Ovulation and ovarian wound healing are impaired with advanced reproductive age

Abstract

Aging is associated with reduced tissue remodeling efficiency and increased fibrosis, characterized by excess collagen accumulation and altered matrix degradation. Ovulation, the process by which an egg is released from the ovary, is one of the most dynamic cycles of tissue wounding and repair. Because the ovary is one of the first organs to age, ovulation and ovarian wound healing is impaired with advanced reproductive age. To test this hypothesis, we induced superovulation in reproductively young and old mice and determined the numbers of eggs ovulated and corpora lutea (CLs), the progesterone producing glands formed post-ovulation. Reproductively old mice ovulated fewer eggs and had fewer CLs relative to young controls. Moreover, reproductively old mice exhibited a greater number of oocytes trapped within CLs and expanded cumulus oocyte complexes within unruptured antral follicles, indicative of failed ovulation. In addition, post-ovulatory tissue remodeling was compromised with age as evidenced by reduced CL vasculature, increased collagen, decreased hyaluronan, decreased cell proliferation and apoptosis, impaired wound healing capacity, and aberrant morphology of the ovarian surface epithelium (OSE). These findings demonstrate that ovulatory dysfunction is an additional mechanism underlying the age-related loss of fertility beyond the reduction of egg quantity and quality.

Keywords: fibrosis; ovary; ovulation; reproductive aging; wound healing.

Figure 1

Reproductively old mice weigh more than young mice but respond similarly to hormone stimulation. (A) Graph showing the average number of each follicle class per ovarian section (every fifth section of serially sectioned ovaries was counted; N = 20 ovaries from 5 reproductively young and 5 old mice). The inset shows the average number of antral follicles and other (degenerate) follicles per ovarian section with a different scale. T-tests were performed within each follicle class; asterisks denote P < 0.0001. (B) Weights of reproductively young and old mice (n = 11 young and 11 old mice). A t-test was performed; asterisks denote P < 0.0001. (C) Serum estradiol levels were assessed (reportable range 3-300 pg/mL). A t-test was performed and there was no difference in estradiol levels between age cohorts. (D) Serum progesterone levels were assessed (reportable range 0.15-40 ng/mL). A t-test was performed and there was no difference between age cohorts. Data are represented as mean ± SEM. N = 20 samples for hormone analysis.

Figure 2

Reproductively old mice ovulate fewer eggs than reproductively young mice, but the proportion of normal and degenerate eggs does not differ between age groups. (A) The number of eggs ovulated into the oviducts of each mouse was tracked, and the graph shows the average number of eggs ovulated per mouse (N = 11 mice per age group, 22 oviducts). A t-test was performed; asterisks denote P < 0.0001. Data are represented as mean ± SEM. The number of morphologically normal eggs ovulated and retrieved from each oviduct from (B) reproductively young and (C) old individual mice. Representative brightfield microscopy images of morphologically (D) normal and (E) degenerate eggs. Scale bars are 100 μm. (F) The graph shows the proportion of morphologically normal and degenerate eggs ovulated from both age cohorts.

Figure 3

Ovaries from reproductively old mice contain fewer CLs than young mice. (A) Representative H&E-stained ovarian tissue sections demonstrate the CL tracking process. Numbers in the upper left corner indicate the tissue section number. Black asterisks identify the same CL that appears across tissue sections. Indicated CL was visible beginning in section 75 and ending in section 240. Scale bars are 100 μm. (B) Graph showing the number of CLs per ovary for both age cohorts. A t-test was performed; asterisks denote significance (P = 0.008) (N = 5 mice, 10 ovaries per age group). Data are represented as mean ± SEM. The number of CLs per ovary for all (C) reproductively young and (D) reproductively old mice.

Figure 4

The discrepancy between eggs ovulated and CLs tracked is partially accounted for by regressing CLs from previous estrous cycles. (A) The number of CLs tracked per ovary and the number of eggs ovulated per oviduct for reproductively young mice (N = 5 mice, 10 ovaries). (B) The number of CLs tracked per ovary and the number of eggs ovulated per oviduct for reproductively old mice (N = 5 mice, 10 ovaries). (C) Representative image of Galectin-3 IHC in an ovary from a reproductively young mouse. All CLs are outlined in blue. Galectin-3 positive CLs are indicated with black asterisks. The inset shows the IgG control. The scale bar is 100 μm. (D) Graph showing the percent of Galectin-3 positive CLs (~24.3%) and Galectin-3 negative CLs (~75.7%) (N = 20 ovaries, 3 slides per ovary). (E) The graph shows the average discrepancy between CLs and eggs for both age cohorts. A t-test was performed and there was no difference in the discrepancy between the age cohorts (P > 0.05).

Figure 5

Reproductively old mice exhibit more failed ovulation events. (A) Representative H&E-stained ovarian tissue sections depicting two phenotypes indicative of failed ovulation events including (I) trapped oocytes within CLs and (II) expanded unruptured antral follicles. Scale bars are 100 μm. (B) Graph showing the percent of failed ovulations per CL number for reproductively young and old mice (N = 10 ovaries per age group). A t-test was performed and P = 0.025. Data are represented as mean ± SEM. The number of failed ovulations per ovary for all (C) reproductively young and (D) reproductively old mice. (E) Representative H&E-stained ovarian tissue section depicting a hemorrhagic CL. The scale bar is 50 μm. (F) Graph showing the percent of hemorrhagic CLs for both age cohorts (N = 10 ovaries per age group). A t-test was performed and there was no significant difference (P > 0.05). Data are represented as mean ± SEM. The percent of hemorrhagic CLs for all (G) reproductively young and (H) reproductively old mice.

Figure 6

CLs exhibit age-associated structural differences. (A) Representative H&E-stained CL. The perimeter of the CL is outlined in blue. The scale bar is 50 μm. (B) Graph showing the area of each CL. Each data point corresponds to one CL. A t-test was performed; asterisk indicates significance (P = 0.022). (C) Images showing how luteinized area was determined at 40X magnification and the same images in ImageJ software. (D) Graph showing the average number of nuclei per defined CL area. Each data point corresponds to one CL. A t-test was performed; asterisks denote P < 0.0001. (E) Representative image of CD31 staining in reproductively young ovarian tissue with the boxed region in (F) and (G) representative image of CD31 staining in reproductively old ovarian tissue with the boxed region in (H). (I) Graph showing the percent of CL area positive for vasculature. A t-test was performed (P = 0.01). Scale bars are (E, G) 200 μm and (F, H) 50 μm. Data are represented as mean ± SEM.

Figure 7

Collagen significantly increases in the whole ovary and around CLs in reproductively old mice. (A) Representative images depicting (I) PSR staining in a reproductively young ovary and (II) the same image in ImageJ software. (III) Representative image of an H&E-stained CL, (IV) the same CL with PSR staining, and (V) the CL in ImageJ. Scale bars are (I-II) 100 μm and (III-V) 25 μm. (B) Representative images of a reproductively old ovary and CL with the same histological techniques in (A). Scale bars are (I-II) 100 μm and (III-V) 25 μm. (C) Graph showing the percent of reproductively young and old ovaries positive for PSR (N = 10 ovaries per age group). A t-test was performed; asterisk denotes P = 0.019. (D) Graph showing analysis of the percent of PSR positive area in and surrounding CLs for both age cohorts. Each data point corresponds to one CL. A t-test was performed; asterisks denote P < 0.0001. Data are represented as mean ± SEM (C and D).

Figure 8

Hyaluronan significantly declines in the ovarian stroma of reproductively old mice. (A) Representative image of the hyaluronan binding protein (HABP) assay performed with reproductively young ovaries. The whole ovarian area is outlined in red. Follicles are outlined in yellow. CLs are outlined in blue. (B) Representative image of the HABP assay performed on reproductively old ovaries with the same ovarian sub-compartments outlined. Scale bars are (A, B) 200 μm. Graphs showing hyaluronan as intensity per area within (C) whole ovaries, (D) follicles, (E) CLs, and (F) the ovarian stroma between age cohorts. T-tests were performed; asterisks indicate significant differences (C: P = 0.043; F: P = 0.012). Data are represented as mean ± SEM (C–F). N = 10 ovaries per age group.

Figure 9

Cell proliferation is reduced post-ovulation in reproductively old mice. (A) Representative image of IHC labeling Ki67 in reproductively young ovaries. The whole ovarian area is outlined in red. Follicles are outlined in yellow. CLs are outlined in blue. (B) Representative image of IHC labeling Ki67 in reproductively old ovaries with the same ovarian sub-compartments outlined. Insets depict the boxed region of the OSE in each respective image. Scale bars are (A, B) 500 μm and (insets) 25 μm. Graphs showing the number of Ki67 positive cells per sub-compartment are within (C) whole ovaries, (D) follicles, (E) CLs, (F) the ovarian stroma, and (G) the OSE. T-tests were performed; asterisks indicate significant differences (C: P = 0.0294; D: P = 0.0091; G: P = 0.0023). Data are represented as mean ± SEM (C–G). N = 5 ovaries per age group.

Figure 10

Cell death is reduced post-ovulation in reproductively old mice. (A) Representative image of IHC labeling CC3 in reproductively young ovaries. The whole ovarian area is outlined in red. Follicles are outlined in yellow. CLs are outlined in blue. (B) Representative image of the IHC labeling CC3 in reproductively old ovaries with the same ovarian sub-compartments outlined. Insets depict the boxed region of the OSE in each respective image. Scale bars are (A, B) 500 μm and (insets) 25 μm. Graphs showing the number of CC3 positive cells per sub-compartment are within (C) whole ovaries, (D) follicles, (E) CLs, (F) the ovarian stroma, and (G) the OSE. T-tests were performed; asterisks indicate significant differences (E: P = 0.022). Data are represented as mean ± SEM (C–G). N = 5 ovaries per age group.

Figure 11

OSE reformation post-ovulation in vivo is reduced with age. Representative image of Troma-1 IHC in (A) reproductively young ovaries and (B) reproductively old ovaries. Scale bars are 200 μm (A, B). Graphs showing the proportion of the ovarian surface that was (C) Troma-1 positive and (D) E-Cadherin positive. T-tests were performed for both parameters and asterisks denote significance (C: P = 0.01; D: P = 0.04). For Troma-1, N = 10 ovaries per age group, each ovary was analyzed in triplicate or quadruplicate (N = 34 measurements for reproductively young ovaries, N = 35 measurements for reproductively old ovaries). For E-Cadherin, N = 10 ovaries per age group (N = 20 measurements for reproductively young ovaries, N = 20 measurements for reproductively old ovaries).

Figure 12

The in vitro wound healing ability of the OSE is compromised with age. (A) Ovary pieces encapsulated in alginate hydrogel beads. (B) Representative images of Troma-1 IHC with reproductively young (n = 7) and old ovary pieces (n = 7) at Day 0 and Day 8 of culture. Ovaries were fixed at each timepoint, therefore, the ovaries are shown at D0 and D8 are not the same. Scale bars are 100 μm. (C) Graph showing the average percent area of ovarian pieces encapsulated by Troma-1 positive cells. T-tests were performed; asterisk denotes significance (P = 0.029). Data are represented as mean ± SEM.

Figure 13

The OSE exhibits aberrant morphologies in reproductively old ovaries. (A) Representative 40X image of the OSE in an ovary from a reproductively young mouse. (B) Representative 40X image of the OSE in an ovary from a reproductively old mouse. Arrowheads indicate invaginations of the OSE into the ovarian stroma. (C) Graph showing the average thickness of the OSE for all reproductively young and old mice. N = 10 ovaries per age group, each ovary was analyzed in triplicate or quadruplicate (N = 34 measurements for reproductively young ovaries, N = 35 measurements for reproductively old ovaries). (D) The graph shows the number of invaginations per ovary section for all mice. N = 10 ovaries per age group. (E) Representative image of an aged ovary. Arrowheads indicate epithelial growths. (F) High magnification region of the boxed region in (E) depicting an epithelial outgrowth. (G) The graph shows the mean number of epithelial outgrowths per ovary section for all mice. N = 10 ovaries per age group. T-tests were performed on all measured parameters with asterisks indicating significant differences (C: P = 0.007; D: P < 0.0001; E: P = 0.0007). Scale bars are (A, B, F) 50 μm and (E) 200 μm. Data are represented as mean ± SEM (C, D, G). N = 10 ovaries per age group.

Figure 14

COX2 expression is decreased in reproductively old ovaries. (A) Representative image of a reproductively young ovary stained for COX2 expression. (B) Magnified image of boxed region in (A), a site of recent follicle rupture. (C) Representative image of a reproductively old ovary stained for COX2 expression. (D) Magnified image of boxed region in (C). The insets are IgG controls of the boxed regions. Scale bars are (A, C) 100 μm and (B, D) 25 μm. (E) Graph showing the reciprocal intensity values of recently ruptured follicles and formed CLs in tissue from reproductively young and old mice. A.u. stands for arbitrary units. A t-test was performed; asterisks indicate P < 0.0001. Data are represented as mean ± SEM.