvements
   in women’s education levels, increased employment pressure, and easier
   access to contraception [[48]3]. Coupled with the rising infertility
   rate, the problem of sustaining and enhancing female reproductive
   health will be crucial in the years to come.

   Several clinical therapeutic strategies have been established to
   improve ovarian function. Primordial follicle in vitro activation
   provides a practical approach for acquiring fully competent oocytes in
   infertile patients [[49]4], but it is traumatic and its safety needs
   further verification. Recently, studies regarding stem cell therapies
   have revealed their potential in improving ovarian functions. Stem cell
   treatment rescues age-related ovarian damage [[50]5] and partially
   restores the fertility of premature ovarian failure patients [[51]6].
   Stem cells play their role via various mechanisms, among which the
   paracrine effect of extracellular vesicles, including exosomes, is
   considered a promising approach that can be applied in multiple fields.
   Exosomes are bilayer-lipid-membrane extracellular vesicles (EVs) with
   diameters ranging from 40 to 200 nm [[52]7, [53]8]. The typical
   biomarkers of exosomes are transmembrane tetrad proteins, TSG101, and
   Flotillin-1. Exosomes can be isolated from cell culture fluids and
   various body fluids, including urine [[54]9], semen [[55]10], saliva
   [[56]11], ascites [[57]12], blood [[58]13], cerebrospinal fluid
   [[59]14], and breast fluid [[60]15]. Because they carry a variety of
   biologically active substances, including DNA, RNA, and so on, exosomes
   serve as important mediators of intercellular communication, delivering
   intracellular contents from their cell of origin to target cells
   [[61]16]. Previous studies have shown that injection with exosomes
   isolated from young mouse serum can reverse the expression pattern of
   aging-associated molecules in the lungs and liver [[62]17].
   Importantly, intravenously injected extracellular vesicles from
   neonatal umbilical cord rejuvenated aged bone marrow–derived
   mesenchymal stem cells [[63]18]. Similarly, age-related cognitive
   decline can be restored after the systemic administration of young
   blood plasma into aged mice [[64]19]. However, whether exosomes from
   follicular fluid can rescue diminished ovarian function remains
   unknown.

   Bovine follicular fluid-derived exosomes contribute to improvements in
   the developmental capacity of oocytes [[65]20]. Equine follicular
   fluid-derived exosomes serve as key regulators in follicle maturation
   [[66]21]. Exosome-secreted miRNAs can regulate the maturation of
   oocytes and sperm [[67]22]. Wei et al. [[68]23] found that miRNAs are
   crucial for the implementation of exosome functions, and the lipid
   membrane structure of exosomes can act as a protective barrier to
   prevent RNase-induced miRNA degradation [[69]24]. Animal studies have
   shown that exosomal miRNA expression profiles in follicular fluid are
   significantly correlated with age. Compared with follicular fluid
   exosomes from young mares, exosomes from aged mares carried higher
   levels of miRNAs, eventually leading to impaired maturation of oocytes
   [[70]21, [71]25]. Sohel et al. [[72]26] found certain differences in
   exosomal miRNA expression patterns in follicular fluid containing
   mature and immature oocytes. Furthermore, high-throughput sequencing
   demonstrated differentially expressed miRNA profiles between young and
   aged women [[73]27]. All this evidence demonstrates that the follicular
   fluid exosomal miRNA expression profile is closely correlated with age
   and ovarian function, indicating the possible involvement of exosomes
   in ovarian aging; thus, the specific underlying mechanisms need to be
   further explored.

   In this study, we established an miRNA expression profile in human
   follicular fluid-derived exosomes, revealed the therapeutic potential
   of follicular fluid exosomes to counteract ovarian aging, identified
   the key regulator miR-320-3p, and unveiled the corresponding mechanism
   underlying miR-320-3p-mediated regulation of ovarian functions, thus
   providing powerful evidence for establishing a novel and promising
   strategy to improve ovarian function in aged women.

Results

Characteristics of follicular fluid-derived exosomes

   To investigate the characteristics of follicular fluid-derived
   exosomes, we isolated follicular fluid-derived exosomes (FF-exos) from
   young (≤ 30-year-old, Y-exos) and aged (> 38-year-old, O-exos) women
   ([74]Fig. S1A) and identified them by transmission electron microscopy
   (TEM), nanoparticle tracking analysis (NTA) and western blot analyses.
   The TEM results showed the circular membrane structure of isolated
   vesicles ([75]Fig. S1B). The number and size distribution of Y-exos and
   O-exos were determined by NTA, showing that the exosome diameters
   mainly ranged from 50 to 150 nm ([76]Fig. S1C). Moreover, no
   significant differences in terms of particle size and concentration
   were observed between Y-exos and O-exos ([77]Fig. S1D and [78]S1E). We
   further performed western blot analysis to examine exosome-specific
   markers to validate the enrichment of exosomes ([79]Fig. S1F).
   Collectively, these data demonstrated the main characteristics of
   Y-exos and O-exos and showed no significant difference between these
   two groups.

Y-exos can enhance functions of aged mouse-derived GCs in vitro

   To investigate whether Y-exos could be internalized by aged
   mouse-derived granulosa cells (GCs), we first isolated exosomes and GCs
   from young and aged mice, respectively. The purity of isolated GCs was
   determined by its specific marker FSHR ([80]Fig. S2A), while that of
   the extracted exosomes was evaluated by TEM, NTA and western blot
   analysis. On the basis of TEM imaging, these exosomes exhibited typical
   circular membrane vesicles ([81]Fig. S2B). The particle size
   distribution and concentration of the exosomes were assessed via NTA,
   and the exosomal size ranges from 50 to 150 nm ([82]Fig. S2C and
   [83]S2D). Western blot analysis showed that the exosomal surface
   markers TSG101 and FLOT1 could be detected in the exosome samples
   ([84]Fig. S2E). Next, the uptake of Y-exos by ovarian GCs of aged mice
   was observed in vitro. As shown in [85]Fig. S2F, PKH67-labeled Y-exos
   could be observed in the GCs at 24 h after co-culture, suggesting that
   Y-exos can be internalized by GCs.

   To examine the effects of Y-exos on GC functions, ovarian GCs of aged
   mice were exposed to Y-exos at different concentrations (2.5, 5, 10
   μg/mL) for 24 h. Y-exos markedly enhanced cell viability, and this
   effect appeared to be concentration dependent (P < 0.05, [86]Fig. 1A).
   Ultimately, we chose 5 μg/mL Y-exos for subsequent experiments. Annexin
   V staining indicated a significant decrease in apoptosis in the Y-exos
   group (P < 0.01, [87]Fig. 1B). We also observed a downregulation in
   cleaved caspase3, cleaved caspase9, and BAX protein levels and a
   substantial upregulation in BCL2 protein levels in the Y-exos group
   (P < 0.05, [88]Fig. 1C and [89]1D). In terms of mitochondrial function
   in GCs, we found higher mitochondrial membrane potential (MMP), ATP
   levels and mitochondrial DNA (mtDNA) copy numbers after Y-exos
   treatment ([90]Fig. 1E–H). Collectively, these results indicate that
   Y-exos can promote proliferation and improve mitochondrial functions in
   aged mouse-derived GCs.

Figure 1.

   [91]Figure 1.
   [92]Open in a new tab

   Y-exos promote proliferative capacity of aged mice derived GCs in
   vitro. (A) Cell viability analyzed by CCK8 assay. (B) Apoptotic rate of
   aged mice derived GCs with or without Y-exos treatment was determined
   by flow cytometry. (C, D) Cleaved caspase-3, cleaved caspase-9, BAX,
   and BCL2 protein expression were analyzed by western blot. (E, F) MMP
   assessed by JC-1 staining. Scale bar, 100 μm. (G) The relative
   expression of cell mtDNA. (H) ATP concentration in GCs was analyzed.

Y-exos can rescue age-related ovarian dysfunction in vivo

   To determine whether Y-exos treatment can rescue age-related ovarian
   dysfunction in vivo, forty-week-old mice received unilateral ovarian
   bursa local injection with Y-exos ([93]Fig. 2A). Fluorescence signals
   were observed in the liver, kidney and ovary, with the highest
   fluorescence intensity appearing in the ovary on both sides ([94]Fig.
   2B and [95]2C), suggesting that exosomes can be transferred to mouse
   ovaries, the liver and the kidney after ovarian bursa local injection.
   Intriguingly, we found that unilateral injection did not cause a
   difference in the volume of the ovaries on both sides within the same
   mouse ([96]Fig. 2B).

Figure 2.

   [97]Figure 2.
   [98]Open in a new tab

   Y-exos can rescue age-related ovarian dysfunction in vivo. (A)
   Schematic illustration of ovarian bursa injection of follicular fluid
   exosomes from young mice (Y-exos, right) and PBS (left) into old mice
   for 3 weeks. (B) Representative fluorescence distribution imaging of
   PKH67-labeled Y-exos in different organs of mice after 48 h and 72 h of
   injection. (C) Quantitation of fluorescence intensities in different
   organs after 48 h and 72 h of injection. (D) Ovaries at 3 weeks after
   injection. Isolated ovaries from control (Ctrl) versus Y-exos
   treatment. (E) Comparison of ovarian weight between control and Y-exos
   treatment (n = 4). (F) Representative image of MII oocytes with normal
   (Norm) or aberrant (Abn) spindles in control and Y-exos treatment
   groups. Scale bar, 100 μm. (G) Percentage of aberrant spindles in
   control and Y-exos treatment groups. *, P < 0.05. (H) Percentage of
   embryo development between two groups. Ctrl, n = 41, Y-exos, n = 42.
   ****, P < 0.0001. (I) Representative ovarian histology images and
   distribution of different stages of follicles with or without Y-exos
   treatment. Scale bar, 100 μm. *, P < 0.05; **, P < 0.01;
   ****, P < 0.0001.

   We next observed the effects of Y-exos on mouse ovaries in vivo. The
   ovarian volume and weight after Y-exos treatment were increased
   compared to those in the control group (P < 0.05, [99]Fig. 2D and
   2[100]E), but no difference was observed between the injected side and
   the noninjected side ([101]Fig. 2D and [102]2E). TUNEL assay results
   indicated a decreased apoptotic rate in the Y-exos group ([103]Fig. S3A
   and [104]S3B). We also analyzed the expression levels of aging-related
   signaling factors, such as Mmp9, P53, Il-1α, P21, Il-8, Mmp1, Ccl20 and
   P16 via real-time qPCR. As shown in [105]Fig. S3C, these factors were
   efficiently downregulated in the Y-exos-treated ovaries, while the
   expression levels of telomerase-related genes, including Terf2, Tep1,
   Men1, Mre11a, Tert, and Tnks, increased significantly ([106]Fig. S3D).
   We next performed TEM analysis to demonstrate the ultrastructural
   change in ovaries after exosome treatment. No significant changes in
   either mitochondrial density or abnormal mitochondrial number were
   observed ([107]Fig. S3E and [108]S3F). However, the number of lipid
   drops in GCs was found to be significantly decreased in the
   Y-exos-treated group ([109]Fig. S3E and [110]S3G). Taken together, our
   results suggest that Y-exos treatment can partially rescue age-related
   ovarian dysfunction.

   To further evaluate the effects of Y-exos on ovarian function, we
   collected GCs and examined the changes in mitochondrial function. After
   Y-exos treatment, both MMP and intracellular ATP production were
   efficiently improved ([111]Fig. S3H and [112]S3I), while mtDNA copy
   number exhibited a slight elevation ([113]Fig. S3J), demonstrating that
   Y-exos treatment can enhance mitochondrial functions in aged mice.

   Next, MII oocytes were collected for oocyte quality analysis.
   Immunofluorescence staining of β-tubulin in MII oocytes showed a lower
   rate of aberrant spindles in the Y-exos-treated group than in the
   control group ([114]Fig. 2F and [115]2G), suggesting that Y-exos
   treatment can alleviate age-related meiosis disorder to a certain
   extent. We then performed in vitro fertilization (IVF) to determine the
   impact of Y-exos on oocyte developmental potential. Although the 2-cell
   rate did not exhibit an obvious change after Y-exos treatment
   (89.92 ± 0.06 vs. 92.00 ± 0.10), the blastocyst rate in the Y-exos
   group was significantly improved compared to that in the control group
   (66.98 ± 0.30 vs. 30.21 ± 0.19) ([116]Fig. 2H), indicating that the
   developmental competence of MII oocytes was improved after Y-exos
   administration. In addition, in the Y-exos group, we found that
   follicles at different developmental stages were obviously increased,
   while atretic follicles were reduced ([117]Fig. 2I). Surprisingly, the
   antral follicle count in the Y-exos-treated group had a nearly 3-fold
   increase compared to that in the control group, suggesting that Y-exos
   treatment largely improved follicular development ([118]Fig. 2I). There
   were no difference in the number of female (P ＞ 0.05, [119]Fig.S3K) in
   the Y-exos group, however, the pups per female have an increasing trend
   (P ＞ 0.05, [120]Fig. S3L).

miRNA sequencing revealed differentially expressed miRNAs in follicular
fluid-derived exosomes between young and aged women

   To further investigate the mechanism underlying the beneficial effects
   of Y-exos in the ovaries of aged, we analyzed the miRNA transcriptome
   of human Y-exos and O-exos via high-throughput sequencing analysis. In
   Y-exos and O-exos, 67.87% of reads were aligned to miRNAs ([121]Fig.
   3A), and a total of 637 miRNAs were identified ([122]Fig. 3B). Among
   all the identified miRNAs, 12 miRNAs exhibited significantly different
   expression levels between Y-exos and O-exos, with 7 miRNAs upregulated
   and 5 miRNAs downregulated (P < 0.05, [123]Fig. 3C). Agarose gel
   electrophoresis and Sanger sequencing were first performed to verify
   the specificity of the miRNA primers ([124]Fig. 3D), and then, these
   verified primers were used for subsequent RT‒qPCR analysis to validate
   the sequencing data. As shown in [125]Fig. 3E, miR-320a-3p was markedly
   upregulated in the Y-exos group and had a significantly higher level
   than other miRNAs, including miR-10399-5p and miR-483-5p. Therefore, we
   speculate that miR-320a-3p may play a key role in exosome-mediated
   effects on ovarian functions.

Figure 3.

   [126]Figure 3.
   [127]Open in a new tab

   Noncoding RNA sequencing revealed differentially expressed miRNAs in
   follicular fluid-derived exosomes between young and old women. (A) Pie
   chart representation of distribution of small RNA categories in Y-exos
   and O-exos. (B) Volcano plot showing the different expression patterns
   of miRNAs in Y-exos and O-exos. (C) Heatmap of differentially expressed
   miRNAs in Y-exos and O-exos. (D) The specificity of miRNA primers was
   examined by 2% agarose gel electrophoresis and Sanger sequencing. (E)
   The expression of seven differentially expressed miRNAs in a larger
   cohort (n = 36) were examined by RT-qPCR. *, P < 0.05; **, P < 0.01.

miR-320-3p enhances mouse GCs proliferation and mitochondrial functions in
vitro

   To examine whether exosomal miR-320-3p is responsible for the
   exosome-mediated effects, we transfected GCs with miR-320-3p mimic and
   confirmed the transfection efficiency via RT‒qPCR ([128]Fig. 4A).
   miR-320-3p mimic enhanced cell viability (P < 0.05, [129]Fig. 4B), and
   decreased cell apoptosis (P < 0.05, [130]Fig. 4C) and downregulation
   cleaved caspase3, cleaved caspase9, BAX expression and upregulation
   BCL2 protein levels in GCs (P < 0.05, [131]Fig. 4D), whereas the
   miR-320-3p inhibitor had the opposite effect on GCs ([132]Fig. 4B,
   [133]4C and [134]4D). These results suggest that miR-320-3p can promote
   proliferation and enhance resistance against apoptosis in GCs. We next
   detected the changes in mitochondrial functions upon miR-320-3p
   overexpression or inhibition and found that miR-320-3p overexpression
   improved MMP, enhanced mtDNA expression and promoted ATP production
   ([135]Fig. 4E–G). These results suggest that miR-320-3p can enhance GCs
   proliferation and mitochondrial functions in vitro.

Figure 4.

   [136]Figure 4.
   [137]Open in a new tab

   miR-320-3p enhances mouse GCs proliferation and mitochondrial functions
   in vitro.(A) RT-qPCR of miR-320-3p in cells transfected with control
   and miR-320-3p mimics. *, P < 0.05. (B) Cell viability analyzed by CCK8
   assay in miR-320-3p mimics, miR-320-3p inhibitor or control. (C)
   Apoptotic rate of aged mice derived GCs was determined by flow
   cytometry. (D) Cleaved caspase-3, cleaved caspase-9, BAX, and BCL2
   protein expression were analyzed by western blot. (E) MMP assessed by
   JC-1 staining. Scale bar, 100 μm. (F) The relative expression of cell
   mtDNA. (G) ATP concentration in GCs was determined.

Follicular fluid-derived exosomes transfer miR-320-3p to GCs derived from
aged mice

   Since the pro-proliferation effect of Y-exos and miR-320-3p had been
   verified in vitro, we next tried to determine whether miR-320-3p
   mediates the Y-exos-induced effects in GCs. We first isolated exosomes
   containing Cy3-labeled miR-320-3p from GCs of young mice and then
   cocultured them with GCs from aged mice. After co-culturing for 24 h,
   the Cy3-labeled miR-320-3p fluorescence signal was observed in GCs
   ([138]Fig. 5A), suggesting that exosomal miR-320-3p was absorbed by
   GCs. Then, we found that the level of miR-320-3p increased as the
   number of exosomes increased (P < 0.0001, [139]Fig. 5B). Co-culture
   with Y-exos increased the miR-320-3p level in GCs from aged mice
   (P < 0.05, [140]Fig. 5C), further validating that miR-320-3p can be
   transferred to GCs via exosomes. By overexpressing miR-320-3p in
   cultured GCs, we found that the miR-320-3p levels in exosomes were
   increased (P < 0.05, [141]Fig. 5D), and the miR-320-3p levels in
   exosomes exhibited a positive correlation with those in GCs (P < 0.05,
   [142]Fig. 5E). Moreover, the level of miR-320-3p in GCs was further
   increased after co-culture with exosomes from GCs overexpressing
   miR-320-3p ([143]Fig. 5F).

Figure 5.

   [144]Figure 5.
   [145]Open in a new tab

   Follicular fluid-derived exosomes transfer miR-320-3p to aged mice
   derived GCs and miR-320a-3p directly binds to Foxq1 to regulate GCs
   proliferation. (A) Existence of Cy3-labeled miR-320-3p in aged
   mouse-derived GCs, which were co-culture with Y-exos isolated from
   young mice derived GCs transfected with Cy3-labeled miR-320-3p mimics.
   Scale bar, 100 μm. (B) miR-320-3p level measured with different amount
   of young follicular fluid exosomes. *, P < 0.05. (C) Relative level of
   miR-320-3p in aged mice derived GCs co-culture with Y-exos. ****, P <
   0.0001. (D, E) Coordinated changes of miR-320-3p level in the cultured
   young mice derived GCs and their secreted exosomes. Gray bars represent
   the control group, red bars represent the miR-320-3p mimics group. Line
   represents linear regression. *, P < 0.05; ***, P < 0.001. (F) Relative
   level of miR-320-3p in aged mice derived GCs co-culture with Y-exos
   isolated from young mice derived GCs transfected with miR-320-3p
   mimics. (G) Differentially expressed genes upon miR-320a-3p
   overexpression profiled by RNA-seq were illustrated by the volcano
   plot. (H) KEGG pathway enrichment analysis of possible miR-320a-3p
   target genes. (I) Foxq1 mRNA quantitation by RT-qPCR in COV434 cells
   line after transfection with miR-320a-3p mimics, inhibitor or control.
   ***, P < 0.001; ****, P < 0.0001. (J) Schematic representation of
   putative miR-320a-3p binding sites on 3ʹUTR of Foxq1 in human, mouse
   and rat. (K) COV434 cells line transfected with wild-type or mutated
   Foxq1 3'UTR luciferase constructs and control mimic or miR-320a-3p
   mimics. ***, P < 0.001.

miR-320a-3p directly binds to Foxq1 to regulate GCs proliferation

   To determine which genes are under the control of miR-320-3p, we
   performed RNA-seq analysis to identify the differentially expressed
   genes upon miR-320a-3p overexpression, and 56 genes were identified as
   downregulated genes (defined as fold change ≥1, P < 0.05, [146]Fig. 5G)
   in the miR-320a-3p mimic group. We then performed KEGG enrichment
   analysis to further identify the possible targets of miR-320a-3p. Among
   all the enriched KEGG terms, “longevity regulating pathway-multiple
   species” and “longevity regulating pathway” were strongly correlated
   with the phenotype that we characterized in previous results ([147]Fig.
   5H). Combined with the results of the functional screening of
   differentially expressed genes, we eventually identified the following
   five genes as possible targets of miR-320a-3p that may contribute to
   the miR-320a-3p-induced effects: forkhead box Q1 (Foxq1), ribosomal
   protein S6 kinase B2 (RPS6KB2), protein phosphatase 2 regulatory
   subunit B’beta (PPP2R5B), nuclear receptor subfamily 1 group H member 2
   (NR1H2), and insulin-like growth factor binding protein 5 (IGFBP5).
   Then, we analyzed the mRNA expression levels of the above five genes in
   the miR-320a-3p mimic and miR-320a-3p inhibitor groups and found that
   these candidate genes were downregulated in the miR-320a-3p mimic group
   and upregulated in the miR-320a-3p inhibitor group ([148]Figs. 5I and
   [149]S4A). However, only Foxq1, PPP2R5B, and IGFBP5 were recognized as
   biological targets of miR-320a-3p by TargetScan. Subsequently, a dual
   luciferase reporter assay was performed to validate the interaction
   between miR-320a-3p and its target transcripts ([150]Fig. 5J and
   [151]5K, [152]S4B, S4C and [153]S4D, S4E). Considering the sequence
   conservation across species and the functions of these candidate genes,
   we selected Foxq1 as the target of miR-320a-3p for further validation
   studies.

miR-320a-3p overexpression promotes GCs proliferation by mediating FOXQ1
downregulation

   To examine whether FOXQ1 regulates proliferation, FOXQ1 was knocked
   down in aged mouse-derived GCs ([154]Fig. S5A and [155]S5B), resulting
   in a significant increase in the viability of old GCs (P < 0.05,
   [156]Fig. S5C). We then examined the apoptosis level and the protein
   expression levels of cleaved caspase3, cleaved caspase9, BAX, and BCL2.
   The rate of apoptosis was decreased (P < 0.05, [157]Fig. S5D), with
   cleaved caspase3, cleaved caspase9 and BAX downregulated, and BCL2
   upregulated in the Foxq1-siRNA group (P < 0.05, [158]Fig. S5E). FOXQ1
   knockdown increased MMP levels, enhanced ATP production and elevated
   cell mtDNA levels (P < 0.05, [159]Fig. S5F–[160]S5H). These results
   suggest that FOXQ1 mediates the pro-proliferative and anti-apoptotic
   effects of miR-320-3p in GCs.

   We next analyzed whether miR-320-3p regulates GCs functions by
   targeting Foxq1. The overexpression efficiency was first determined by
   western blot and RT‒qPCR analyses ([161]Fig. 6A). Then, we analyzed
   whether FOXQ1 overexpression could at least partially impede the
   miR-320-3p-induced effects in GCs. We found that FOXQ1 overexpression
   disturbed the pro-proliferative and anti-apoptotic effects ([162]Fig.
   6B–E) and hindered the enhancement of mitochondrial functions mediated
   by miR-320-3p overexpression ([163]Fig. 6F–I). These results
   demonstrate that the effects of miR-320-3p in GCs are achieved by
   targeting the Foxq1 transcript.

Figure 6.

   [164]Figure 6.
   [165]Open in a new tab

   miR-320a-3p overexpression promotes GCs proliferation by mediating
   FOXQ1 downregulation.(A) Western blot and RT-qPCR results of FOXQ1
   levels in FOXQ1-overexpressed cells or its control. ****, P < 0.0001.
   (B) Cell viability analyzed by CCK8 assay in cells of the control,
   miR-320-3p mimics, miR-320-3p mimics + control OE and miR-320-3p
   mimics + FOXQ1 OE groups. OE: overexpression. (C) Apoptotic rate of
   aged mice derived GCs was determined by flow cytometry. *, P < 0.05.
   (D) Cleaved caspase-3, cleaved caspase-9, BAX, and BCL2 protein
   expression were analyzed by Western blot. (E) and summary graph. (F)
   MMP assessed by JC-1 staining. Scale bar, 100 μm. (G) and quantitative
   graph. *, P < 0.05. (H) ATP concentration in aged mice derived GCs. **,
   P < 0.01; ***, P < 0.001. (I) The relative expression of cell mtDNA.
   ****, P < 0.0001; *, P < 0.05.

hnRNPA2B1 controls the loading of miR-320-3p into exosomes

   To investigate the molecular mechanism by which miR-320-3p is loaded
   into exosomes, an RNA pull-down assay was performed, and proteins bound
   with miR-320-3p were subjected to mass spectrometry for identification
   ([166]Table S1). The precipitated proteins included various
   heterogeneous nuclear ribonucleoproteins (hnRNPs), including hnRNPA2B1,
   hnRNPL, hnRNPR, and hnRNPU. We then confirmed the specific binding
   between hnRNPA2B1 and miR-320-3p ([167]Fig. 7A) and examined the level
   of hnRNPA2B1 protein expression in exosomes and in GCs from aged mice
   ([168]Fig. 7B).

Figure 7.

   [169]Figure 7.
   [170]Open in a new tab

   hnRNPA2B1 controls the loading of miR-320-3p into exosomes. (A) RNA
   pull-down western blot analysis showing the binding of hnRNPA2B1 to
   miR-320-3p in Y-Cyt and O-Cyt. Y-Cyt: young mice cytoplasm, O-Cyt: old
   mice cytoplasm. (B) Representative Western blot analysis showing the
   expression of hnRNPA2B1 in O-exos and O-GCs. O-exos: old mice derived
   exosomes, O-GCs: old mice derived GCs. (C) Western blot analysis of
   hnRNPA2B1 in HeLa cells from KO or overexpression hnRNPA2B1. KO:
   knockout, hnRNPA2B1-Flag: overexpression hnRNPA2B1. (D) RT-qPCR
   analysis of miR-320-3p in exosomes from KO or overexpression hnRNPA2B1.
   *, P < 0.05; ***, P < 0.001.

   To reveal the role of hnRNPA2B1 in the process of miR-320-3p loading
   into exosomes, we knocked out and overexpressed hnRNPA2B1 in HeLa cells
   ([171]Fig. 7C) and measured miR-320-3p levels in both exosomes and GCs.
   We found that hnRNPA2B1 depletion led to a significant reduction in
   miR-320-3p in exosomes (P < 0.05, [172]Fig. 7D). In contrast, hnRNPA2B1
   overexpression resulted in an elevation in exosomal miR-320-3p levels
   (P < 0.05, [173]Fig. 7C and [174]7D), suggesting that hnRNPA2B1 is
   indispensable for miR-320-3p loading from GCs to exosomes.

Discussion

   The intercellular communication between GCs and oocytes provides
   nutrients for the development and maturation of oocytes [[175]28], and
   oocyte-secreted factors promote the proliferation and differentiation
   of GCs [[176]29]. GCs dysfunction impairs oocyte developmental capacity
   and quality. In the co-culture experiment of exosomes and the mouse
   oocyte-cumulus complex, the type of cells absorbing exosomes was
   clarified. Only cumulus GCs in the oocyte-cumulus complex absorbed
   exosomes, while oocytes did not. This may be attributed to the zona
   pellucida surrounding the oocytes blocking the transmission of
   exosomes.

   We evaluated how the behaviors of GCs isolated and cultured in vitro
   from aged mice were affected by follicular fluid exosomes from young
   mice. The results showed that follicular fluid exosomes from young mice
   could be absorbed by ovarian GCs of aged mice, significantly enhancing
   the proliferation ability and mitochondrial function of GCs and
   alleviating GCs apoptosis. Next, in vivo experiments revealed that
   young mouse-derived exosomes significantly increased the weight of
   mouse ovaries and the number of follicles at all developmental stages
   in aged mice, attenuated apoptosis in ovarian tissue, enhanced the
   mitochondrial function of GCs, alleviated aging-related demonstrations
   and significantly reduced the number of lipid droplets in ovarian
   follicles of aged mice. The accumulation of intracellular lipids leads
   to high levels of free fatty acids that cause oxidative damage and
   produce highly active oxidative metabolites that cause irreversible
   cell damage [[177]30], and long-term damage and oxidative processes
   impair cellular homeostasis, ultimately leading to apoptosis [[178]31].
   The function of GCs is closely related to oocyte quality, and
   enhancement of GCs function improves the quality of oocytes. This may
   be part of the mechanism by which young mouse-derived exosomes improve
   the ovarian function of aged mice.

   Furthermore, transplantation of follicular fluid exosomes from young
   mice reduced the number of abnormal spindles and increased the
   percentage of blastocysts in aged mice. The results of mating
   experiments showed that young follicular fluid exosomes have an
   increasing litter size trend, validating the conclusion that young
   mouse-derived exosomes may improve oocyte quality and ovarian function
   in aged mice. Consistent with our results, intravenous injection of
   serum exosomes from young mice reversed the increase in the expression
   of aging-related markers in aged male mice [[179]17]. Another study
   found that intraovarian injection of exosomes from human umbilical cord
   mesenchymal stem cells increased the number and quality of oocytes and
   restored diminished fertility in aged female mice [[180]32]. These
   findings confirm that exosomes can modulate ovarian functions after
   being absorbed by the ovaries.

   Current evidence has shown that exosomes contain a large number of
   miRNAs. These miRNAs can be transferred to target cells via exosomes,
   thus changing the gene expression and biological activity of target
   cells [[181]33–35]. In this study, higher levels of miR-320a-3p were
   found in follicular fluid exosomes from young women, indicating that
   loading disorder of miRNAs into exosomes may occur during the ovarian
   aging process. We found that miR-320-3p can be transferred by young
   mouse-derived exosomes to the ovarian GCs of aged mice. miR-320-3p was
   found to be involved in regulating GCs proliferation, apoptosis, and
   mitochondrial functions. Furthermore, miR-320-3p overexpression and
   co-culture with young mouse-derived exosomes exerted similar effects on
   ovarian GCs of aged mice, suggesting that follicular fluid-derived
   exosomes regulate GCs function by delivering miR-320-3p.

   miRNAs are short noncoding RNA molecules that act as gene regulators by
   inhibiting translation or inducing the degradation of target mRNAs via
   binding to the 3’UTR of target mRNAs [[182]36–39]. miR-320a-3p is a
   member of the miR-320a family. In acute pancreatitis, miR-320-3p
   promotes cell proliferation by inhibiting the inflammatory response
   through the PTK2 axis [[183]40]. Furthermore, miR-320-3p facilitates
   myogenic progenitor proliferation and differentiation by regulating
   actin remodeling [[184]41]. In addition, a recent study found that
   miR-320-3p regulates hypoxia-induced proliferation, migration and
   apoptosis of pulmonary artery smooth muscle cells [[185]42]. These
   findings suggest that upregulation of miR-320-3p may promote cell
   proliferation.

   Several target genes of miR-320a-3p have already been identified, among
   which Foxq1, a human-mouse homologous gene, may be responsible for the
   effect of miR-320a-3p in regulating the function of ovarian GCs in aged
   mice. The interaction between miR-320a-3p and Foxq1 was validated by
   dual luciferase experiments, indicating that the positive effect of
   exosomal miR-320-3p may, at least partially, rely on suppressing FOXQ1
   expression. These results link follicular fluid exosomal miR-320-3p to
   Foxq1, providing new clues for unveiling the impact of young follicular
   fluid exosomes on aged ovarian GCs.

   Selective loading of miRNAs into exosomes is directly related to
   pathological processes in several diseases, including cancer,
   neuroinflammation, diabetes, and heart disease [[186]43]. Various
   mechanisms have been indicated in the loading of miRNAs into
   extracellular vesicles, including cytosol-mediated miRNA delivery,
   specific modifications and the interaction between miRNAs and
   RNA-binding proteins [[187]44]. Previous studies have confirmed that
   RNA-binding proteins play an important role in the loading of miRNAs
   into exosomes. In hepatocytes, the RNA-binding protein SYNCRIP controls
   the sorting and loading of miRNAs into hepatocyte exosomes [[188]45].
   In lung cancer cells, miR-122-5p is selectively secreted into lung
   cancer exosomes by binding to the RNA-binding protein hnRNPA2B1
   [[189]46]. In the present study, hnRNPA2B1 was identified as a
   miR-320-3p binding protein. hnRNPA2B1 is responsible for controlling
   the sorting and packaging of miRNAs into exosomes by binding to
   specific motifs [[190]47]. Here, we revealed the interaction between
   miR-320a-3p and the RNA-binding protein hnRNPA2B1 and elucidated that
   hnRNPA2B1 depletion can lead to a disorder in miR-320-3p loading into
   exosomes, partially revealing the molecular mechanisms underlying the
   process of miRNA loading into follicular fluid-derived exosomes.
   However, whether and how hnRNPA2B1 is involved in regulating the
   biological functions of follicular fluid-derived exosomes needs further
   exploration.

   Compared with other RNA therapies, exosomes have been considered an
   ideal vehicle for drug delivery. As a natural product of the human
   body, follicular fluid exosomes have the advantages of low
   immunogenicity, low ethical risk, nontumorigenicity and high safety.
   Moreover, the double-membrane structure of exosomes prevents their
   contents from being degraded. The biggest challenge in RNA therapy is
   its cytotoxicity. In this study, co-culture with follicular fluid
   exosomes and miR-320-3p overexpression in GCs exerted similar effects
   on ovarian function in aged mice, suggesting that exosomes are a
   promising strategy for RNA therapy.

   In conclusion, these results confirmed that follicular fluid-derived
   exosomes of young mice can promote ovarian functions by transferring
   miR-320a-3p to GCs ([191]Fig. 8), indicating that follicular
   fluid-derived exosomes can be considered a potential therapy for
   ovarian aging. Follicular fluid is an unneeded clinical product
   generated during the process of assisted reproductive technology.
   Compared with exosomes derived from other cells, follicular fluid
   exosomes have the advantages of easy accessibility and high yield,
   which gives them promising clinical application prospects. Our research
   results verify the antiaging effects of follicular fluid exosomes from
   young patients and provide a highly valuable treatment strategy to
   counteract ovarian aging.

Figure 8.

   [192]Figure 8.
   [193]Open in a new tab

   Schematic diagram. hnRNPA2B1 mediated miR-320a-3p sorting into young
   exosomes, which inhibited aged granulosa cells apoptosis, promoted
   cells proliferation, and improved mitochondrial function by inhibiting
   FOXQ1 expression of aged mouse in vitro.

Research limitations

   There are still some limitations to this study. First, the present
   study did not further evaluate the exact pathway mechanism by which
   miR-320-3p regulates FOXQ1 expression to promote ovarian GC
   proliferation and enhance mitochondrial function in older aged mice. In
   addition, the contents of follicular fluid exosomes are complex, and it
   cannot be excluded that some other factors also play important
   biological functions, such as lncRNAs, circRNAs, or proteins. That may
   be helpful in the future screening of key regulatory molecules.

Methods

Extraction of follicular fluid-derived exosomes

   Human follicular fluid samples were collected from follicles that
   reached a diameter of 18 mm or more by ultrasound inspection. The
   samples were centrifuged at 300 g for 10 min to remove dead cells, cell
   debris, and large EVs and then at 2000 g for 10 min and 10,000 g for
   30 min. Then, the samples were filtered through a 0.22 μm filter, and
   the supernatants were ultracentrifuged and washed with
   phosphate-buffered saline (PBS) at 100,000 g for 90 min. The exosome
   precipitate was suspended in 100 μL of PBS.

   Mouse follicular fluid-derived exosomes were extracted by mechanical
   method. Specifically, 6- to 8-week-old C57BL/6J female mice were
   treated with 10 IU pregnant mare serum gonadotropin (PMSG) at 46−48 h
   prior to exosome extraction. After execution, mouse ovaries were then
   dissected and placed in Dulbecco's Modified Eagle Medium (DMEM)/F12
   complete medium to remove blood and fat tissue. Ovaries were then
   transferred to a fresh DMEM/F12 complete medium and the ovarian
   follicles were punctured with 1 mL syringes to release the follicular
   fluid. The follicular fluid was then filtered using a 70 μm filter to
   remove oocytes and remaining ovarian tissues. The centrifugation steps
   were performed at 4°C, including 300 g for 10 min, 2000 g for 10 min,
   and 10,000 g for 30 min, and the supernatant was collected after each
   centrifugation step. The filtrate was then passed through a 0.22 μm
   filter and subjected to EV extraction using the ExoQuick-TC™ reagent
   kit according to the manufacturer’s instructions. The EVs were
   resuspended in 100 μL of PBS, frozen at −80°C, and identified using
   electron microscopy, Western blot, and NTA, as described in the
   protocol for human follicular fluid EV identification.

Exosome uptake assay

   Exosomes were stained with the green fluorescent cell membrane labeling
   dye PKH67 (Sigma, USA) according to the manufacturer’s instructions at
   room temperature and then blocked with 1% bovine serum albumin
   (BSA)/PBS solution. An ExoQuick-TC™ kit (SBI, USA) was used to extract
   exosomes and remove the unconjugated dye. Labeled exosomes were
   incubated with GCs for 24 h. After treatment, GCs were collected, and
   nuclei were stained with Hoechst 33342 (Invitrogen, USA). The
   fluorescence signal was observed using a fluorescence microscope.

Intrabursal injection of exosomes

   C57BL/6J mice (40 weeks) were used for modeling, and 1% pentobarbital
   was used for anesthesia. The skin near the ovary side adjacent to the
   spine was shaved, a small incision was made, and the fat pad at the
   ovary was fixed to expose the ovary. Intracapsular ovary injection was
   performed, and the modeling time was 3 weeks. The mice were sorted into
   a PBS injection group and a Y-exos (young C57BL/6J mouse follicular
   fluid exosomes) injection group. The injection volume was 10 µL, and
   the exosome mass was 20 µg. In the ovarian weight experiment,
   unilateral injection of ovaries (injected side and noninjected side)
   was performed, and in the other experiments, bilateral injection of
   ovaries was performed.

Research ethics

   This project was approved by the Ethics Committee of Reproductive
   Medicine Center, Tongji Medical College, Huazhong University of Science
   and Technology ([2020] ethical approval number: 007) and Institutional
   Animal Care and Use Committee, Huazhong University of Science and
   Technology (ethical approval number: S3745). Follicular fluid samples
   were collected with the informed consent of patients.

Statistical analysis

   Data are expressed as the mean ± standard deviation (SD). The results
   between two groups that conformed to a normal distribution were
   compared using unpaired t tests and those that did not conform to a
   normal distribution were compared via Mann‒Whitney tests. Results among
   three or four groups were compared via one-way analysis of variance.
   Statistical analysis was performed using the Statistical Package for
   the Social Sciences, version 12.0 (SPSS Inc., Chicago, IL, USA).
   P < 0.05 indicates that a difference is statistically significant. Each
   experiment was repeated three times.

Supplementary Material

   lnae013_suppl_Supplementary_Figures_S1
   [194]lnae013_suppl_Supplementary_Figures_S1.tif^ (3.2MB, tif)
   lnae013_suppl_Supplementary_Figures_S2
   [195]lnae013_suppl_Supplementary_Figures_S2.tif^ (2.9MB, tif)
   lnae013_suppl_Supplementary_Figures_S3
   [196]lnae013_suppl_Supplementary_Figures_S3.pdf^ (744.6KB, pdf)
   lnae013_suppl_Supplementary_Figures_S4
   [197]lnae013_suppl_Supplementary_Figures_S4.tif^ (2.2MB, tif)
   lnae013_suppl_Supplementary_Figures_S5
   [198]lnae013_suppl_Supplementary_Figures_S5.pdf^ (4.7MB, pdf)
   lnae013_suppl_Supplementary_Data
   [199]lnae013_suppl_Supplementary_Data.docx^ (30.5KB, docx)

Contributor Information

   Yu Liu, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China;
   Department of Gynaecology, Hubei Maternal and Child Health Hospital,
   Wuhan 430070, China.

   Hongbei Mu, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Yu Chen, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Kexin Li, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Qiaojuan Mei, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Lingjuan Wang, Department of Obstetrics and Gynecology, Tongji
   Hospital, Tongji Medical College, Huazhong University of Science and
   Technology, Wuhan 430030, China.

   Tianyu Tang, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Qiuzi Shen, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Huaibiao Li, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Ling Zhang, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

   Jing Li, State Key Laboratory of Reproductive Medicine, Nanjing Medical
   University, Nanjing 211166, China.

   Wenpei Xiang, Institute of Reproductive Health, Tongji Medical College,
   Huazhong University of Science and Technology, Wuhan 430030, China.

Author contributions

   Yu Liu (Conceptualization [Equal], Investigation [Equal], Methodology
   [Equal], Visualization [Equal], Writing – original draft [Equal],
   Writing – review & editing [Equal]), Hongbei Mu (Investigation [Equal],
   Writing – original draft [Equal], Writing – review & editing [Equal]),
   Yu Chen (Investigation [Equal], Methodology [Equal]), Kexin Li
   (Investigation [Equal]), Qiaojuan Mei (Investigation [Equal]), Lingjuan
   Wang (Investigation [Equal]), Tianyu Tang (Investigation [Equal]),
   Qiuzi Shen (Investigation [Equal]), Huaibiao Li (Methodology [Equal]),
   Ling Zhang (Conceptualization [Equal], Methodology [Equal], Project
   administration [Equal], Supervision [Equal]), Jing Li
   (Conceptualization [Equal], Supervision [Equal]), and Wenpei Xiang
   (Conceptualization [Equal], Funding acquisition [Equal], Methodology
   [Equal], Project administration [Equal], Supervision [Equal], Writing –
   review & editing [Equal])

Conflict of interest

   The authors declare no conflict of interests.

Acknowledgements