Neonatal Mouse Ovary Culture and Whole-mount Follicle Quantification: An Efficient In Vitro Approach for Studying Primordial Follicle Regulation

Oogenesis is a highly regulated and complex process that results in the production of mature oocytes. In mammals, oogenesis initiates in fetal ovaries, where primordial germ cells (PGCs) differentiate into primary oocytes. Each primary oocyte is enclosed by a layer of squamous pregranulosa cells, forming a primordial follicle. Most of these primordial follicles become dormant after formation, serving as the ovarian reserve^1,2,3. In the adult ovary, periodic activation of primordial follicles for follicular development, known as folliculogenesis, is essential for sustaining the production of mature oocytes and ovarian steroid hormones ^4,5. In the absence of activation signals, human primordial follicles can remain quiescent for up to 50 years. The prolonged dormancy may contribute to reduced oocyte quality, as well as an increased rate of primordial follicle loss in aging ovaries⁶. The mechanisms underlying how a specific pool of follicles becomes activated while others stay quiescent and undergo periodic cell death remain open questions.

The process of oogenesis is highly conserved in mammals, making mice an ideal model for studying mammalian oogenesis⁷. In mice, primordial follicle formation is complete by postnatal day 4 (P4). A small proportion of primordial follicles undergo follicular development immediately after their formation. During follicle development, the oocyte increases in size due to enhanced organogenesis, and mRNA and protein synthesis⁴. Squamous granulosa cells transition to cuboidal granulosa cells and become proliferative⁸. These developing follicles grow to the ovulatory stage around day 21, which is when female mice reach puberty. This phase of follicular development in postnatal ovaries is commonly referred to as 'first-wave folliculogenesis.' Since first-wave folliculogenesis closely mirrors the process of follicle development, postnatal mouse ovaries provide an ideal model for studying primordial follicle regulation and ovarian folliculogenesis⁹.

The classification of ovarian follicles is primarily based on the nuclear morphology of follicle somatic cells and the size of the oocyte. In mouse ovaries, the dormant primordial follicle is identified by a single layer of squamous pregranulosa cells that encase a primary oocyte, which has a diameter of approximately 20 µm. Developing follicles can be categorized into primary, secondary, tertiary, and antral follicles. The primary follicle contains an oocyte that is typically over 25 µm in diameter and is surrounded by a single layer of cuboidal granulosa cells (Supplementary Figure 1). The secondary follicle features two layers of granulosa cells, while the tertiary follicle comprises three or more layers of granulosa cells. The antral follicle is larger than the tertiary follicle, contains multiple layers of granulosa cells, and has a fluid-filled cavity known as the antrum. As follicles develop, the oocyte increases in size, reaching approximately 80 µm by the time of ovulation. Additionally, the developing follicles are encircled by several layers of thin theca cells, which are located outside the granulosa cells^10,11,12.

The relatively small size of postnatal mouse ovaries makes whole ovary culture a practical approach for research. This method effectively enables the study of ovarian and follicle development within an intact ovary, as it preserves the physical and physiological microenvironments while minimizing interference from surrounding tissues¹². This approach allows for experiments that would be challenging to conduct in in vivo models. Examples include live imaging of ovarian development, time-controlled multi-drug treatments, and the analysis of ovarian secretory activity through protein profiling of the culture media^13,14. Furthermore, this approach can be applied to investigate the effect of secretory factors without direct cell-cell interaction through co-culturing experiments. In these experiments, different types of tissues can be placed on separate membrane inserts within a shared medium¹⁵. Additionally, using conditional gene-knockout mouse ovaries in vitro could help elucidate key mechanisms involved in primordial follicle activation and growth.

In this article, we introduce methods for culturing postnatal mouse ovaries using membrane inserts and for whole-mount follicle quantification in an entire ovary (Figure 1). Techniques outlined here include 1) dissection of neonatal mouse ovaries, 2) culture of neonatal mouse ovaries, 3) media change during culture, 4) tissue fixation and whole-mount antibody staining, and 5) tissue imaging and follicle quantification. Additionally, this study employed doxorubicin to experimentally induce primordial follicle loss, providing a model to investigate the mechanisms underlying ovarian reserve depletion. The results showed that the number of ovaries per insert and their positioning within the insert during culture influenced primordial follicle counts. This highlights the importance of maintaining a consistent culture setup to prevent non-biological variations in the experimental outcomes.

CD-1 male and female mice were housed together at a 1:1 ratio as breeding pairs. The date of birth of new pups was designated as postnatal day 0. All animal experiments were approved by the Institutional Animal Care & Use Committee (IACUC) at the University of Missouri (protocol number: 36647).

1. Reagents and Culture Media

1. All reagents used in this study are listed in the Table of Materials. Prepare all materials and media before ovary culture.
2. Culture media: Make ~5 mL of DMEM/F-12 media supplemented with 3 mg/mL bovine serum albumin (BSA), 10% fetal bovine serum (FBS), and 10 mIU/mL follicular-stimulating hormone (FSH).
   NOTE: The media should be freshly prepared by adding the BSA (30 mg/mL), FBS, and FSH stocks (10 IU/mL) to DMEM/F-12 media. Equilibrate the media in a petri dish for about 30 min in a 5% CO₂ incubator at 37 °C. Store the stocks at -20 °C and thaw them immediately before use. Aliquot stocks in proper volumes to avoid repeated freezing and thawing of the stocks.

2. Ovary culture

1. Euthanize P5 female mice by decapitation. Remove the ovaries and the attached bursa tissue using surgical scissors and forceps. Place the ovaries in a 60 mm petri dish with pre-chilled Dulbecco's phosphate-buffered saline (DPBS) containing Antibiotic-Antimycotic (Figure 2A,B).
2. Using a pair of insulin syringes (1 mL), dissect and remove the bursa tissue from the ovary under a stereoscope (Figure 2C,D). Transfer the ovaries using a 1 mL pipette to a 30 mm petri dish containing 3 mL of culture media and place the dish in a CO₂ incubator.
   NOTE: Avoid poking or squeezing the ovaries. Make clean and uniform incisions through the ovarian bursa using the insulin syringes to release the ovary. Prefill the pipette tip with approximately 50 µL of media, place the pipette tip next to the ovary, and aspirate it into the tip for transfer.
3. Under the laminar flow hood, prepare the plate by placing 400 µL of media and an insert in each well (Figure 2E,E'). The media is outside the insert, and the membrane of the insert turns opaque once it is soaked by the media.
4. For the culture with 24 h Dox treatment, prepare media containing Dox at 0.1 mg/mL concentration freshly by diluting Dox stock (10 mg/mL) with the media.
   NOTE: Media containing Dox should be disposed of properly in accordance with the guidelines of the institutional committees for biosafety and environmental health and safety.
5. Under a stereoscope, transfer up to three ovaries to an insert using a 1 mL pipette (Figure 2F,G). Minimize the amount of media transferred into the insert and immediately remove any excess media after transfer. Once the ovaries are gently placed in the insert, use a 200 µL pipette to transfer approximately 200 µL of media from the well into the insert, ensure the membrane is completely covered by the media. Transfer the media back to the well and center the ovaries on the membrane. Fill the pipette tip with culture media to prevent the ovary from being sucked into the tip when centering it. Make sure to leave adequate space between the ovaries and that they are stabilized on the membrane insert (Figure 2H), rather than remaining suspended in the medium (Figure 2I). Take care not to poke the ovaries and the membrane.
6. In a 5-day culture, change media every two days by removing 200 µL of media (half of the total amount) from the well and adding 200 µL of fresh culture media back to the well (Figure 2J).
7. For the culture with 24 h drug treatment, transfer the insert with ovaries to a new well with 400 µL of fresh culture media after the initial 24 h, using the following steps:
   1. Remove the insert containing the ovaries from the well using a pair of forceps.
   2. Rinse the bottom of the insert briefly by submerging it in a 60 mm dish with fresh culture media.
   3. Gently tap the bottom of the insert on an empty 60 mm dish to remove excess media on the insert. Place the insert into the well with fresh culture media.
      NOTE: Remove excess media from the insert to ensure the ovaries are stabilized on the membrane immediately after changing media or transferring the insert.

3. Fixation and whole-mount immunostaining

1. To fix the cultured ovaries, remove the insert from the plate using a pair of forceps and place it into a 60 mm petri dish containing 5 mL of DPBS. Transfer approximately 100 µL of DPBS next to the ovaries and resuspend them using a 1 mL pipette.
2. Using a 1 mL pipette, transfer the ovaries from the insert into a 1.5 mL Eppendorf tube. Remove DPBS from the tube and add 500 µL of 4% paraformaldehyde. Fix the ovaries at 4°C for 2 h.
   NOTE: The end of the pipette tip may need to be trimmed wider to prevent damaging the tissues if the ovaries are too large to pass through it.
3. Wash the fixed ovaries in 1 mL of PBST₂ (PBS, 0.1%Tween [v/v], 0.5% Triton [v/v]) at least three times, 1 h each on a rotating shaker. It is recommended to leave the tissues in PBST₂ overnight (12-16 h) at 4 °C to enhance antibody penetration.
   NOTE: This can be a stopping point; the tissues may be stored in PBST₂ up to three days before primary antibody incubation.
4. Incubate the ovaries with anti-dead-box helicase 4 (DDX4) in 100 µL of antibody dilution buffer (1% BSA [w/v], 10% Donkey serum [v/v], 0.1 M glycine, 0.1% Tween, and 0.5% Triton in PBS) at a dilution of 1/400 (2.3 µg/mL) at 4 °C for 12-16 h.
5. On the next day, remove the primary antibody and add 1 mL of PBST₂ to the tube. Wash the tissues at least three times, 1 h each on a rotating shaker.
   NOTE: This can be a stopping point; the tissues can remain in PBST₂ up to three days after primary antibody incubation.
6. Dilute the secondary antibody donkey anti-rabbit AF488 at 1/500 dilution (1.5 µg/mL) in PBST₂. Incubate the tissue with at least 100 µL of the diluted secondary antibody at 4 °C for 12-16 h.
7. On the next day, wash the tissues extensively in PBST₂ at least three times, 1 h each on a rotating shaker_. It is recommended to wash the tissues for up to 8 h to reduce the background. Make sure to cover the tubes with foil to protect them from light from this step forward.
8. Remove PBST₂ and add 100 µL of DAPI at a dilution of 1/1000 (1 µg/mL) in PBS to the tube. Incubate the tissues in DAPI for 30 min at room temperature.
9. Wash the ovaries in PBST₂ once for 30 min and mount them on a microscopic slide with an imaging spacer. Transfer the ovaries to the center of the spacer using a 1 mL pipette and remove excess PBST₂. Add about 5 µL of the antifade mounting medium and place a microscope slide on the top of the spacer. Seal the edges of the spacer with clear nail polish for long-term storage.
   NOTE: Before imaging the slides, ensure nail polish is completely dry and that there is no mounting medium on the outside of the cover slip. These reagents may damage the objectives of the microscope. Additionally, avoid using excess mounting media, as the ovaries need to remain stationary during imaging.
10. Image the ovaries with a confocal microscope at 10x magnification with a Z-step size of 10 µm. Adjust settings and use the Z-compensation (excitation gain) to image the entire ovary.This can be done using the following steps:
    1. Adjust the excitation power and detector gain by monitoring signal saturation using the histogram, as it may vary between tissues. Begin with low laser power and gradually increase as needed.
    2. For imaging deeper regions, enable Z-compensation for the gain detector, as signal intensity may decrease with depth. Use a frame averaging of at least 10 and enable bidirectional X-scanning at a speed of 600 Hz.
       NOTE: With these settings, imaging should not exceed 10 min per ovary (~200 µm). Whole-mount immunofluorescent staining of ovaries with anti-DDX4 is typically robust and rarely photobleaches.

4. Follicle quantification

1. To quantify follicles in each ovary, open the images using ImageJ software. Upon opening, ensure that the color mode is selected as colorized and that split channels is checked in the box under split into separate windows. Select the image of interest, then use the image/transform/rotate tool to place the ovary symmetrically in the image.
   NOTE: The images acquired by confocal can be viewed and analyzed in 3D using Imaris software (Figure 3A,B) or ImageJ software¹⁶.
2. Use the analyze/tools/grid tool to divide the ovary into grids (Figure 3C,C').
3. Use the plugins/analyze/cell counter tool to mark all the grids covering the ovary (Figure 3D,D', grids marked by 4). Of these grids, mark every other grid in each row for follicle quantification (Figure 3E,E', grids marked by 8). Grids on the edge of the ovary that have fewer than 10 oocytes are not included for follicle quantification.
   NOTE: Initialize the cell counter and choose the counter types that you consider easy to read.
4. Oocytes can be identified by the presence of a DDX4-positive ring-shaped structure surrounding a nucleus that is positive for DAPI. Primordial follicles have oocytes at a size of approximately 20 µm in diameter surrounded by squamous somatic cells. Developing follicles have oocytes that are greater than 25 µm in diameter and surrounded by one or more layers of cuboidal granulosa cells (Supplementary Figure 1).
5. Count the primordial follicles in each marked grid (as seen in Figure 3E,E' with counter Type 8). Go over the follicles by using optical sections/stacks in the grid, and mark them using the cell counter tool, respectively. Ensure not to double-count primordial follicles on two consecutive optical sections/stacks. Count developing follicles in all the grids marked with counter Type 4.
   NOTE: Use the zoom option to help visualize and count the oocytes through the stacks.
6. Calculate the number of primordial follicles in the entire ovary (nf) by multiplying the average number of primordial follicles per grid (nfgrid) by the number of grids encompassing the ovary (ngrids). nf = nf grid × ngrids.
   NOTE: Statistical analyses were performed, and graphs were generated using Prism 10.2.0 software (GraphPad Software, version 10, Inc., CA). Differences among groups were evaluated using one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. Statistical significance was considered at p < 0.05. Data is presented as the mean ± standard error (SE).

In this article, we demonstrate the methods of neonatal ovary culture, whole-mount antibody staining, confocal imaging, and follicle quantification to assess whether the number and position of the ovaries in the insert during culture affect the experimental outcome, specifically, follicle counts in the ovary. We cultured ovaries with the following setups: one, two, or three ovaries per insert for stabilized culture, and three ovaries per insert for floating culture. The ovaries were cultured under two conditions: a control group or a drug-treated group with a 24 h treatment of Dox at 0.1 µg/mL. After 24 h of culture, ovaries from both the control and Dox-treated groups were switched to fresh culture media for an additional 4 days of culture.

We found that among the ovaries cultured in control media using the stabilized culture approach, on average, the one-ovary culture group had more primordial follicles and the three-ovary group had less primordial follicles. However, there was no significant difference between the three groups. Notably, the three-ovary group cultured with the floating culture approach had the lowest number of primordial follicles among all four culture conditions. However, no significant difference was observed between three-ovary cultured in stabilized and floating approaches (Figure 4A,B,E). Additionally, ovaries cultured using the floating approach also had the fewest developing follicles. In contrast, ovaries cultured using the stabilized approach had a similar number of developing follicles regardless of the number of ovaries cultured in each insert. No significant difference in the numbers of developing follicles was observed among the four groups (Figure 4F).

For the ovaries cultured with 24 h Dox treatment, among the three groups with stabilized culture, the one-ovary group had the most primordial follicles, although no significant difference was observed among the three groups. Strikingly, three ovaries cultured in the floating approach had significantly fewer primordial follicles compared to those cultured in the stabilized approach, regardless of the number of ovaries in the insert (Figure 4C,D,G). There was no significant difference in the number of developing follicles across the four culture groups. On average, the three ovaries cultured using the floating approach had the fewest developing follicles (Figure 4H).

Figure 1: A diagram illustrating the key experimental steps for ovary culture and whole-mount antibody staining. Step numbers in green indicate possible stopping points. The diagram was created with BioRender.com. Please click here to view a larger version of this figure.

Figure 2: Step-by-Step Illustration of Neonatal Mouse Ovary Culture. (A) The anatomic location of the ovary in the postnatal day 5 mouse. (B) Two ovaries embedded in the bursa and the associated reproductive tracts. (C) Dissection of the ovary from the surrounding bursa and reproductive tracts. (D) A dissected ovary. (E) A 24-well plate. The wells in the top row each contain an insert and 400 µl of media. (E') A close-up image showing a well containing an insert. (F-G) Transfer of the ovaries from the petri dish to the insert. (H) A well with three ovaries set up for stabilized culture. (I) A well with three ovaries set up for floating culture. (J) Media change for cultured ovaries. Diagrams in H and I were created with BioRender.com. Please click here to view a larger version of this figure.

Figure 3: Whole-mount follicle quantification of the cultured ovaries. (A-B) 3D images generated by Imaris showing control and Dox-treated ovaries cultured under the condition of stabilized culture of three ovaries. (C-E) Ovarian follicle quantification using the Grids tool in ImageJ, indicating the total number of grids selected in the ovary (marked with the number 4) and the number of grids counted for primordial follicles (marked with the number 8). (C'-E') Zoomed-in images of the boxed area in C-E. Please click here to view a larger version of this figure.

Figure 4. Follicle counts in the ovaries cultured using two different setups. (A-D) Representative images of an optical section from a series of confocal images of ovaries cultured under the following conditions: (A) control stabilized culture (SC) of three ovaries, (B) control floating culture (FC) of three ovaries, (C) Dox-treated SC of three ovaries, and (D) Dox-treated FC of three ovaries. (E-F) Numbers of primordial and developing follicles per ovary after 5 days of culture under different conditions (1 ovary, 2 ovaries, or 3 ovaries in stabilized condition, and 3 ovaries in the floating condition). (G-H) Numbers of primordial and developing follicles per ovary after culture with Dox under different culture conditions. Data is presented as mean ± standard error (S.E.). *p < 0.05; ****p < 0.0001 calculated by Tukey's test. Please click here to view a larger version of this figure.

Supplementary Figure 1: Classification of primordial follicles and developing follicles in confocal images. (A) An optical section from a series of confocal images of a cultured mouse ovary stained with DDX4 for oocytes and DAPI for nuclei. (A') A close-up view of the boxed area in A. Arrowheads: primordial follicles; circle: a primary follicle. Please click here to download this File.

In this study, we demonstrated methods for neonatal mouse ovary culture and for whole-mount ovarian follicle quantification. The culture approach can sustain primordial and early developing follicles in vitro. The whole-mount follicle quantification method requires minimal tissue processing, which saves time and preserves tissue integrity, ensuring accurate follicle quantification.

The number of primordial follicles observed in the control group suggests that, while there is a slight reduction compared to the literature's reported follicles counts in postnatal day 10 ovaries (approximately 4,300 follicles per ovary), our results are still comparable¹⁷. This decrease is likely due to the less-than-optimal conditions of in vitro culture compared to in vivo conditions. In addition to primordial follicles, developing follicles, including primary, secondary, and tertiary follicles, were observed in cultured ovaries. This observation is consistent with folliculogenesis reported in CD1 ovaries at postnatal day 10, indicating that the culture approach can effectively support follicle development¹⁷.

The results of follicle quantification indicate that it is crucial to maintain the ovaries using a stabilized approach, which provides consistent physical support. Ovaries cultured using a floating approach exhibited an increased loss of primordial follicles, likely due to cell death, as no corresponding increase in the number of developing follicles was observed. The difference in primordial follicle counts between the stabilized culture approach and the floating culture approach may lead to non-biological variations in experimental results. This discrepancy is likely a result of different mechanical stresses experienced by the ovaries^14,18. Maintaining stabilization during culture is especially important when conducting drug treatments; significantly fewer primordial follicles were found in the Dox-treated ovaries cultured using the floating approach. Additionally, we noted that the number of ovaries cultured in each insert slightly affected the average primordial follicle counts when using the stabilized approach. Therefore, it is essential to maintain a consistent number of ovaries per well throughout the study to minimize non-biological variations.

The whole-mount follicle quantification method is suitable for postnatal cultured ovaries, as the number of primordial follicles obtained closely aligns with existing literature. Several approaches have been reported for identifying and quantifying primordial follicles, ranging from traditional morphology-based histological sectioning to whole-mount immunofluorescence combined with software-assisted or deep learning-based analyses^19,20,21. The immunofluorescence-based whole-mount follicle quantification protocol introduced here provides a simpler and faster alternative. This method enables imaging without the need for extensive tissue clearing and permeabilization, allowing for data acquisition in under a week using freely available software. The present study employed a DDX4 antibody to identify oocytes, a method that has been well-established and used in over 500 studies for decades, according to information on the vendor's webpage. Additionally, recent studies have used p63 and GCNA to mark oocytes^22,23. These proteins, located in the nucleus of the oocyte, may offer an alternative for computer-based automated whole-mount follicle quantification.

Whole ovary culture is a valuable in vitro approach that preserves the ovary's full architecture. This approach supports essential intercellular communication and signaling required for follicle development while enabling the isolation of therapeutic or treatment effects specific to the ovary. However, this system has limitations, including the lack of consideration of systemic factors, such as immune responses. Additionally, the use of serum in the culture media can complicate analyses, such as proteomics. Appropriate controls should always be considered based on experimental goals.

In summary, neonatal mouse ovary culture and whole-mount follicle quantification are effective approaches for studying the regulation of primordial follicle dormancy, death, and development. It is crucial to keep the culture conditions as consistent as possible to prevent non-biological variations.