No evidence for neo-oogenesis may link to ovarian senescence in adult monkey

Authors

  • Jihong Yuan,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    2. Key Laboratory of Ministry of Health on Hormones and Development, Metabolic Diseases Hospital, Tianjin Medical University, Tianjin, China
    Search for more papers by this author
  • Dongdong Zhang,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Lei Wang,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Mengyuan Liu,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Jian Mao,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Yu Yin,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Xiaoying Ye,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Na Liu,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Jihong Han,

    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    Search for more papers by this author
  • Yingdai Gao,

    1. The State Key Laboratory of Experimental Hematology, Institute of Hematology, Center for Stem Cell Medicine, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China
    Search for more papers by this author
  • Tao Cheng,

    1. The State Key Laboratory of Experimental Hematology, Institute of Hematology, Center for Stem Cell Medicine, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China
    Search for more papers by this author
  • David L. Keefe,

    1. Department of Obstetrics and Gynecology, New York University Langone Medical Center, New York, USA
    Search for more papers by this author
  • Lin Liu

    Corresponding author
    1. State Key Laboratory of Medicinal Chemical Biology, The 2011 Collaborative Innovation Center of Tianjin for Medical Epigenetics and College of Life Sciences, Nankai University, Tianjin, China
    • Correspondence: Lin Liu, Ph.D., College of Life Sciences, Nankai University, Tianjin 300071, China. Telephone: 86-22-23500752; Fax: 86-22-23500752; e-mail: liulin@nankai.edu.cn

    Search for more papers by this author

  • Author contributions: J.Y., D.Z., L.W., M.L., Y.Y., J.M., X.Y., and N.L.: collection/assembly of data and data analysis/interpretation; J.H., Y.G., and T.C.: provision of study materials and data analysis/interpretation; D.L.K. and L.L.: conception and design, manuscript writing, and revision. J.Y., D.Z., and L.W. contributed equally to this article.

Abstract

Female germline or oogonial stem cells transiently residing in fetal ovaries are analogous to the spermatogonial stem cells or germline stem cells (GSCs) in adult testes where GSCs and meiosis continuously renew. Oocytes can be generated in vitro from embryonic stem cells and induced pluripotent stem cells, but the existence of GSCs and neo-oogenesis in adult mammalian ovaries is less clear. Preliminary findings of GSCs and neo-oogenesis in mice and humans have not been consistently reproducible. Monkeys provide the most relevant model of human ovarian biology. We searched for GSCs and neo-meiosis in ovaries of adult monkeys at various ages, and compared them with GSCs from adult monkey testis, which are characterized by cytoplasmic staining for the germ cell marker DAZL and nuclear expression of the proliferative markers PCNA and KI67, and pluripotency-associated genes LIN28 and SOX2, and lack of nuclear LAMIN A, a marker for cell differentiation. Early meiocytes undergo homologous pairing at prophase I distinguished by synaptonemal complex lateral filaments with telomere perinuclear distribution. By exhaustive searching using comprehensive experimental approaches, we show that proliferative GSCs and neo-meiocytes by these specific criteria were undetectable in adult mouse and monkey ovaries. However, we found proliferative nongermline somatic stem cells that do not express LAMIN A and germ cell markers in the adult ovaries, notably in the cortex and granulosa cells of growing follicles. These data support the paradigm that adult ovaries do not undergo germ cell renewal, which may contribute significantly to ovarian senescence that occurs with age. Stem Cells 2013;31:2538–2550

Introduction

A remarkable decline in the ovarian follicle pool accompanies ovarian aging in women, leading to reproductive aging and menopause [1-3]. Ovarian aging reduces fertility in women who delay attempts at childbearing, and contributes to menopause-related health problems, psychological stress, and increased risk of genetic diseases in offspring. Oocyte renewal or germ cell regeneration, if it did take place in adult human ovaries, would have important implications for the millions of women worldwide who attempt childbearing at midlife, and others who have lost ovarian function following chemo or radiation therapy [4, 5].

Extensive evidence accumulated since the 1950s supports the paradigm that in females of most mammalian species, oogonia, the germline stem cells (GSCs) that produce oocytes, exist only during a brief window of oogenesis in the fetal period of development [6] and that females are born with a nonrenewable pool of follicles that declines progressively with age [4, 5, 7-11]. In mice, oogonia and oocytes in early stages of meiosis, with their chromosomes undergoing recombination, are abundant during fetal development, but disappear shortly after birth [12, 13]. This well-substantiated paradigm was challenged by observations suggesting that ovarian surface epithelium (OSE) of adult mice may contain cells that can sustain postnatal follicular renewal [14]. Recently, putative oogonial stem cells with proliferative capacity were identified in adult mouse and human ovaries [15], but these findings and the functional tests were not confirmed by other laboratories [16-19]. Oocytes can be generated in vitro from embryonic stem cells, induced pluripotent stem cells, and stem cells from skin [20-23]. Yet in vitro conversion of the OSE cells [10, 24, 25] and reprogramming cells to germ-like cells does not prove the existence of GSCs and neo-oogenesis in vivo [13, 18].

GSCs are present as spermatogonial stem cells (SSCs) and maintain their self-renewal and germ cell renewal continuously in the adult testis [26-28]. A similarity in self-renewal and survival mechanisms between human and mouse SSCs exists [29], and mouse and human SSCs share conserved gene expression of GSC regulatory molecules [30]. These data potentially allow for the extrapolation of the knowledge about mouse SSCs to the human germline, which is difficult to study [30]. DAZL is remarkably highly expressed in enriched SSCs of mice and humans, in contrast to testis somatic cells [30]. Mice and rhesus spermatogonia also express similar markers of germ cells (VASA, DAZL) and stem/progenitor spermatogonia (PLZF and GFRα1) [31]. Although spermatogonia express their own specific genes, germline markers for SSCs were validated and are conserved from mouse systems to monkey and human testis, and might be used also for putative female GSCs in monkey and human ovaries. In addition, absent or minimal expression of Lamin A/C also can be used to identify somatic stem cells or progenitor stem cells in adult mammalian tissues [32, 33]. Lamin A/C expression marks both mouse and human embryonic differentiation [34, 35].

Meiotic prophase I encompasses many unique features, including homologous chromosomal searching and pairing, and formation of synaptonemal complexes (SCs) [36, 37]. SCs composed of three main SC proteins (Scp 1–3) form a tripartite structure in early meiotic germ cells, begin to assemble at leptotene, and complete SC formation at pachytene [38]. Both the expression and distribution of SCP3 are frequently used as meiotic markers [39]. Also, early meiocytes form telomere bouquets, and their perinuclear distribution at the termini of SC lateral or axial filaments, indicative of homologous pairing at leptotene to pachytene of prophase I, represents an essential step for subsequent meiotic recombination [40]. Homologous pairing and perinuclear distribution of telomeres can serve as decisive indicators of neo-oogenesis in postnatal mammalian ovaries.

The rhesus monkey provides a suitable model for the ovarian aging that is ubiquitous in women. Macaca mulatta (rhesus monkey) females reach sexual maturity at 3–5 years of age, remain reproductively active over 20 years, and undergo pathological and hormonal changes characteristic of the human female climacteric [41, 42]. As in women, ovaries in rhesus monkeys suffer depletion of follicular reservoir [43]. Whether the adult rhesus monkey ovary exhibits GSCs and neo-oogenesis has not been systematically examined. We rigorously searched for evidence of GSCs and neo-meiosis in ovaries from adult monkeys of various ages, using various experimental approaches.

Materials and Methods

Experimental Design

  1. We attempted to isolate and characterize GSCs from testis using mice and compared with those of female mouse ovaries. Proliferative GSCs were identified by conventional bromodeoxyuridine (BrdU) incorporation, and by strong nuclear PCNA and KI67 expression and absence of nuclear membrane protein LAMIN A, costaining with DAZL or VASA using immunofluorescence microscopy.
  2. We searched in adult monkey ovaries from different age group for proliferative GSCs marked by nuclear PCNA or KI67 colocalized with cytoplasmic DAZL or VASA and additional pluripotency-associated genes LIN28 or SOX2, in comparison with those of adult monkey testis served as positive controls.
  3. We looked for neo-oogenesis in adult monkey ovaries using defined markers specific for early meiocytes, SCP3 lateral element structures ending with perinulcear distribution of telomeres by immunofluorescence microscopy, and compared with the early spermatocytes served as positive controls. SCP3 element structures also were examined for mouse fetal and adult ovaries and testis, to validate the monkey data.
  4. We counted follicles of various stages and measured serum hormone levels from ovaries of monkeys at different age.
  5. We performed experiments to analyze expression of genes specific for primordial germ cells (PGCs), cell proliferation, early meiotic pairing and recombination, and folliculogenesis by conventional PCR, and expression of selected proteins by immunoblot in monkey ovaries compared with testis.

Collection of Tissues

The general care and housing of rhesus monkeys were provided by Guangxi Xiong Sen Primate Experimental Animal Development Corporation. The use of monkeys and mice for this study and the protocol for collecting monkey and mouse tissues were approved by the Institutional Animal Care and Use Committee at Nankai University. A minimal number of monkeys were used in this study to minimize ethical concerns. Female monkeys were randomly chosen from three groups according to their reproductive ages at 3–4 years (young), 7–8 years (middle-age), and 18–19 years (old), with three monkeys in each age group. Male monkeys were approximately 5 years old. Female and male rhesus monkeys were euthanized, and ovaries, kidney, liver, serum, and testis were collected. Half of each ovary and testis were fixed in 4% paraformaldehyde (PFA) for 24 hours at room temperature and then embedded in paraffin. Another half of each ovary and testis, liver and kidney were cut into small pieces, snap-frozen in liquid nitrogen, and stored at −80°C for RNA, DNA and protein extraction, and serum for hormone assays was stored at −80°C.

Isolation and Purification of Mouse Testicular Cells

Isolation of testicular cells was performed based on methods previously published [44, 45], with slight modifications. Testis were isolated from 4 and 32-week-old mice and washed in phosphate-buffered saline (PBS) containing 3% penicillin and streptomycin. Testis tissues were dissected and minced by sterile surgical instrument, rinsed with PBS, and centrifuged (Eppendorf, Hamburg, Germany) at 1,000 rpm for three times to remove the blood cells. The tissue was placed in a Petri dish containing 1 mg/ml Collagenase IV (GIBCO, Grand Island, NY) and 10 μg/ml DNase I (QIAGEN, Chatsworth, CA), digested at 37°C in 5% CO2 incubator for 15 minutes, and then centrifuged at 1,000 rpm for 3 minutes. The tissue was digested in 1 mg/ml Dispase II (Roche, Berlin, Germany) and 10 μg/ml DNase I for another 15 minutes and filtered with a 40-μm diameter filter. After centrifugation at 1,200 rpm for 3 minutes, a small mass of red blood cells at the bottom of the pellet was carefully removed and the pellet was resuspended in SSCs medium without growth factor, the cells counted, and prepared in suspension at the concentration of 1 × 107 cells per milliliter.

The isolated testicular cells were cultured in 100 mm dish and purified by a differential plating method for three times [46]. After culture in incubator for 4–8 hours, Sertoli cells and fibroblast cells adhere to the bottom of the dish while SSCs could not. SSCs were resuspended carefully by pipeting and transferred to a new gelatin-coated dish. The procedure was repeated twice such that SSCs were enriched, and the attached cells were cultured further. Both SSCs and attached cells were collected and washed with Hanks' balanced saline solution (HBSS).

Isolation of Ovarian Cells

Isolation of ovarian cells was performed based on the method described [15]. Briefly, ovaries from six mice of 7–8-week were pooled and dissociated by mincing, followed by a two-step enzymatic digestion involving a 15-minute incubation with 800 U/ml collagenase (type IV; prepared in HBSS) and a 10-minute incubation with 0.05% trypsin-EDTA. Digestions were carried out in the presence of 1 μg/ml DNase I (Sigma-Aldrich, St. Louis, MO) to minimize stickiness within the cell preparations, and trypsin was neutralized by addition of 10% fetal bovine serum (Hyclone, Logan, UT). Ovarian dispersants were filtered through a 70-μm nylon mesh, and the filtrate was centrifuged at 1,200 rpm/minute and washed with HBSS.

Flow Cytometry Analysis

For each experiment, ovarian cells, SSCs, and attached cells were designed for both DDX4 staining and negative control (staining without the first antibody). After washing, cells were blocked in a solution composed of 1% fatty-acid-free bovine serum albumin with either 1% normal goat serum in HBSS for 20 minutes on ice. Cells were then reacted for 20 minutes on ice with a 1:10 dilution of primary antibody (DDX4, ab13840). After washing with HBSS, cells were incubated with a 1:500 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor 488 (Invitrogen, Carlsbad, CA) for 20 minutes on ice, and washed with HBSS. Labeled cells were then filtered again (35-μm pore diameter) and subjected to flow cytometry analysis using a BD FACSCalibur flow cytometer (BD Biosciences, San Jose, CA), gated against negative controls. Ovarian cells and SSCs that were DDX4 positive were sorted according to the method used by White et al. [15], and cell cycles were analyzed using a BD FACSCalibur flow cytometer (BD Biosciences).

In Vivo BrdU Incorporation and BrdU Staining

BrdU incorporation assay was performed according to a method described by Staszkiewicz et al. [47]. Briefly, 8–10-week-old female and male C57BL/6J mice (n = 3) were injected intraperitoneally with BrdU at a concentration of 50 μg/g of b.wt. (Sigma Co., St. Louis, MO) twice daily, at 7:00 a.m. and 6:00 p.m. for 3 consecutive days. Control animals (littermates; n = 2) were injected with saline. The ovaries and testis were collected and fixed in 3.7% PFA 1 day after BrdU injection and further preparation method for section was the same as the method for follicle counts.

Coimmunostaining of BrdU and PCNA, VASA, or LAMIN A was performed based on the methods described [48], with slight modifications. Briefly, sections were deparaffinized, hydrated, treated with hydrogen peroxide (0.3% in PBS) for 10 minutes), subjected to high-pressure antigen recovery sequentially in 0.01% citrate buffer (pH 6.0) for 2 minutes, and denatured in 2 N HCl for 20 minutes at room temperature. The sections were then incubated in 5% normal goat serum, followed by incubation in anti-BrdU (1:200; A21304; Invitrogen) and PCNA (1:400) overnight at 4°C, and incubated with appropriate secondary antibody diluted by 1:200 for 2 hours at room temperature. For costaining with LAMIN A and VASA, the section was incubated in anti-BrdU overnight at 4°C, and then with LAMIN A or VASA overnight at 4°C. After washing in PBS three times, the sections were incubated with the secondary antibodies. The sections were rinsed thoroughly three times and stained with 1 μg/ml Hoechst 33342 for 20 minutes to stain nuclei, washed in PBS three times, mounted in Vectashield (H-1000, Vector Laboratories, Burlingame, CA), and photographed with Zeiss Axio Imager Z1 Carl Zeiss, Oberkochen, Germany.

Immunocytochemistry and Fluorescence Microscopy

Briefly, after deparaffinizing, rehydrating, and washing in 0.01 M PBS (pH 7.2–7.4), sections were incubated with 3% H2O2 for 20 minutes at room temperature to block endogenous peroxidase, subjected to high-pressure antigen recovery sequentially in 0.01% citrate buffer (pH 6.0) for 2 minutes, incubated with blocking solution (5% goat serum in PBS) for 20 minutes at room temperature, and then incubated with the diluted primary antibodies overnight at 4°C. Blocking solution without the primary antibody served as negative control. After washing with PBS, sections were incubated with appropriate secondary antibodies (Alexa Fluor 568, 488, or 594, Invitrogen). The sections were then stained with 1 μg/ml Hoechst 33342 for 20 minutes for reveal of nuclei, washed, mounted in Vectashield (H-1000, Vector Laboratories), and photographed with a Zeiss Axio Imager Z1 (Carl Zeiss). The following primary antibodies were used for immunocytochemistry: VASA (ab13840, Abcam, Cambridge, MA) (1:500), SCP3 (NB 300-232, Novus Biologicals, Littleton, CO) (1:400), KI-67 (AB9260, Millipore, Billerica, MA; and 14–5699-82 eBioscience, San Diego, CA) (1:500), PCNA (SC25280, Santa Cruz, CA) (1:400), TRF2 (05–521, Millipore) (1:400), LAMIN A (ab26300, Abcam), SOX2 (AB5603, Millipore) (1:100), LIN28 (ab46020, Abcam) (1:100), and DAZL (ab134139, Abcam) (1:500).

Testis and ovaries were continuously sectioned as described below, and sections at every 30 section intervals were subject to immunocytochemistry, and carefully examined under ×40 objective to search for putative GSCs as shown by colocalized immunostaining of strong nuclear PCNA, cytoplasmic DAZL, nuclear LIN28, or SOX2, and absence of LAMIN A, and early meiocytes marked by SCP3 filament structure with bouquet clustering and perinuclear distribution of telomere indicated by TRF2. These sections showed consistent results by examination under the fluorescence microscopy. About 14–22 section images from adult ovaries depending on the age group were examined to estimate the number of GSCs and early meiocytes, in comparison with similar number of testis sections served as positive control.

Follicle Counts

Fixed specimens from monkey testis and ovaries were subsequently dehydrated with graded alcohols, cleared in xylene, and embedded in paraffin wax. It was impractical to section the entire ovary of each animal, considering that the aim of the study was to demonstrate ovarian senescence over time, so serial 5-μm sections of half of each ovary were cut and placed on silanized slides. One out of every 20 serial sections was stained with hematoxylin and eosin Y (H&E), and analyzed for the number of follicles in four different developmental stages, and numbers of primordial, primary, secondary, and antral follicles were classified and counted using slightly modified standard methods [43, 49]. Primordial, primary, and intermediate-stage follicles were identified by the presence of an oocyte surrounded by a single layer of flat, squamous, or cuboidal cells. Growing follicles were characterized as having more than one layer of granulosa cells with no visible antrum. Antral follicles possessed small areas of follicular fluid (antrum) or a single large antral space. Only those follicles containing an oocyte with a clearly visible nucleus were scored, so we could not exclude the possibility of losing de novo forming follicles if any.

Immunohistochemistry for Detection of SCs

Briefly, slides were deparaffinized and rehydrated, incubated in 3% H2O2 for 20 minutes at room temperature to block endogenous peroxidase, and incubated sequentially with 5% goat serum for 20 minutes after high-pressure antigen recovery, SCP3 rabbit polyclonal antibody (NB 300-232, Novus) diluted 1:400 in blocking solution at 4°C overnight, and then HRP polymer-goat anti-mouse IgG (Maixin_Bio, Beijing, China) for 15 minutes. Signals were detected by 3,3-diaminobenzidine substrate (Maixin_Bio). Slides were slightly counterstained with hematoxylin (Sigma-Aldrich) and examined using light microscopy. Negative controls were incubated with blocking solution containing no antibodies.

RT-PCR

Cryopreserved ovarian tissues were fully ground in liquid nitrogen and total RNA was isolated using the RNeasy Mini Kit (74104, Qiagen, Valencia, CA). Total RNA was initially treated with RNase-free DNase I (79254, Takara Bio Inc., Shiga, Japan) to remove contaminating genomic DNA, and 1 μg of RNA was used to synthesize the first strand cDNA using the M-MLV Reverse Transcriptase (28025-013, Invitrogen) with oligod(T)18 (D511, Takara) primers. Amplification via 33 cycles of PCR was then performed using Takara Ex Taq Hot Start Version (DRR006, Takara) with primer sets specific for each gene designed with dnaman5.2.2 (supporting information Table S1). Parallel amplification of GAPDH was used to verify cDNA synthesis from total RNA isolated from each tissue. For each cycle, the template was denatured at 94°C for 30 seconds, annealed at 58°C for 30 seconds, and extended at 72°C for 1 minute. Products of RT-PCR were run on 1% agarose gels and visualized by staining with ethidium bromide.

Western Blot

Ovarian tissues and testis or liver were fully ground with glass homogenizer and lysed in SDS sample buffer at 99°C for 10 minutes, and 25 μg of whole-tissue extract proteins was separated on 10% SDS-polyacrylamide gels and transferred to polyvinylidine difluoride membranes (Millipore). Nonspecific binding was blocked by incubation in 5% skim milk in TBST for 2 hours at room temperature. Blots were incubated for 12–16 hours at 4°C with anti-PCNA (Mouse mono-IgG2a; Santa Cruz, sc-25280), (Tris-Buffered Saline and Tween 20), anti-LAMIN A (Rabbit polyclonal; Abcam, ab26300-100), anti-VASA (Rabbit polyclonal; Abcam, ab13840), anti-TRF2 (Mouse monoclonal; Upstate, 05–521), anti-DAZL (Rabbit polyclonal; Abcam, ab34139), or anti-β-ACTIN (Rabbit polyclonal; Santa Cruz, sc1616R), followed by wash for 10 minutes three times. Blots were incubated at room temperature for 2 hours with horseradish peroxidase conjugated monkey-anti-rabbit IgG (GE Healthcare 371624) or goat anti-mouse IgG (H+L) (ZB2305) and washed three times for 30 minutes. Bound antibody was detected using Enhanced ECL Amersham prime western blotting detection reagent (GE Healthcare RPN2232).

Hormone Assays

Serum samples from nine female monkeys were assayed twice for progesterone (P4), testosterone (T), and estradiol (E2). P4, T, and E2 levels were determined by direct radioimmunoassay using commercial kits (R0205PR-B, R0204PR-B, and R0206PR-B, China Diagnostics Medical Corporation, Beijing, China). Quality control serum, sterilized distilled water, and five series diluted standard samples for a standard curve were tested for each serum sample. The intra- and inter-assay coefficients of variability for P4, T, and E2 were below 10% and 15%.

Statistical Analysis

Statistical analyses were performed by ANOVA and means compared by Fisher's protected least-significant difference using StatView software from SAS Institute Inc. (Cary, NC). p-Value <.05 was considered statistically significant.

Results

Identification of GSCs Using Mouse Testis by Live Cell Fluorescence-Activated Cell Sorting and Immunofluorescence

Spermatogonia stem cells (SSCs), GSCs in testes, renew continuously in the male [26-28]. The putative GSCs isolated from adult mouse and human ovaries were selected by fluorescence-activated cell sorting of cells expressing DDX4 (DEAD (Asp-Glu-Ala-Asp) box polypeptide 4, also called VASA or MVH) [15], so we used this same method [15] to isolate DDX4 expressing cells from mouse testes and ovaries. SSCs exhibited very strong DDX4 fluorescence intensity, in the range of 100–1,000. Some cells from testes also showed weaker fluorescence, in the range of 10–100 (supporting information Fig. S1), while other cells expressed only background fluorescence, below 10. These cells appeared to be spermatocytes and somatic cells, respectively. In contrast, cells isolated from adult ovaries exhibited only low levels of DDX4 fluorescence (10–100). Cells with fluorescence in the intense range (100–1,000), like that seen in SSCs, were not detectable in cells extracted from the ovary (supporting information Fig. S1).

Next, we examined mouse testis for markers known to label actively dividing cells and germ cells using immunofluorescence microscopy. Actively dividing stem cells take up BrdU [48]. Labeling for proliferating cell nuclear antigen (PCNA) also indicates the presence of dividing cells, including stem cells [50-52], and serves as an alternative to BrdU incorporation. In testis, SSCs with small nuclei (white arrowheads, around 6 μm) located close to the basal membrane and showed high BrdU incorporation, but they did not label for VASA (supporting information Fig. S2A). SSCs had small nuclei and stained both for BrdU and PCNA (supporting information Fig. S2B). Testes cells showing cytoplasmic VASA staining had relatively large nuclei (white arrows, 9–10 μm) (supporting information Figs. S2C, S3A), lacked BrdU incorporation, and their location and appearance suggest that they are primary spermatocytes. Some spermatocytes also stained for diffuse, weaker PCNA. It appears that VASA marked germ cells, but not necessarily GSCs. Indeed, VASA protein is not expressed in oogonia or gonocytes during the first trimester human fetal ovary [53]. These data suggest that VASA may not reliably identify proliferative GSCs in monkeys or humans.

Nuclear LAMIN A is expressed in differentiated cells, but not in somatic stem cells [33, 54, 55]. Thus, absence of LAMIN A could provide an additional marker to identify stem cells. We verified that BrdU-positive nuclei lacked LAMIN A, while basal membrane cells showed LAMIN A staining in mouse testis (supporting information Fig. S2D). Notably, DAZL (deleted in azoospermia-like), a germ cell marker, labeled cells with both small (with BrdU incorporation and PCNA staining) and large nuclei (without BrdU incorporation, and with only diffuse PCNA staining) (supporting information Fig. S2E, S2F). Together, male mouse GSCs have relatively small nuclei, about 6 μm in diameter, incorporate nuclear BrdU, lack nuclear LAMIN A, and stain for nuclear PCNA and cytoplasmic DAZL. These criteria could be used to identify GSCs (supporting information Fig. S3B).

GSCs Are Not Detectable in Monkey and Mouse Ovaries

KI67 also labels proliferative cells, including stem cells [56-59]. Monkey spermatogonia (GSCs) with round or oval shaped nuclei stained strongly for both KI67 and PCNA (Fig. 1A), and also exhibited cytoplasmic DAZL staining (arrowheads) (Fig. 1B). Meiocytes or spermatocytes showed punctuate staining for KI67 and strong cytoplasmic staining of DAZL. Strong nuclear staining for KI67 in proliferative cells also was found in the cortex of ovaries from monkeys of various ages, but these cells showed no cytoplasmic staining for DAZL (arrowheads) (Fig. 1B), although punctuate staining of KI67 also was found in oocytes with weak peripheral staining for DAZL in primordial and primary follicles. By careful examination of sections processed by immunocytochemistry, we estimated that an average of 24.3 ± 3 (mean ± SE) cells/section (refer to Materials and Methods) were positive for both PCNA and DAZL which may represent actively dividing GSCs in adult monkey testis. In contrast, putative GSCs by these criteria were not detectable in all immunostained sections from adult monkey ovaries for each age group. In addition, some oocytes showed weak staining for DAZL in young and middle-aged adult monkey ovaries, but not in old monkey ovaries (Fig. 1B). Spermatocytes at different stages also showed punctuate nuclear staining for KI67 and strong cytoplasmic staining for VASA, and oocytes with these properties were found in monkey ovaries (supporting information Fig. S4). Some granulosa cells of growing follicles showed strong KI67 staining, indicative of active proliferation, but they were negative for VASA or DAZL. Together, these observations demonstrate that GSCs homologous to SSCs in testis by these specific criteria were undetectable in adult monkey ovaries.

Figure 1.

Germline stem cells (GSCs) found in adult monkey testis but not in adult monkey ovaries by coimmunostaining of DAZL and KI67. (A): Colocalized immunostaining of PCNA with KI67 in spermatogonial stem cells (SSCs) of monkey testis. (B): Identification of SSCs (GSCs) by colocalized strong positive staining of KI67 and cytoplasmic DAZL staining in small nuclei from monkey testis, in contrast to the absent cytoplasmic DAZL staining and strong KI67-positive small nuclei (white arrowheads) of adult monkey ovaries. Note, spotted KI67 pattern in the oocyte appears to be similar to that of spermatocytes. Scale bar = 20 μm.

Likewise, staining for strong nuclear PCNA and cytoplasmic DAZL also was found in monkey SSCs localized to the peripheral seminiferous tubules (Fig. 2). Comparatively, early spermatocytes in monkey testes showed diffuse, weak staining for PCNA and strong cytoplasmic DAZL staining. Small nuclei (<10 μm) with strongly positive PCNA staining also were found in ovaries of young, middle-age, and old monkeys, but these cells had no cytoplasmic DAZL staining colocalized with the nuclear PCNA (Fig. 2), suggesting that these cells represent proliferative cells but not necessarily GSCs.

Figure 2.

Immunocytochemistry of markers for germ cells and meiocytes by coimmunostaining of DAZL and PCNA in adult monkey ovaries compared with testis. While meiocytes and germ cells in the testis are positive for DAZL in the cytoplasm and diffusely and lightly stained by PCNA in the nucleus, spermatogonia stem cells are strongly stained by PCNA and weakly stained by DAZL in the cytoplasm. Nuclei strongly stained for PCNA in monkey ovarian cortex do not exhibit typical cytoplasmic DAZL staining, in contrast to spermatogonial stem cells in the adult testis. Young monkey at the age of 3–4 years; middle-aged monkey at 7–8 years; old, reproductive aging monkeys at 18–19 years. Scale bar = 20 μm.

In adult mouse ovaries, cytoplasmic VASA staining appeared only in cells with large nuclei (white arrow) not incorporating BrdU. This antibody more likely labels primary oocytes (supporting information Fig. S5A). Notably, nuclei positive for both BrdU and PCNA appeared in the cortex of mouse ovaries (indicated by arrowhead), with some nuclei showing diffuse PCNA but negative for BrdU (arrows) (supporting information Fig. S5B). Strongly PCNA-positive nuclei lacked cytoplasmic VASA (arrowhead), but oocytes stained for cytoplasmic VASA and diffuse PCNA (arrow) (supporting information Fig. S5C). Some cells incorporated BrdU and expressed LAMIN A (arrow) (supporting information Fig. S5D), consistent with their differentiated cell types. BrdU-positive cells lacking nuclear LAMIN A (white arrowheads) were found in adult mouse ovaries, but these BrdU-positive cells lacked cytoplasmic DAZL staining (yellow arrows, supporting information Fig. S5E), as did the strongly PCNA-positive cells in the epithelia (yellow arrows) (supporting information Fig. S5F), providing evidence against their germ cell origin and consistent with these cells being somatic stem cells.

Pluripotency-associated gene LIN28 orchestrates PGC specification [60], and is strongly expressed in spermatogonia in mice, non-human primates, and human, maintains the GSC state in the developing human ovary, and might be a marker for rare GSCs [61-63]. Likewise, SOX2 may also mark GSCs in the testis [64]. LIN28 was found to express in the nuclei, colocalized with PCNA and in the cytoplasm of putative GSCs in the monkey testis, whereas PCNA-positive cells were found in adult monkey ovaries, but did not show specific nuclei and cytoplasmic staining pattern for LIN28, unlike GSCs in the testis (supporting information Fig. S6A). Also, nuclear staining of SOX2 colocalized with strong nuclear staining of PCNA was found in the monkey testis, but no specific nuclear SOX2 coimmunostaining with strong PCNA in ovaries (supporting information Fig. S6B).

Neo-Oogenesis Is Absent in Adult Mouse and Monkey Ovaries

During early meiosis in both males and females, telomere bouquet clustering and perinuclear distribution are associated with homologous bivalents identified by SC protein 3 (SCP3) elements. These structures are essential for homologous chromosome pairing and synapsis and provide definitive markers for prophase I of meiosis [36, 38-40, 65], and their presence in adult ovaries would provide evidence of neo-meiosis. We validated the presence of neo-meiosis in adult monkey testis by coimmunostaining of DAZL with the telomere-associated protein TRF2 (supporting information Fig. S7A), or coimmunostaining for SCP3 and TRF2 (supporting information Fig. S7B). DAZL was strongly expressed in the cytoplasm of spermatocytes, which had perinuclear, bouquet clustering of telomeres, as shown by TRF2 immunostaining (supporting information Fig. S7A). DAZL also stained the cytoplasm of premeiotic cells, with telomeres centered in the nuclei. Early spermatocytes exhibited typical perinuclear distribution of telomeres and distinct SCP3 lateral filaments connecting telomeres, as shown by TRF2 at termini of SCP3 filaments during pachytene stage, and bouquets at the leptotene stage (supporting information Fig. S7B), and thus are distinguishable from GSCs.

DAZL staining was found in mouse early spermatocytes, meiocytes, and to a lesser extent in oocytes at different stages (supporting information Fig. S8). Short, punctuate foci of SCP3 immunostaining were found in fetal testis, with increased staining in day 10 testis. SCP3 immunostaining formed lateral filament structures in adult mouse testis (supporting information Fig. S9), like those of adult monkey testis. Aggregates of SCP3 presumably represent degraded or fragmented SCP3 proteins, appearing after prophase I of meiosis. Alternatively, insufficient levels of SCP3 proteins may have failed to assemble distinct lateral filament structures, forming punctuated and fragmented SCP3 staining. Likewise, no distinct SCP3 lateral filaments but punctuate SCP3 staining were found in mouse newborn day 10 and adult ovaries, while distinct staining of SCP3 lateral filaments found in early meiocytes of mouse fetal ovaries (supporting information Fig. S9), consistent with the notion that neo-meiosis arises in fetal ovaries and adult testis, but not in postnatal ovaries.

Adult monkey testes undergo continuous early meiosis, as shown by perinuclear telomere TRF2 distribution and homologous pairing indicated by SCP3 filaments, but these characteristic early meiocytes were absent in adult monkey ovaries, regardless of age (Fig. 3). We estimated an average of 46.4 ± 0.7 early meiocytes/section (refer to Materials and Methods) in adult monkey testis, but none in all immunostained sections from adult monkey ovaries for each age group, substantially different from those of testis by statistical analysis (SAS). Only fragmented and punctuate SCP3 staining was found in oocyte nuclei, resembling postnatal day 10 and adult mouse ovaries (supporting information Fig. S9). Immunohistochemistry using anti-SCP3 antibody identified SCP3 in germ cells at various stages as well as in OSE cells, but the morphology of these SCP3-positive structures in adult monkey ovaries differed from that of testes and fetal ovaries-no clear SCP3 containing filamentous structures could be seen (supporting information Fig. S10). Immunohistochemistry does not permit resolution of SCP3-bearing filamentous structures, and thus cannot image meiotic synapsis and homologous pairing. Further, we performed additional experiments by coimmunostaining of TRF2 to indicate telomeres and MVH/VASA to indicate germ cells. Again, meiocytes with perinuclear distribution of telomeres were evident in adult testis, but not in adult monkey ovaries (supporting information Fig. S11). Together, the results of these experiments are not consistent with neo-oogenesis in adult monkey ovaries.

Figure 3.

Coimmunostaining of SCP3 and TRF2 in adult monkey ovaries compared with testis. SCP3 lateral filaments and the telomere perinuclear distribution at the termini of the SCP3 filaments in meiocytes likely at the pachytene stage are found in testis but not in adult monkey ovaries. H&E shows histological sections. Inset, magnified image of a neo-meiocyte showing SCP3 axis filaments (red) attached with telomeres (green). Scale bar = 20 μm.

Proliferative Somatic Stem Cells Found in Adult Monkey and Mouse Ovaries

While searching for GSCs in adult monkey ovaries, we unexpectedly found cells with characteristics of somatic stem cells in the adult ovary. LAMIN A is expressed exclusively in differentiated cells, including Sertoli, theca, epithelial cells as well as germinal vesicle oocytes of mouse testis and ovaries (supporting information Figs. S2, S12). In the OSE and ovarian cortex of mouse postnatal day 10 and adult ovaries, some cells stained for PCNA, while others stained both for PCNA and LAMIN A, suggestive of a differentiated state. Granulosa cells from growing follicles also were positive for PCNA, but not for nuclear LAMIN A, whereas oocytes all expressed LAMIN A. As a positive control for GSCs, monkey spermatogonia stained strongly for PCNA and negatively for LAMIN A, whereas spermatocytes stained diffusely and weakly for PCNA, but not for LAMIN A (Fig. 4; supporting information Fig. S13).

Rare cells in the cortex or OSE of monkey ovaries stained strongly positive for PCNA and lacked LAMIN A (Fig. 4), but these cells did not stain for cytoplasmic DAZL (Fig. 2), so these cells most likely were somatic stem cells, not GSCs. Primary oocytes in ovaries from young and middle-aged monkeys diffusely stained for PCNA and exhibited punctuate staining for KI67, and were negative for LAMIN A (supporting information Figs. S13, S14). Most OSE cells, ovarian cortex, and the single theca cell layer of primordial and primary follicles strongly stained for LAMIN A, but not for PCNA or KI67, consistent with their differentiated state. In contrast, granulosa cells from growing follicles exhibited proliferative, stem cell-like properties, as evidenced by staining strongly for PCNA and KI67, but negative for LAMIN A. KI67 and LAMIN A costaining of ovaries from monkeys of different ages revealed rare LAMIN A-negative, KI67-positive cells in the cortex and tunica albuginea (arrows, supporting information Fig. S15). Consistently, mouse ovaries contained rare proliferative cells with stem cell properties in the cortex, as shown by BrdU incorporation and LAMIN A-negative staining (supporting information Fig. S16). Nuclei of these cells in monkey ovaries were smaller (4–5 μm) than testis GSCs (6 μm). Since these cells did not stain for DAZL, they are likely somatic stem cells, not GSCs.

Figure 4.

Proliferative somatic stem cells identified by coimmunostaining of LAMIN A with PCNA in ovaries of monkeys at different age compared with adult testis from a 5-year-old monkey. Proliferative spermatogonia are strongly stained by PCNA without LAMIN A in the nuclear envelope (white arrows), in contrast to strong LAMIN A positive but PCNA-negative staining in Sertoli cells and basal membrane cells. Most cells in the ovarian cortex are strongly stained by nuclear envelope LAMIN A, but few PCNA strongly positive cells are negative for LAMIN A (white arrows). Scale bar = 20 μm.

To test whether these cells originated from hematopoietic stem cells (HSCs), we performed immunocytochemistry using anti-CD45 and -CD34 antibodies. As expected, HSCs exhibited CD45 and CD34-positive staining in their cell membranes, but the monkey ovarian cells showed no specific membrane staining for CD45 and CD34 (supporting information Fig. S17). Background staining for CD45 did not colocalize with KI67 staining. Together, these data suggest that rare proliferative stem cells exist in the cortex of adult ovaries, but that these stem cells are not of germ cell origin, and therefore they are not GSCs.

Age-Associated Ovarian Senescence in Monkey

Despite the extensive search, we failed to find compelling evidence for GSCs and early meiocytes in adult monkey ovaries. We hypothesized that follicular renewal should be evident in adult ovaries, if GSCs existed and neo-oogenesis took place in the adult monkey. Hence, we examined follicle reserve by counting follicles, measuring serum sex steroid hormone levels, and analyzing expression of genes related to senescence in ovaries of monkeys at various ages.

Ovaries from old monkeys contained only rare follicles. The numbers of primordial, primary, and secondary follicles were reduced markedly in ovaries of old monkeys (18–19 years) compared with those of young (3–4 years) and middle-aged (7–8 years) monkeys (Fig. 5A, 5B). Only few follicles and no mature or antral follicles could be found in ovaries from old monkeys. The number of primordial, primary, and secondary follicles was significantly greater in ovaries from 3- to 4-year-old than from 7- to 8-year-old monkeys. Fewer antral follicles appeared in ovaries from 3- to 4-year-old than in ovaries from 7- to 8-year-old monkeys.

Figure 5.

Ovarian ageing as evidenced by follicle depletion and decreased levels of sex hormones in monkeys with increasing age. (A): Representative morphology of primordial and primary, secondary, and antral follicles in the ovaries of monkeys at various reproductive ages (young, Y, 3–4 years; middle-age, M, 7–8 years; and old, O, 18–19 years). For young and middle-age monkey ovaries, (a) indicates primordial and primary follicles, (b) secondary, and (c) antral follicles. For old monkey ovaries, (a) indicates rare primary follicles, (b) secondary, and (c) the absence of antral follicles. Scale bar = 100 μm. (B): Number of ovarian follicles at various developmental stages from monkeys at different reproductive ages. (C): Levels of serum steroid hormones progesterone, testosterone, and estradiol from monkeys at various age. (D): Relative expression levels of p21 and p16 in the monkey ovaries. n = 3 (mean ± SEM). Different superscripts shown above the bars indicate significant differences (p < .05).

Levels of serum progesterone P4 were higher in the 18–19-year-old compared to other monkeys. Levels of testosterone (T) were increased from young (3–4 years) to middle-age (7–8 years), but dropped significantly in old monkeys (18–19 years). Similarly, levels of estradiol E2 were higher in 7–8-year-old than in younger and older monkeys (Fig. 5C), consistent with their increased number of antral follicles (Fig. 5B). Levels of T and E2 were much lower in monkeys at 18–19 years of age. Follicle depletion, with reduced levels of E2 and T, corroborated with an age-related reduction of ovarian function in aging monkeys. Furthermore, expression of genes associated with cell senescence, p21 and p16 [66, 67], was higher in ovaries from old than from young and middle-aged monkeys (Fig. 5D). These data further support the notion that the absence of GSCs and neo-oogenesis is associated with ovarian aging in monkeys.

Furthermore, if GSCs and neo-oogenesis persist in adults, their ovaries should express genes encoding proteins specific to premeiotic or early prophase stages of oogenesis [7]. Early germ cell development is controlled by a number of genes, including NANOS3 [68], PRDM1 (also known as Blimp1) [69], PRDM14 [70], and DAZL [71]. Msh5, Dmc1, and Scp1–3 are crucial for homologous pairing and synapsis. PRDM9 [9, 72] and Spo11 are involved in double-strand break formation.

To confirm the previous findings obtained by flow cytometry and immunofluorescence, we performed PCR analysis and validated that stem cell and early meiosis genes are expressed in testis, but not or at minimal levels in adult ovaries. PRDM1 was expressed in monkey ovaries of all ages, but NANOS3 and PRDM14 only at low or minimal levels in adult monkey ovaries unlike testis (Fig. 6A). Expression of genes for homologous recombination, SPO11, PRDM9, DMC1, and REC8, was low or undetectable in ovaries from monkeys of all ages, and SCP3 or SCP1 mRNA levels were very low or absent in young and middle-aged monkey ovaries, in contrast to testis (Fig. 6B, 6C). These results supported the immunocytochemistry data, and together show that activities necessary for meiotic synapsis from leptotene to pachytene and recombination at prophase I unlikely take place in adult monkey ovaries. Genes for germ cell development (OCT4, DPPA3, and VASA) (Fig. 6D) and cell proliferation (TERT and KI67) (Fig. 6E) were expressed at higher levels in young and middle-age than in old monkey ovaries.

Figure 6.

mRNA and protein levels of genes required for proliferation, meiosis, and germ cell development in monkey ovaries compared with other tissues by RT-PCR analysis (A–G) and Western blot (H). (A) Markers for primordial germ cells formation, migration, and specification; (B) markers of double-strand breaks; (C) markers for homologous pairing and meiotic synapsis; (D) markers of germ cells; (E) markers for cell proliferation; (F) gene related to follicular development; (G) housekeeping gene and negative control. GAPDH was coamplified as an internal control. MOCK, mock transcribed RNA samples. Y, young monkey (age 3–4 years); M, middle-aged monkey (7–8 years); O, old monkeys (18–19 years). Liver and kidney served as negative controls for meiosis and germ cell development. (H) Protein expression required for cell proliferation and germ cell development by Western blot using antibodies against PCNA, TRF2, LAMIN A, VASA, and DAZL. β-ACTIN served as loading control.

Notably, NOBOX, involved in follicular development [73], was expressed in young and middle-aged monkey ovaries, but its expression was very low or absent in testis and ovaries from old monkeys (Fig. 6F), consistent with the active follicular development in young and middle-aged ovaries, but not in testis or old monkey ovaries (Fig. 5). Moreover, expression of DAZL, VASA, and KI67 was detected in young and middle-aged monkey ovaries, but only minimally in ovaries from old monkeys. At the protein level, PCNA was significantly enriched in testis, and reduced greatly in ovaries from old monkeys. TRF2 protein also was highly expressed in testis but much reduced in monkey ovaries (Fig. 6H). By contrast, LAMIN A levels were high in adult ovaries. Notably, DAZL and VASA protein levels were remarkably higher in testis than in ovaries. Liver served as a negative control did not express DAZL and VASA, as expected (data not shown), in agreement with the PCR data. These data are not consistent with GSCs and early germ cell formation in adult monkey ovaries.

Discussion

We are unable to find evidence for proliferative GSCs and germ cell renewal in adult monkey and mouse ovaries. The lack of neo-oogenesis is further substantiated by the universal observation of robust ovarian failure in reproductively aged monkeys confirmed here. Continuous meiosis and germ cell renewal, as occurs in testes, relies on a constant supply of proliferative GSCs, but our experiments show no evidence of these in ovaries from mice or monkey. Consistently, ovaries from female rhesus monkeys approaching or undergoing the menopausal transition (about 20 years of age) demonstrated evidence of ovarian senescence, with decreasing numbers of primordial follicles [43], concomitant hormone changes, and a molecular signature of cell senescence. Ovaries from old monkeys contain very few follicles, in association with their reduced levels of estrogen and testosterone. These data are consistent with ovarian senescence in older monkeys, similar to that observed in women, and absence of neo-oogenesis which presumably contributes to the ovarian senescence. However, our data cannot exclude the possibility that dormant GSCs may exist in adult females and are activated to replicate by ovotoxic damage [74, 75].

We did find evidence of proliferative stem cells in the cortex of adult ovaries, but these were somatic rather than GSCs. Interestingly, granulosa cells also showed proliferative, stem cell-like properties, consistent with recent findings [76]. These proliferative somatic stem cells in the adult ovaries exhibit strong positive staining for PCNA, but negative for LAMIN A (Fig. 4). GSCs in the adult testis are positive for SOX2 and LIN28, but negative for nuclear LAMIN A. However, cells positive for SOX2 or LIN28 were not found in adult monkey ovaries, regardless of PCNA-positive staining (supporting information Fig. S6A, S6B). It is unlikely that the proliferative somatic stem cells found in adult monkey ovaries are positive for markers of SOX2 and LIN28. In the ovarian cortex, the small nuclei with diameters of 4–5 μm were found, but were negative for markers of hematopoietic stem cells (CD45 and CD34). These cells might be related to very small embryonic-like stem cells found in adult human tissues and organs, for example, bone marrow [77, 78]. Similar population of cells was recently found in adult human ovaries [24, 79]. It is not excluded that these small putative “somatic” stem cells are the prestage of GSCs. The nature of these somatic stem cells in the ovary remains to be determined, but presumably they replenish granulosa cells and ovarian cortex, both highly proliferative in the ovary. The lack of new meiocytes, marked decline in follicle number, and elevated molecular signature of senescence in the aging ovary suggest that these cells lack significant capacity to undergo neo-oogenesis and folliculogenesis in vivo.

Together, GSCs, defined as proliferative stem cells that stain for nuclear PCNA, KI67, and LIN28 and cytoplasmic DAZL but not for LAMIN A, are readily found in testis but not in ovary (Table 1). Neo-oogenesis as marked by SCP3 lateral filaments associated with perinuclear telomere distribution at termini also readily appears in adult testes but not in ovaries, yet these decisive structures marked for homologous pairing in early meiocytes were not demonstrated, but only SCP3 punctuate staining shown in previous studies [14, 15]. We found SCP3 punctuate staining in the primary oocytes.

Table 1. Comparison of markers for GSCs and neo-oogenesis in testis and ovaries
Male/femaleGSC markersNeo-oogenesis markers
KI67PCNALAMIN ADAZL/LIN28VASASCP3 lateral filamentsTelomere TRF2 perinuclear distributionDAZLVASA
  1. N/A: TRF2 antibody was not specific to mouse cells, or NA data not available. ++, strong staining; +, positive staining; , negative staining; +/, few small cells negative staining.

  2. Abbreviation: GSC, germline stem cell.

Monkey
Adult testis+++++/+++++++
Ovary         
Young+++/−/++
Middle-age+++/−/++
Old+++/−/
Mouse
Fetal ovary+++/NA+N/A++
Adult testis+++++/NA+N/A++

Conclusions

The well-established paradigm of female reproduction is that the number of oocytes and the pool of follicles are fixed at birth and continuously decline to the point when few oocytes are available after menopause, and the supply of follicles dictates the length of her reproductive lifespan [37, 80]. The oocytes themselves orchestrate and coordinate ovarian follicular development [81]. By rigorous searching using a variety of experimental approaches, we do not demonstrate GSCs or neo-oogenesis in adult monkey ovaries, although our data also cannot rule out the possibility that experimental conditions could reprogram ovarian adult somatic stem cells to germ cells, and also the possible existence of LAMIN A-positive female GSCs cannot yet be refuted, yet they do support the existing paradigm that germ cells and follicle reserve are not replenished in vivo by neo-oogenesis in adult ovaries [8, 9, 11, 16, 18, 82-84]. They also raise the distinct possibility that prior studies may have confounded a mixture of proliferative somatic stem cells and primary oocytes, which do normally reside in the ovary, with GSCs.

Acknowledgments

We thank Shouquan Zhang, Xinglong Zhou, Yan Geng, Lixin Feng, Yifei Liu, Xiangdong Tang, Jing Li, Yuanli Chen, and Yahui Ding for help with the materials, experiments, or discussion. This work was supported primarily by the China National Basic Research Program (2010CB94500 and 2011CBA01002 to L.L., 2012CB911202 to N.L., 2011CB964801 to T.C.), and Natural Science Foundation of Tianjin (12JCZDJC24800, L.L.).

Disclosure of Potential Conflicts of Interest

The authors indicate no potential conflicts of interest.

Ancillary