Iron accumulation in ovarian microenvironment damages the local redox balance and oocyte quality in aging mice

Abstract

Accumulating oxidative damage is a primary driver of ovarian reserve decline along with aging. However, the mechanism behind the imbalance in reactive oxygen species (ROS) is not yet fully understood. Here we investigated changes in iron metabolism and its relationship with ROS disorder in aging ovaries of mice. We found increased iron content in aging ovaries and oocytes, along with abnormal expression of iron metabolic proteins, including heme oxygenase 1 (HO-1), ferritin heavy chain (FTH), ferritin light chain (FTL), mitochondrial ferritin (FTMT), divalent metal transporter 1 (DMT1), ferroportin1(FPN1), iron regulatory proteins (IRP1 and IRP2) and transferrin receptor 1 (TFR1). Notably, aging oocytes exhibited enhanced ferritinophagy and mitophagy, and consistently, there was an increase in cytosolic Fe2+, elevated lipid peroxidation, mitochondrial dysfunction, and augmented lysosome activity. Additionally, the ovarian expression of p53, p21, p16 and microtubule-associated protein tau (Tau) were also found to be upregulated. These alterations could be phenocopied with in vitro Fe2+ administration in oocytes from 2-month-old mice but were alleviated by deferoxamine (DFO). In vivo application of DFO improved ovarian iron metabolism and redox status in 12-month-old mice, and corrected the alterations in cytosolic Fe²⁺, ferritinophagy and mitophagy, as well as related degenerative changes in oocytes. Thereby in the whole, DFO delayed the decline in ovarian reserve and significantly increased the number of superovulated oocytes with reduced fragmentation and aneuploidy. Together, our findings suggest that aging-related disturbance in ovarian iron homeostasis contributes to excessive ROS production and that iron chelation may improve ovarian redox status, and efficiently delay the decline in ovarian reserve and oocyte quality in aging mice. These data propose a novel intervention strategy for preserving the ovarian reserve function in elderly women.

Graphical abstract

Fig. 1

The decrease in ovarian follicle number and increase in ovarian iron content during mice natural aging. The iron content in mouse ovary, hippocampus, liver and oocytes was detected by inductively coupled plasma mass spectrometry (ICP-MS). A. Representative images of ovarian HE staining section of 2, 6, 12 months group mice. The arrows point to: (a). Antral follicle; (b). Secondary follicle; (c). Primary follicles; (d). Primordial follicle; (e). Atretic follicle. (n = 3). Scale bar = 20 μm. B. Statistical analysis of the number of follicles at each level in ovarian HE staining section of 2, 6, 12 months group mice. C. Iron content per gram in ovary, hippocampus and liver in mice at age of 2, 6, 12 and 18 months. (n = 8/group). D. The average iron content in oocyte of mice at age of 2, 6, 12 months. (2 m: 0.8663 ± 0.07412, 6 m: 0.8736 ± 0.03956, 12 m: 1.329 ± 0.2243; n = 280/group). Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001.

Fig. 2

Dynamic changes of iron metabolic proteins in natural aging ovaries. A. Western blot analysis of FTH, FTL, FTMT, DMT1, FPN1 and TFR1 in the ovaries from mice at age of 2, 6, 12 months. B. Western blot analysis of HO-1, IRP1, IRP2 and NCOA4 in the ovaries from mice at age of 2, 6, 12 months. C. Western blot analysis of p53, p21, p16, Tau and GPX4 in the ovaries from mice at age of 2, 6, 12 months. D-R. Statistical analysis of difference in protein levels of FTH, FTL, FTMT, DMT1, FPN1, TFR1, HO-1, IRP1, IRP2, NCOA4, p53, p21, p16, Tau and GPX4 in the ovaries from mice at age of 2, 6 and 12 months. (n = 8). Total proteins values were compared to the 2 months. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001.

Fig. 3

Changes in protein expression relevant to iron metabolism, autophagy and oxidative stress in natural aging oocytes. A. Western blot analysis of FTH, FTL and FTMT in the oocytes from mice at age of 2 and 12 months. B. Western blot analysis of DMT1, FPN1, TFR1, IRP1 and IRP2 in the oocytes from mice at age of 2 and 12 months. C. Western blot analysis of NCOA4, LC3, SQSTM1 and Beclin1 in the oocytes from mice at age of 2 and 12 months. D. Western blot analysis of MDA, SOD2, GPX4, PRX and MFN1 in the oocytes from mice at age of 2 and 12 months. E-U. Statistical analysis of difference in protein levels of FTH, FTL, FTMT, DMT1, FPN1, TFR1, IRP1, IRP2, NCOA4, LC3 II, SQSTM1, Beclin1, MDA, SOD2, GPX4, PRX and MFN1 in the oocytes from mice at age of 2 and 12 months. (n = 3–5). Total proteins values were compared to the 2 months, which was normalized to 1.0. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01, ****P < 0.001.

Fig. 4

Fe²⁺ accumulation in mitochondria and lysosome in aging oocytes. A. Representative images of LysoTracker, MitoTracker and FeRhonox-1 signal in the oocytes at age of 2 and 12 months. Hochest was showed in blue, Lysotracker was showed in green, Mitotracker was showed in red, FeRhonox-1 was showed in purple. (a) and (b) were enlarged diagram of marked square in representative images. (c) was enlarged diagram of arrow pointed coincident point in representative images. Scale bar = 10 μm. B-D. Relative fluorescence intensity of LysoTracker, MitoTracker and FeRhonox-1 signal in the oocytes of 2 and 12 months. (n = 35–46). E. Statistical analysis of mitochondrial clustering index in the oocytes of 2 and 12 months. (n = 28–29). F. Statistical analysis of spearman's rank correlation value (ρ) of co-location between FeRhonox-1 with LysoTracker or Mitotracker in the oocytes of 2 and 12 months. (n = 38–48). G. Statistical analysis of co-localization migration rate of Fe²⁺ from mitochondria to lysosomes. The formula for calculation is as follows: ρ (co-location between FeRhonox-1 and LysoTracker)/ρ (co-location between feRhonox-1 and MitoTracker). (n = 37–47). Three 2-month-old mice and six 12-month-old mice were used in this experiment. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001.

Fig. 5

The aging oocytes exhibit high ROS, early apoptosis and DNA damage. A. Representative images of DCFH-DA and MitoTracker signal in the oocytes at age of 2 and 12 months. DCFH-DA was showed in green, MitoTracker was showed in red. Scale bar = 10 μm. B. Relative fluorescence intensity of DCFH-DA signal in the oocytes at age of 2 and 12 months. (n = 30–36). C. Representative images of JC-1 signal in the oocytes at age of 2 and 12 months. JC-1/monomers were showed in green; JC-1/aggregates were showed in red. Scale bar = 10 μm. D. Relative fluorescence intensity of JC-1 red/green in the oocytes at age of 2 and 12 months. (n = 29–49). E. Representative images of Annexin V signal in the oocytes at age of 2 and 12 months. Scale bar = 10 μm. F. Relative fluorescence intensity of Annexin V signal in the oocytes at age of 2 and 12 months. Annexin V was showed in green. (n = 31–33). G. Fluorescence intensity profiling of phalloidin in representative images. Lines were drawn through the oocytes, and pixel intensities were quantified along the lines. H. Western blot analysis of γH2AX in the oocytes at age of 2 and 12 months. I. Statistical analysis of difference in protein levels of γH2AX in the oocytes at age of 2 and 12 months. Total fluorescence intensity or proteins values were compared to the 2 months. Nine 2-month-old mice and eighteen 12-month-old mice were used in this confocal experiment. Data were presented as mean ± SEM of at least three times independent experiments. **P < 0.01, ****P < 0.001.

Fig. 6

Ammonium iron (II) sulfate hexahydrate (FAS) leads to an increase of Fe²⁺ and ROS in oocytes, altering the state of lysosomes and mitochondria, which could be reversed by Deferoxamine (DFO). Mice oocytes at age of 2 months were processed with the LysoTracker, MitoTracker, FeRhonox-1 and DCFH-DA staining after 6 h incubation in DMSO, 3000 μm FAS, 3000 μm FAS +3000 μm DFO. A. Representative images of Lysotracker, Mitotracker, FeRhonox-1 and DCFH-DA signal in the oocytes of DMSO, FAS and FAS + DFO groups. Scale bar = 10 μm. B-D. Relative fluorescence intensity of LysoTracker, MitoTracker and FeRhonox-1 signal in the oocytes of DMSO, FAS and FAS + DFO groups. (n = 26–34). E. Statistical analysis of mitochondrial clustering index in the oocytes of DMSO, FAS and FAS + DFO groups. (n = 23–32). F. Relative fluorescence intensity of DCFH-DA signal in the oocytes of DMSO, FAS and FAS + DFO groups. (n = 38–44). Relative fluorescence intensity was compared to the DMSO group. Twelve 2-month-old mice were used in this experiment. Data were presented as mean ± SEM of at least three times independent experiments. *^/#P < 0.05, **^/##P < 0.01, ***P < 0.005.

Fig. 7

Amelioration on free Fe²⁺, ROS level and degenerative changes in aging oocytes incubated with DFO. GV oocytes were collected from mice at age of 12 months, and cultured in DMSO or 3000 μm DFO for 6 h, then processed with live imaging with FeRhonox-1, DCFH-DA and Annexin V or fixation for immunofluorescence staining. Some oocytes were cultured for 16 h in DMSO or 3000 μm DFO and collected for meiotic maturation analysis. A. Representative images of FeRhonox-1 and DCFH-DA signal in the oocytes of DMSO and DFO groups. Scale bar = 10 μm. B–C. Relative fluorescence intensity of FeRhonox-1 and DCFH-DA signal in the oocytes of DMSO and DFO groups. (n = 45–60). D. Representative images of Annexin V signal in the oocytes of DMSO and DFO groups. Scale bar = 10 μm. E. Relative fluorescence intensity of Annexin V signal in the oocytes of DMSO and DFO groups. (n = 42–44). F. Statistics of the rate of oocytes matured to MII phase after 16 h incubation in DMSO and DFO. G. Representative images of spindle in the oocytes of 2 m, 12 m- DMSO and 12 m- DFO group. The calculation model of the relative width of chromosome plate region. The radius of the oocyte is represented as R; the width of the chromosome plate is represented as d. The relative chromosomes region width = d/R. Scale bar = 10 μm. H. Statistical analysis of the abnormal rate of spindle in the oocytes from different groups. (n = 55–58). I. Statistical analysis of the d/R value of spindle in the oocytes from different groups. (n = 55–58). Relative fluorescence intensity was compared to the 12 m- DMSO. Fifteen 12-month-old mice were used in this confocal experiment. Data were presented as mean ± SEM of at least three times independent experiments. *^/#P < 0.05, **^/##P < 0.01, ****P < 0.001.

Fig. 8

DFO can rectify iron content and the expression of iron regulatory and autophagy related proteins in the ovaries of aging mice. The 12 months mice were injected intraperitoneally with DFO for 14 days, while the control group was injected with PBS. A. Iron content per gram in the ovaries of 2 m, 12m- PBS and 12m- DFO group. (2 m: 170 ± 15.33, 12 m-PBS: 412.4 ± 18.48, 12 m-DFO: 347.7 ± 9.666; n = 7/group). B. Western blot analysis of FTH, FTL and FTMT in ovaries of PBS and DFO group. C. Western blot analysis of IRP1, IRP2, DMT1, TFR1 and FPN1 in the ovaries of PBS and DFO group. D. Western blot analysis of NCOA4, Beclin1, LC3, and SQSTM1 in ovaries of PBS and DFO group. E. Western blot analysis of p53 in the ovaries of PBS and DFO group. F–R. Statistical analysis of difference in protein levels of FTH, FTL, FTMT, IRP1, IRP2, DMT1, FPN1, TFR1, NCOA4, Beclin1, LC3 II, SQSTM1 and p53 in the ovaries of PBS and DFO group. (n = 5–8). Total proteins values were compared to the PBS group, which was normalized to 1.0. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01.

Fig. 9

DFO can correct ferritinophagy, mitophagy and oxidative stress in aging oocytes. The 12 months mice were injected intraperitoneally with DFO for 14 days, while the control group was injected with PBS. A. Western blot analysis of NCOA4, LC3, SQSTM1, FTMT and FTH in the oocytes of PBS and DFO group. B. Western blot analysis of RAB7 and Parkin in the oocytes of PBS and DFO group. C. Western blot analysis of MDA and γH2AX in the oocytes of PBS and DFO group. D-L. Statistical analysis of difference in protein levels of NCOA4, LC3, SQSTM1, Parkin, RAB7, FTMT, FTH, MDA and γH2AX in the oocytes of PBS and DFO group. (n = 3–4). Total proteins values were compared to the PBS group, which was normalized to 1.0. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005.

Fig. 10

DFO can delay the decline of ovarian reserve in aging mice. The 12 months mice were injected intraperitoneally with DFO for 14 days, while the control group was injected with PBS. A. Representative images of ovarian HE staining section of PBS and DFO group mice. The arrows point to: (a). Antral follicle; (b). Secondary follicle; (c). Primary follicles; (d). Primordial follicle; (e). Atretic follicle. (n = 3–4). Scale bar = 50 μm. B. Statistical analysis of the number of follicles at each level in ovarian HE staining section of PBS and DFO group mice. C. Statistical analysis of AMH content in serum of PBS and DFO group mice. (n = 6/group). Data were presented as mean ± SEM of at least three times independent experiments. *^/#P < 0.05, ****P < 0.001.

Fig. 11

DFO can reduce the levels of iron accumulation, fibrotic changes, and 8-OHdG and 4-HNE in aging ovaries. The 12 months mice were injected intraperitoneally with DFO for 14 days, while the control group was injected with PBS. A. Representative images of Prussian blue staining in ovary sections of 2 m, 6 m, 12 m- PBS and 12 m- DFO group mice. B. Statistical analysis of the intensity of Prussian blue signal in ovary sections of 2 m, 6 m, 12 m- PBS and 12 m- DFO group mice. C. Representative images of Masson staining in ovary sections of 2 m, 6 m, 12 m- PBS and 12 m- DFO group mice. D. Statistical analysis of the intensity of Masson signal in ovary sections of 2 m, 6 m, 12 m- PBS and 12 m- DFO group mice. E. Representative images of 8-OHdG immunohistochemical staining in ovary sections of 2 m, 6 m, 12 m- PBS and 12 m- DFO group mice. F. Statistical analysis of 8-OHdG level in ovary sections from different groups. G. Representative images of 4-HNE immunohistochemical staining in ovary sections of 2 m, 6 m, 12 m- PBS and 12 m- DFO group mice. H. Statistical analysis of 4-HNE level in ovary sections from different groups. This portion of the experiment involved sectioning and staining wax blocks prepared from one side of the ovaries of three different mice. Specifically, the statistical approach involved serially sectioning approximately 45 slices near the largest surface of the mouse ovary. Every 15th slice was chosen for staining and statistical analysis. The average of the statistical results from three selected slices represented the proportion of positive staining or antibody expression area in the ovary. Finally, statistical analysis was performed on the results from the three mice. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001.

Fig. 12

DFO can alleviate the levels of free Fe²⁺, ROS, apoptosis, and also changes in lysosome and mitochondria in aging oocytes. The 12 months mice were injected intraperitoneally with DFO for 14 days, while the control group was injected with PBS. A. Representative images of LysoTracker, MitoTracker and FeRhonox-1 signal in the oocytes of PBS and DFO group. Scale bar = 10 μm. B. Representative images of DCFH-DA, JC-1 and Annexin V signal in the oocytes of PBS and DFO group. Scale bar = 10 μm. C-E. Relative fluorescence intensity of LysoTracker, MitoTracker and FeRhonox-1 signal in the oocytes of PBS and DFO group. (n = 44–62). F. Statistical analysis of mitochondrial clustering index in the oocytes of PBS and DFO group. (n = 35–39). G. Statistical analysis of spearman's rank correlation value (ρ) of co-location between FeRhonox-1 with LysoTracker or MitoTracker in the oocytes of PBS and DFO group. (n = 42–49). H. Statistical analysis of co-localization migration rate of Fe²⁺ from mitochondria to lysosome. The formula for calculation is as follows: ρ (co-location between FeRhonox-1 and Lysotracker)/ρ (co-location between FeRhonox-1 and Mitotracker). (n = 44–49). I. Relative fluorescence intensity of DCFH-DA signal in the oocytes of PBS and DFO group. (n = 42–48). J. Relative fluorescence intensity of JC-1 red/green in the oocytes of PBS and DFO group. (n = 37–42). K. Relative fluorescence intensity of Annexin V signal in the oocytes of PBS and DFO group. (n = 38–40). Relative fluorescence intensity was compared to the PBS group. Twenty-four 12-month-old mice treated with PBS and twenty-four 12-month-old mice treated with DFO were used in this experiment. Data were presented as mean ± SEM of at least three times independent experiments. *P < 0.05, **P < 0.01.

Fig. 13

DFO can improve the reproductive potential of aging mice. The 12 months mice were injected intraperitoneally with DFO for 14 days, while the control group was injected with PBS. A. Representative images of super-ovulated oocytes from mice in injected with PBS or DFO. The arrows point to fragmented oocytes. B. Statistical analysis of the number of super-ovulated oocytes in PBS and DFO group mice. (n = 6/group). C. Statistical analysis of fragmentation rate of oocytes in PBS and DFO group. (n = 6/group). D. Representative images of spindle in the oocytes of 2 m, 12 m - PBS and 12 m - DFO group. Scale bar = 10 μm. E. Statistical analysis of the rate of abnormal spindle in the oocytes from different groups. (n = 82–96). F. Statistical analysis of the d/R value of spindle in the oocytes from different groups. (n = 82–96). G. Representative images of chromosome spreading of MII oocytes in PBS and DFO group. Scale bar = 5 μm. H. Statistical analysis of aneuploid rate of MII oocytes in PBS and DFO group. (n = 8–23). I. Statistical analysis of the proportion of oocytes with compacted chromosomes in 12 m - PBS and 12 m - DFO groups. (n = 23–30). Data were presented as mean ± SEM of at least three times independent experiments. *^/#P < 0.05, **P < 0.01, ^####P < 0.001.