Summary

Accurate Follicle Enumeration in Adult Mouse Ovaries

Published: October 16, 2020
doi:

Summary

Here, we describe, compare, and contrast two different techniques for accurate follicle counting of fixed mouse ovarian tissues.

Abstract

Sexually reproducing female mammals are born with their entire lifetime supply of oocytes. Immature, quiescent oocytes are found within primordial follicles, the storage unit of the female germline. They are non-renewable, thus their number at birth and subsequent rate of loss largely dictates the female fertile lifespan. Accurate quantification of primordial follicle numbers in women and animals is essential for determining the impact of medicines and toxicants on the ovarian reserve. It is also necessary for evaluating the need for, and success of, existing and emerging fertility preservation techniques. Currently, no methods exist to accurately measure the number of primordial follicles comprising the ovarian reserve in women. Furthermore, obtaining ovarian tissue from large animals or endangered species for experimentation is often not feasible. Thus, mice have become an essential model for such studies, and the ability to evaluate primordial follicle numbers in whole mouse ovaries is a critical tool. However, reports of absolute follicle numbers in mouse ovaries in the literature are highly variable, making it difficult to compare and/or replicate data. This is due to a number of factors including strain, age, treatment groups, as well as technical differences in the methods of counting employed. In this article, we provide a step-by-step instructional guide for preparing histological sections and counting primordial follicles in mouse ovaries using two different methods: [1] stereology, which relies on the fractionator/optical dissector technique; and [2] the direct count technique. Some of the key advantages and drawbacks of each method will be highlighted so that reproducibility can be improved in the field and to enable researchers to select the most appropriate method for their studies.

Introduction

The immature, meiotically-arrested oocytes stored within primordial follicles in the ovary are the storage unit for the female germline and comprise an individual’s lifetime ovarian reserve. Primordial follicle numbers decline naturally with age1, or alternatively, can be prematurely depleted following exposure to exogenous chemicals, including some pharmaceuticals and environmental toxicants in air, food and water2. Given that the primordial follicle number is finite, the quantity and quality of follicles present within the ovary largely determine female fertility and offspring health. Thus, accurate quantification of primordial follicle number in women is essential for evaluating the off-target impacts of exogenous insults on the ovarian reserve.

In women, analysis of the whole ovary is generally not possible, thus non-invasive surrogate measures of the ovarian reserve must be utilized in a clinical setting. Anti-Mϋllerian hormone (AMH) is the most widely used surrogate biomarker clinically3. Serum AMH levels are often measured in women of advanced maternal age, or before and after cancer treatment, such as chemotherapy. However, AMH is produced by growing follicles and not by primordial follicles, and thus, serum levels do not inform on absolute primordial follicle number.

With the absence of methods to accurately determine primordial follicle number in women in situ, counting ovarian follicles in small animal models, such as rodents, remains an essential research tool to assess the degree by which exogenous insults impact on primordial follicles and thus, fertility. Unfortunately though, reports throughout the literature of primordial follicle numbers in rodent models are highly variable4. A major reason for this is widely reported technical differences in the counting method employed. Predominately, there are two different techniques described in the literature for enumerating primordial follicles in mice. These include stereology, which employs the fractionator optical dissector method, and direct follicle counts.

Stereology is widely regarded the gold standard as it uses systematic uniform random sampling5, making it the most accurate method of quantifying primordial follicle number in whole mouse, or rat ovaries4,6,7. Stereology is unbiased, as it accounts for the three-dimensional structure of the object of interest8. Using an optical dissector/fractionator method, three levels of sampling are applied to quantify primordial follicles using thick tissue sections (e.g., 20 µm) within a known fraction of the total mouse ovary. Firstly, the sampling interval is chosen (e.g., every 3rd section) at a random start (sampling fraction 1, f1)4. Sections are then sampled in a systematic, uniform manner from this point through the whole ovary. Then, an unbiased counting frame is superimposed over the ovarian section and progressively moved along a defined, randomized counting grid (sampling fraction 2, f2)8. Lastly, a known fraction of the section thickness is optically sampled (e.g., 10 µm) and follicles within this area are counted (sampling fraction 3, f3)4. The raw follicle number is multiplied by the inverse of these sampling fractions to obtain the final value. This method requires expert training and equipment, including a microscope with a motorized stage driven by stereological software. Tissues should be preserved in a specialized Bouin’s fixative, and embedded in glycolmethacrylate resin to allow for thick tissue sections to be cut using a glycolmethacrylate resin microtome with a glass knife. This method is designed to account for tissue shrinkage and deformation, to best preserve the three-dimensional morphological structure of the ovary and follicles9.

Direct follicle counting is the most frequently used method for counting follicles10. More common fixatives (i.e., formalin) can be used, followed by paraffin wax embedding and exhaustive serial sectioning using a standard microtome at a thickness of between 4-6 µm. Follicles are systematically counted in the entire tissue section at a defined interval, and then multiplied by the inverse of the sampling interval to obtain the total follicle estimate. This method is quick, easy, can be performed using archived tissues, and prepared using standard histological techniques. It requires only a light microscope with standard imaging capabilities. However, despite these advantages, direct follicle counting lacks the accuracy and strict counting parameters of stereology, making it more prone to investigator bias. Additionally, tissues may undergo shrinkage and deformation during processing, disrupting the integrity and morphology of the ovary and thus making follicle classification and quantification difficult.

The aim of this article is to describe two commonly-used methods to quantitatively assess primordial follicle number in mouse ovaries: stereology and direct follicle counting. We will provide detailed protocols for these two methods and highlight some of their strengths and weaknesses, in order to enhance reproducibility in our field and allow researchers to make an informed decision of the most appropriate counting method for their studies.

Protocol

Ovaries were collected from female C57BL6J mice. All animal procedures and experiments were performed in accordance with the NHMRC Australian Code of Practice for the Care and Use of Animals and approved by the Monash Animal Research Platform Animal Ethics Committee.

NOTE: A chemotherapy agent shown to deplete primordial follicle oocytes, as determined using stereology11 and direct counts12,13 was used in this report to compare the two counting methods in the same animal. Female, 8-week-old (young adult) mice were weighed prior to a single intraperitoneal injection of 75 mg/kg/bodyweight of cyclophosphamide, or saline vehicle control (n=5/group). This dose has been shown to cause an approximate 50% depletion of primordial follicles, but not reported to cause morbidity or mortality in mice14. Ovaries were harvested 48 hours after treatment. One ovary from each animal was fixed in 10% (v/v) neutral buffered formalin solution for 24 hours, and the other fixed in Bouin’s solution for 24 hours. Tissue was then embedded in either glycolmethacrylate resin and serially sectioned at 20 µm, or in paraffin and serially sectioned at 5 µm. All tissues were stained with periodic acid Schiff and haematoxylin.

1. Histological preparation: fixation, processing, embedding and sectioning mouse ovaries

  1. Dissect mouse ovaries by trimming the oviduct and all surrounding fat, without damaging or cutting the ovary itself. If necessary, use a dissecting microscope and fine blade for this step (Figure 1A).
  2. Fix tissues immediately by placing into a small labelled tube containing either Bouin’s fixative (stereology), or formalin fixative (direct counts) for 24 hours (Figure 1B), before transferring tissues into 70% ethanol.
    NOTE: Follicle morphology is conserved best within Bouin’s fixed ovarian tissue, embedded into glycolmethacrylate resin (Figure 2).
  3. Process whole fixed ovaries and then embed in glycolmethacrylate resin for stereology (Supplementary File 1), or paraffin wax for direct counts using a standard histological protocol.
    CAUTION: Resin is toxic, so ensure all tissue processing steps are performed in a fume hood and gloves are worn at all times.
  4. Use a specialized resin methacrylate microtome (Figure 1C) fitted with a glass knife (Figure 1D) to exhaustively cut thick glycolmethacrylate resin sections (e.g., 20 µm). Collect the sections at a regular interval (e.g., every 3rd section) onto glass microscope slides for stereology.
  5. Use a standard microtome to cut thin paraffin sections (e.g., 4-6 µm). Collect tissue sections at a regular interval (e.g., every 9th section) onto a glass microscope slide for direct follicle counts.
  6. Stain the slides with periodic acid Schiff and haematoxylin (Supplementary File 2).
  7. Coverslip with standard DPX for paraffin sections, or thick DPX for glycolmethacrylate resin sections (Figure 1E).
    CAUTION: Glycolmethacrylate resin DPX is hazardous, so perform this step in to fume hood.
    NOTE: Glycolmethacrylate resin DPX is extremely viscous. The glass coverslip must be adhered firmly by pressing it down with a spatula to ensure the DPX is evenly and thinly dispersed, and there are no air bubbles present under the coverslip (Figure 1F).

2. Stereological estimation of primordial follicle number using the optical fractionator

  1. Turn on the computer, the multi-control unit, the camera and the light source within the stereology setup and set the microscope objective to a low magnification (e.g., 10x).
  2. Open the stereology software.
  3. Put the first slide securely on the microscope stage.
  4. Adjust the light exposure by checking Automatic under Exposure in the Camera Settings panel (Supplementary Figure 1A). Alternatively, manually adjust the light exposure.
  5. Use the joystick to locate the first tissue sample and bring the sample into focus.
  6. Adjust the white balance, either by clicking on More Settings (located bottom right of the Camera Settings panel), and then click White Balance and click on Automatic (Supplementary Figure 1B). Alternatively, click the White Balance button adjacent to the More Settings button (or Select Area de More Settings), to set the white balance manually by selecting a white area on the section.
  7. Go to the Probes drop down menu and click on Optical Fractionator Workflow. Then click Start New Project and click OK.
    1. If an existing sampling configuration has been saved, under Sampling Parameters click Yes | and select the desired sampling configuration.
    2. If not, click No and manually enter the serial section information (Supplementary Figure 1C) and define the probe configuration at step 2.13.
  8. Click on Next, set the microscope to Low Magnification and choose 10x magnification from the dropdown menu.
  9. Click on Next, and then trace around the entire ovarian section – start by left clicking around the section and at the end, right click and choose Close Contour.
  10. Click on Next, set the microscope to High Magnification and choose 100x Oil magnification from the dropdown menu.
  11. Place a drop of oil on the tissue section on the slide and move the microscope objective to 100x magnification.
  12. Adjust the light exposure (as in step 2.4) and click Next.
  13. Set up the Sampling Parameters to define the probe configuration. Here, the counting frame was set to 47.5 µm x 47.5 µm (2,256.25 µm2) and the step length was set to 100 µm x 100 µm (10,000 µm2) (Supplementary Figure 2). Once the sampling parameters are established, save the template and re-open during subsequent analysis sessions at step 2.7.
  14. Click on Start Counting (Supplementary Figure 1D). Focus to the top of the sample, click OK and begin counting.
  15. Use the focusing knob to move through the 10 µm sampling depth and count any primordial follicles whose oocyte nucleus comes into focus. Click Next to move to the next area.
    1. Classify follicles as a primordial if the oocyte is surrounded by squamous (flattened) granulosa cells, but no cuboidal granulosa cells (Figure 2A).
      NOTE: Primordial follicles are distinct from intermediate/transitional follicles, which comprise a combination of cuboidal and squamous granulosa cells (Figure 2B), and primary follicles, which are surrounded predominantly by cuboidal granulosa cells (Figure 2C). These follicle classes should be quantified separately.
    2. Only count follicles in which the oocyte nucleus is visible. The oocyte nucleus must appear within the counting frame or be touching the green inclusion lines of the counting frame (Supplementary Figure 1E,F).
    3. If the oocyte nucleus falls outside the counting frame (Supplementary Figure 1G) or touches the red exclusion lines of the counting frame, do not count this follicle.
    4. When assessing primordial follicle depletion in response to an exogenous chemical (e.g., chemotherapy), ensure all follicles counted are healthy and thus have normal morphology (Figure 2). Count any abnormal or atretic follicles separately. Often, follicle death is induced by insults such as chemotherapy, and quantification of the atretic follicles should be obtained separately in order to distinguish between healthy and atretic follicles, as only healthy follicles comprise the ovarian reserve15.
  16. Once counting is complete on that section, do one of the following:
    1. Click Add New Section, and then return to step 2.3 to set up the next section for counting.
    2. Click I’ve Finished Counting to end the session. Return the objective to 10x, exit the stereology software and turn off the light source, camera, multi-control unit and computer.
  17. Obtain the sum raw follicle number (Q) from each tissue section sampled from the entire ovary, then using a spreadsheet, use the equation below to obtain the final value from each replicate animal (N)4.
    Equation 1, where:
    N = Total estimated number of follicles within the ovary.
    Q = Raw primordial follicle count.
    f1 = Sampling interval. Every 3rd section was sampled thus Equation 2.
    f2 = Relationship between the counting frame (sample area) and stepper, calculated as Equation 3. Since the sample area was 2256 µm2 (47.5 µm x 47.5 µm) and the stepper area was 10000 µm2 (100 µm x 100 µm), Equation 4.
    f3 = Fraction of ovarian section sampled. Since 10 µm of the 20 µm section was sampled, Equation 5.
    Therefore, Equation 6.
    NOTE: This protocol describes how to apply these principles of stereological analyses using a widely cited stereology software (Table of Materials); however, other stereological software is available. The principles applied during stereological analyses of ovarian follicles are the same, regardless of the software used to set up the parameters. Stereology is most accurate when 100 or more objects are counted in an adult mouse ovary4, as this gives a coefficient of error for the estimate below 10%16. The sampling parameters outlined here have been optimized to ensure a minimum of approximately 100 objects (i.e., primordial, transitional and primary follicles) can be counted in control adult wild-type C57BL6J ovaries. A pilot study can be conducted including a small sample size to establish the optimal sampling parameters, such as the interval and number of sections to be analyzed and the number of optical dissectors within the sampled sections17.

3. Estimation of primordial follicle number by direct ovarian follicle counts

  1. Place the microscope slide securely under a standard light microscope and perform direct counts to obtain raw primordial follicle number.
    1. Classify follicles as a primordial if the oocyte is clearly visible and is surrounded by squamous (flattened) granulosa cells, but no cuboidal granulosa cells (Figure 2D).
      NOTE: Primordial follicles are distinct from intermediate/transitional follicles, which comprise a combination of cuboidal and squamous granulosa cells (Figure 2E), and primary follicles, which are surrounded predominantly by cuboidal granulosa cells (Figure 2F). These follicle classes should be quantified separately.
    2. Ensure all follicles counted are healthy and thus have normal morphology (Figure 2). Count any abnormal or atretic follicles separately, as only healthy follicles comprise the ovarian reserve.
  2. Alternatively, take multiple, or stitched high-power (e.g., 20x) photomicrographs of the entire ovarian tissue section to perform counts by opening the image file(s). This can be done manually, or using an automated slide scanner.
  3. Obtain the sum raw follicle number (Q) from each tissue section sampled from the entire ovary at the predetermined interval. Multiply this number by the inverse of the sampling fraction to obtain the final value for each replicate animal (N), using the equation below.
    Equation 7, where:
    N = Total estimated number of follicles within the ovary.
    Q = Raw follicle count (of each individual type, calculated separately).
    f1 = Sampling interval. Every 9th section was sampled thus Equation 8.
    Therefore, Equation 9.

Representative Results

A well-characterized model of follicle depletion was used, whereby young adult female mice were administered a single dose of cyclophosphamide chemotherapy, or saline vehicle control (n=5/group) and both ovaries were harvested from each animal after 48 hours. One ovary per animal was prepared as described in Step 1 for each of the two methods: stereology or direct counts. The left and right ovary from each animal was randomly assigned to each group. These data show that when using stereology, a significant depletion of mouse primordial follicles can be detected following chemotherapy treatment (387 ± 11 follicles), versus control (1043 ± 149; p=0.0024) (Figure 3). In contrast, using the contralateral ovaries from the same animals, direct follicle counts failed to detect a significant reduction of the ovarian reserve after chemotherapy (490 ± 40), when compared to control (752 ± 139; p=0.1063) (Figure 3). Of note, it is clear that primordial follicle number in young adult wild type animals is variable, since the distribution of the saline treated animals is wider, compared to the cyclophosphamide groups, even when counts were performed using stereology (Figure 3).

Figure 1
Figure 1: Histological preparation of Bouin’s-fixed, glycolmethacrylate resin-embedded ovaries for stereological analysis. A) An adult mouse ovary (circle, arrow) was closely trimmed of all fat and the oviduct from the mouse ovary using a Feather blade. B) Contralateral mouse ovaries (circle, arrow) from the same female adult were dissected, trimmed, then either fixed in Bouin’s fixative (left) for stereology, or formalin for direct counts (right). C) A specialized resin methacrylate microtome D) fitted with a glass knife (arrow) was used to exhaustively cut glycolmethacrylate resin blocks into 20 µm sections, collected onto glass microscope slides. E) Periodic acid Schiff-stained ovarian tissue sections on glass microscope slides (arrows) were dipped into fresh histolene (green container) in a fume hood, then a glass coverslip was added on top of drops of GMA DPX. A spatula was used to remove excess DPX and air bubbles. Slides were air dried in the fume hood overnight. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Primordial follicle classification. Representative images of primordial, transitional and primary follicles in glycolmethacrylate resin-embedded (A-C) and paraffin-embedded (D-F) ovarian sections. A: Primordial follicle surrounded by squamous granulosa cells (arrow). Bar = 10 µm. B: Transitional follicle surrounded by both squamous (arrow) and cuboidal (arrowhead) granulosa cells. Bar = 10 µm. C: Primary follicle surrounded by cuboidal (arrowhead) granulosa cells. Bar = 25 µm. D: Primordial follicle surrounded by squamous granulosa cells (arrow). Bar = 10 µm. E: Transitional follicle surrounded by both squamous (arrow) and cuboidal (arrowhead) granulosa cells. Bar = 10 µm. F: Primary follicle surrounded by cuboidal (arrowhead) granulosa cells. Bar = 25 µm. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Comparison of the optical fractionator and direct counting methods to evaluate mouse primordial follicles. A well-established model of follicle depletion was used as a model, in which 8-week old female C57BL6J mice were treated intraperitoneally with saline, or a 75 mg/kg/bodyweight dose of the chemotherapeutic, cyclophosphamide (n=6/group). This enabled comparison of the two follicle counting methods to detect changes to the ovarian reserve. In the same cohort of animals, total follicle estimates failed to detect significant follicle depletion, in contrast to stereology which detected a larger and significant depletion of the ovarian reserve. Please click here to view a larger version of this figure.

STRENGTHS WEAKNESSES
STEREOLOGY Most accurate way to estimate follicles Time consuming and costly
Uses several sampling parameters, making it statistically robust Requires expert equipment and software
Highly sensitive, can detect smaller changes in primordial follicle numbers Samples must be prepared using specific, specialized histological techniques
Follicle structure is better preserved during tissue processing
Uniform rules that can be applied by different investigators, thus reducing bias
DIRECT COUNTS Time and cost effective Less accurate and sensitive than stereology
Requires standard laboratory equipment, e.g. light microscope Only one sampling parameter used, making it less effective at detecting small changes in follicle numbers
Retention of ovarian sections that can be used for immunohistochemical or other analyses Tissue is more susceptible to changes in volume and follicle structure during processing
No uniform set of rules that can be applied, thus more prone to bias from the investigator

Table 1: Comparing the strengths and weaknesses of follicle counting methods: stereology vs. direct counts.

Supplementary Figure 1: Screenshots of the stereology software protocol. A) Adjust the exposure by ticking Automatic (red box) under Exposure in the Camera Settings panel (white box). B) Set the white balance by first clicking the More Settings Icon under Other in the Camera Settings panel. Then, select White Balance and click Automatic (red box). C) If an existing sampling configuration has not been set, click No (red box) and enter the serial section information (green box). D) To begin counting, click click Start Counting (white arrow). E) If the nucleus of a primordial follicle is visible within the counting frame, this follicle is counted (white dotted circle). F) If the nucleus of a primordial follicle is touching the green inclusion lines of the counting frame, this follicle is also counted (white dotted circle). G) If the nucleus of a primordial follicle is not visible within the counting frame or is touching the red exclusion lines of the counting frame, this follicle is not counted (white dotted circle). Please click here to download this figure.

Supplementary Figure 2: Setting up the Sampling Parameters. Follow sequential steps in left-hand panel to define the Probe Configuration. A) Measure mounted thickness. Ensure ‘Refocus to top of section at each grid site’ is unticked (red box). Tick ‘Manually enter the average mounted thickness’ (green box). Enter average mounted thickness as 20.00 µm (blue box). B) Define the counting frame size. Under Counting Frame Display, tick ‘Force the counting frame display to be square’ and ‘Centre on live image’ (red box). Under Counting Frame Size, enter 47.5 µm for X and Y (green box). C) Define SRS grid layout. Manually enter the grid size as 110 for X and Y (red box). D) Define dissector options. For top guard zone height, enter 1.00 µm (red box). For optical dissector height, enter 10 µm (green box). Under Focus Method, select ‘Manual focus’ (blue box). E) Save sampling parameters. To save these settings, enter a Name and Comment then click ‘Save your Current Settings’ (white arrow). F) Under Current Sampling Parameter Settings, ensure your settings match those displayed. Please click here to download this figure.

Supplementary File 1. Please click here to download this file.

Supplementary File 2. Please click here to download this file.

Discussion

This article provides a step-by-step protocol for the gold standard technique for enumerating mouse primordial follicles, stereology, and the more commonly employed method of direct follicle counting. Chemotherapy treatment was used to compare and contrast the results obtained from these two different methods within the left and right ovary from the same animal. Both methods revealed high inter-animal variability in primordial follicle numbers. A significant depletion of the ovarian reserve was recorded using stereology, but direct counting failed to detect a significant reduction in primordial follicle number after chemotherapy administration versus control.

Notably, it is clear that even in an inbred strain of mice, such as C57BL6J, the ovarian reserve in young adult wild type (control) females is widely distributed, as is the case in humans. Factors that influence this and contribute to the establishment of the ovarian reserve remain under investigation18. High variability amongst studies of mouse ovarian function poses numerous challenges for the field to compare, or replicate data and to advance clinical translation of novel therapeutics to protect oocyte number and quality4. This variability is likely due to a number of factors including not only the counting methods employed, but additional factors such as differences between animal strain and age; as well as treatment regimens, such as dose and timing. These factors must all be considered when comparing data from different studies. Regardless, the follicle depletion model employed in this study overall highlights the accurate and sensitive nature of the stereology protocol, whilst also demonstrating that the widely reported direct counts method is unable to detect a known significant primordial follicle depletion.

There are strengths and limitations of each counting method detailed in this article (Table 1). Stereology is regarded as the gold standard method due to its accuracy when used to determine cell number in a variety of different organs5. However, drawbacks of this technique include the time and cost involved, as well as the requirement for specialized equipment and expertise. Together, this has limited the wide-spread use of this procedure for obtaining the most sensitive data.

In contrast, the direct counts method is quick, easy, can be done using archived tissues, and prepared using standard histological techniques. It requires only a light microscope with standard imaging capabilities. However, it does not account for volume changes induced by histological processing, which may disrupt the three-dimensional structure of the ovary. Consequently, follicle morphology is not always adequately preserved, making identification and counting accuracy challenging, especially for inexperienced investigators. While formalin was the fixative used for direct counts in this study, Bouin’s-fixed tissues can be embedded into paraffin and this may provide better morphology and help researchers to more easily identify primordial follicles. Furthermore, within the literature, sampling fractions for the direct counts method vary widely, ranging from every 3rd to every 10th 5 µm paraffin section19, which may contribute to large discrepancies in follicle counts between studies. Given that paraffin wax sections are so thin, counting every 9th section helps to prevent oversampling and needless counting.

Additional methods not outlined in this guide, but commonly reported in the literature, are counting in only one central section from the entire ovary and considering this to be representative of the entire ovary, or follicle density. This first method is highly problematic, since primordial follicles are not evenly distributed in the ovary and can be found in clusters. This means a single, or even a few sections of the ovary are not representative of the entire organ, making systematic sampling an essential component to any counting method. Follicle density involves counting follicles in an ovarian tissue sample, then expressing the counts per area of tissue. Given the limitations of obtaining primary human tissue, follicle density is often used as a surrogate measure for absolute follicle number in human ovary, though reports in mice are also common. However, this quantification method does not account for the uneven distribution of follicles throughout the ovary, or fluctuations in ovarian volume, which occur routinely throughout the ovarian endocrine (luteal) cycle. This is important because an accurate measure of density amongst samples relies on conserved tissue area. Furthermore, this method often only samples a small biopsy or fraction of the ovary when analyzing human tissues, thus is not representative of the entire ovary. Follicle density lacks the accuracy achievable with stereology. Even using the same whole ovary, comparative measures of follicle quantification following stereological analysis, with follicle density, did not correspond in mouse ovaries20.

Although it is the most commonly used experimental animal, the mouse is only one model species. Obtaining ovaries from larger species, including domestic animals or non-human primates, is possible, thus follicle counts from whole ovarian specimens can be accomplished using a modified stereological protocol21,22. Finally, while the ability to analyze follicle counts in whole human ovaries is indeed very difficult, it nonetheless can be done when specimens are available, using a modified stereological protocol23,24,25.

Future directions in the field may include the expansion and uptake of new techniques. For example, a new methodology of automated detection and counting for primordial follicle oocytes has been described26. This method uses convolutional neural networks driven by labelled datasets and a sliding window algorithm to select test data. However, this algorithm was only tested over two ovaries and widespread uptake remains to be determined. Alternatively, recent advances in optical tissue clearing and light sheet microscopy have permitted comprehensive analysis of intact tissues and some studies have investigated this for follicle enumeration in the mouse ovary27,28.

In conclusion, whilst direct follicle counting is an easy, quick and cost-effective method for enumerating primordial follicles within the ovary, this technique may lack sensitivity, accuracy and be prone to investigator bias. Therefore, despite its drawbacks, stereology remains the best available technique for accurately and reproducibly defining primordial follicle numbers.

Divulgations

The authors have nothing to disclose.

Acknowledgements

This work was made possible through Victorian State Government Operational Infrastructure Support and Australian Government NHMRC IRIISS and supported by funding from National Health and Medical Research Council (ALW #1120300) and Australian Research Council (KJH #FT190100265). The authors would like to acknowledge the technical support of the Monash Animal Research Platform, Monash Histology Platform and Monash Micro Imaging facility.

Materials

1-Butanol (HPLC) Fisher Chemical #A383-1
Acid alcohol Amber Scientific #ACDL
Bouin’s fixative Sigma-Aldrich #HT10132 Picric acid 0.9% (w/v), formaldehyde 9% (v/v), acetic acid 5% (w/v)
Cyclophosphamide Sigma-Aldrich #C0768-5G
Dibutylphthalate Polystyrene Xylene (DPX) Sigma-Aldrich #06522
Ethanol Amber Scientific #ETH Ethanol 100%
Micro Feather opthalmic scalpel with aluminium handle Designs for Vision #FEA-745-SR Feather blade for dissections (seen in Figure 1)
Formalin fixative Australian Biostain #ANBFC
Glass coverslip Thermo Scientific #MENCS22501GP 22 mm x 50 mm
Glycomethacrylate resin RM2165 microtome Leica Microsystems #RM2165
Glycolmethacrylate DPX *made in house *Mix 1.5 L Xylene; 800 g polystyrene pellets; 100mL Dibutyl phthalate for 3 weeks
Histolene Trajan #11031
Mayer’s haematoxylin Amber Scientific #MH
Olympus BX50 microscope Olympus #BX50 Brightfield microscope fitted with 10x dry & 100x oil immersion objective (numerical aperture 1.3)
Olympus immersion oil type-F Olympus #IMMOIL-F30CC
Olympus TH4-200 light source Olympus #TH4-200
Paraffin wax Sigma-Aldrich #03987
Periodic acid Trajan #PERI1% Periodic acid 1%
Rotary Microtome CUT 4060 MicroTec #4060R/F Used to cut paraffin sections
Schiff’s reagent Trajan #SCHF
Scott's tap water Amber Scientific #SCOT Potassium carbonate, magnesium sulphate, water
StereoInvestigator Stereological System MBF Bioscience Includes StereoInvestigator software, multi-control unit, automatic stage and joystick
Superfrost microscope slides Thermo Scientific #MENSF41296SP 1 mm, 72 pcs
Technovit 7100 Plastic embedding system Emgrid Australia #64709003 500 mL/5 x 1 g/40 mL
Technovit 3040 yellow Emgrid Australia #64708805 100 g/80 mL

References

  1. Stringer, J. M., Winship, A., Liew, S. H., Hutt, K. The capacity of oocytes for DNA repair. Cellular and Molecular Life Sciences. 75 (15), 2777-2792 (2018).
  2. Winship, A. L., Stringer, J. M., Liew, S. H., Hutt, K. J. The importance of DNA repair for maintaining oocyte quality in response to anti-cancer treatments, environmental toxins and maternal ageing. Human Reproduction Update. 24 (2), 119-134 (2018).
  3. Broer, S. L., Broekmans, F. J., Laven, J. S., Fauser, B. C. Anti-Mullerian hormone: ovarian reserve testing and its potential clinical implications. Human Reproduction Update. 20 (5), 688-701 (2014).
  4. Myers, M., Britt, K. L., Wreford, N. G., Ebling, F. J., Kerr, J. B. Methods for quantifying follicular numbers within the mouse ovary. Reproduction. 127 (5), 569-580 (2004).
  5. Boyce, R. W., Dorph-Petersen, K. A., Lyck, L., Gundersen, H. J. Design-based stereology: introduction to basic concepts and practical approaches for estimation of cell number. Toxicologic Pathology. 38 (7), 1011-1025 (2010).
  6. Ristic, N., et al. Maternal dexamethasone treatment reduces ovarian follicle number in neonatal rat offspring. Journal of Microscopy. 232 (3), 549-557 (2008).
  7. Soleimani Mehranjani, M., Mansoori, T. Stereological study on the effect of vitamin C in preventing the adverse effects of bisphenol A on rat ovary. International Journal of Reproductive Biomedicine (Yazd). 14 (6), 403-410 (2016).
  8. Brown, D. L. Bias in image analysis and its solution: unbiased stereology. Journal of Toxicologic Pathology. 30 (3), 183-191 (2017).
  9. Hasselholt, S., Lykkesfeldt, J., Overgaard Larsen, J. Thick methacrylate sections devoid of lost caps simplify stereological quantifications based on the optical fractionator design. Anatomical Record (Hoboken). 298 (12), 2141-2150 (2015).
  10. Tilly, J. L. Ovarian follicle counts–not as simple as 1, 2, 3. Reproductive Biology and Endocrinology. 1, 11 (2003).
  11. Nguyen, Q. N., et al. Loss of PUMA protects the ovarian reserve during DNA-damaging chemotherapy and preserves fertility. Cell Death and Disease. 9 (6), 618 (2018).
  12. Meirow, D., Assad, G., Dor, J., Rabinovici, J. The GnRH antagonist cetrorelix reduces cyclophosphamide-induced ovarian follicular destruction in mice. Human Reproduction. 19 (6), 1294-1299 (2004).
  13. Winship, A. L., et al. The PARP inhibitor, olaparib, depletes the ovarian reserve in mice: implications for fertility preservation. Human Reproduction. , (2020).
  14. Meirow, D., Lewis, H., Nugent, D., Epstein, M. Subclinical depletion of primordial follicular reserve in mice treated with cyclophosphamide: clinical importance and proposed accurate investigative tool. Human Reproduction. 14 (7), 1903-1907 (1999).
  15. Nguyen, Q. N., et al. Cisplatin- and cyclophosphamide-induced primordial follicle depletion is caused by direct damage to oocytes. Molecular Human Reproduction. , (2019).
  16. Gundersen, H. J., Jensen, E. B. The efficiency of systematic sampling in stereology and its prediction. Journal of Microscopy. 147, 229-263 (1987).
  17. Olesen, M. V., Needham, E. K., Pakkenberg, B. The Optical Fractionator Technique to Estimate Cell Numbers in a Rat Model of Electroconvulsive Therapy. Journal of Visualized Experiments. (125), e55737 (2017).
  18. Findlay, J. K., Hutt, K. J., Hickey, M., Anderson, R. A. How Is the Number of Primordial Follicles in the Ovarian Reserve Established. Biology of Reproduction. 93 (5), 111 (2015).
  19. Omari, S., et al. Mcl-1 is a key regulator of the ovarian reserve. Cell Death and Disease. 6, 1755 (2015).
  20. Winship, A. L., Bakai, M., Sarma, U., Liew, S. H., Hutt, K. J. Dacarbazine depletes the ovarian reserve in mice and depletion is enhanced with age. Scientific Reports. 8 (1), 6516 (2018).
  21. Miller, P. B., Charleston, J. S., Battaglia, D. E., Klein, N. A., Soules, M. R. An accurate, simple method for unbiased determination of primordial follicle number in the primate ovary. Biology of Reproduction. 56 (4), 909-915 (1997).
  22. Miller, P. B., Charleston, J. S., Battaglia, D. E., Klein, N. A., Soules, M. R. Morphometric analysis of primordial follicle number in pigtailed monkey ovaries: symmetry and relationship with age. Biology of Reproduction. 61 (2), 553-556 (1999).
  23. Charleston, J. S., et al. Estimating human ovarian non-growing follicle number: the application of modern stereology techniques to an old problem. Human Reproduction. 22 (8), 2103-2110 (2007).
  24. Hansen, K. R., Craig, L. B., Zavy, M. T., Klein, N. A., Soules, M. R. Ovarian primordial and nongrowing follicle counts according to the Stages of Reproductive Aging Workshop (STRAW) staging system. Menopause. 19 (2), 164-171 (2012).
  25. Hansen, K. R., et al. A new model of reproductive aging: the decline in ovarian non-growing follicle number from birth to menopause. Human Reproduction. 23 (3), 699-708 (2008).
  26. Sonigo, C., et al. High-throughput ovarian follicle counting by an innovative deep learning approach. Scientific Reports. 8 (1), 13499 (2018).
  27. McKey, J., Cameron, L. A., Lewis, D., Batchvarov, I. S., Capel, B. Combined iDISCO and CUBIC tissue clearing and lightsheet microscopy for in toto analysis of the adult mouse ovarydagger. Biology of Reproduction. 102 (5), 1080-1089 (2020).
  28. Kagami, K., Shinmyo, Y., Ono, M., Kawasaki, H., Fujiwara, H. Three-dimensional evaluation of murine ovarian follicles using a modified CUBIC tissue clearing method. Reproductive Biology and Endocrinology. 16 (1), 72 (2018).

Play Video

Citer Cet Article
Winship, A. L., Sarma, U. C., Alesi, L. R., Hutt, K. J. Accurate Follicle Enumeration in Adult Mouse Ovaries. J. Vis. Exp. (164), e61782, doi:10.3791/61782 (2020).

View Video