Ovarian tissue cultures can be used as models of follicle development, ovulation, and follicle atresia and indicate regulatory mechanisms of dynamic ovarian processes.
Mammalian females periodically ovulate an almost constant number of oocytes during each estrus cycle. To sustain such regularity and periodicity, regulation occurs at the hypothalamic-pituitary-gonadal axis level and on developing follicles in the ovary. Despite active studies, follicle development mechanisms are not clear because of the several steps involved from the dormant primordial follicle activation to ovulation, and because of the regulation complexity that differs at each follicular stage. To investigate the mechanisms of follicle development, and the dynamics of follicles throughout the estrus cycle, we developed a mouse ovarian tissue culture model that can be used to observe follicle development using a microscope. Systematic follicle development, periodical ovulation, and follicle atresia can all be reproduced in the cultured ovary model, and the culture conditions can be experimentally modulated. Here, we demonstrate the usefulness of this method in the study of the regulatory mechanisms of follicle development and other ovarian phenomena.
Female mouse ovaries contain several thousand follicles1, and periodic ovulation matures approximately ten oocytes at each estrus cycle. Follicles are classified into several developmental stages: primordial, primary, secondary, antral, and Graafian follicles, depending on the form of the granulosa cellular layer surrounding each oocyte. Most primordial follicles are dormant, and some of them are activated and grow into primary follicles at each estrus cycle2. After the secondary follicular stage, follicle development is mainly regulated by gonadotropins, follicular stimulating hormone (FSH), and luteinizing hormone (LH). However, primordial and primary follicle development is independent of gonadotropin, and the regulatory mechanisms that govern these stages remain poorly inderstood3,4,5. In addition to growth factors and hormones, the primordial and primary follicle is regulated by the interactions among follicles6,7. Therefore, we performed analyses of follicle dynamics in mouse ovary tissues, and investigated the associated regulatory mechanisms using ovarian tissue cultures8,9,10.
Herein, we introduce two ovarian tissue culture model methods. The first is used to analyze follicle development by measurement of follicular areas, and the second is used to study the regulatory mechanism during early follicle development from primordial to secondary follicle stage with transgenic mice. For follicle development analysis, we mainly used ovaries of 4-week old female mice because they allow for easy visualization of follicles. To induce periodical ovulation and model in vivo follicle development, we reproduced LH surge and observed ovulation, follicle atresia, and secretion of estradiol under tissue culture conditions. Images of the cultured ovaries were captured, and the follicle development processes were analyzed by tracing changes in the follicular area. However, in bright field microscopy analyses, the distinction between primordial and early primary follicles was unclear. Thus, we developed a method to detect small follicles, and distinguish between the primordial, primary, and secondary follicle in cultured ovarian tissues using Oogenesin1 (Oog1) pro3.9 and R26-H2B-mCherry transgenic mice ovaries at days 0 and 4 after birth11. Oog1 expression is detectable in oocytes after entry into meiosis, and gradually increases with follicle development, allowing observation of the transition from primordial to primary follicles using time-lapse images of cultured ovary tissue11,12. Although morphological methods have been used to study factors that activate dormant primordial follicles13,14,15,16, physiological follicle development in ovaries is difficult to observe, and the effects of various factors remain uncharacterized. The present culture methods were designed to address this paucity in real time analyses of target factors.
In the present study, we tracked follicular development using a time-lapse imaging method and characterized the process of follicle development. Our novel methods offer an unprecedented tool for investigating the physiology of ovaries.
Mice were housed in an environmentally controlled room at 23 ±1 °C with a 12 h light/12 h dark cycle. Animal care protocols and experiments were conducted in accordance with the Guidelines for Animal Experimentation at Aichi Medical University and were approved by the incumbent Animal Care and Use Committee.
1. Preparation of Culture Medium and Dishes
2. Preparation of Ovarian Tissue
3. Ovarian Tissue Culture
4. Microscope Images of Cultured Ovaries
5. Time-lapse Imaging of Cultured Ovaries
6. Follicle Growth Analysis
7. Analysis of Follicle Development Using Transgenic Mice
Figure 1 shows the protocol for changes in media during ovarian tissue culture. Following this program, 4-week-old ICR mice ovaries were cultured and imaged at 24-h intervals using confocal microscopy (Figure 2). During culture of ovary tissues for 3 weeks, most antral and secondary follicles were degenerated by follicle atresia and some were ovulated (Figure 2D and Table 2). In analysis of follicle areas using ImageJ (Figure 3), some follicles developed in groups during each LH surge cycle (Figure 3B and Supplementary Movie 1). Moreover, ovulation occurred almost simultaneously in separate cultured ovaries from right and left ovaries of single mice. Whereas, the timing of ovulation and numbers of ovulated oocytes (Table 2) differed between mouse ovaries (data not shown), ovulation from cultured ovaries predominantly occurred within 48 h of LH surges (Figure 3, and Supplementary Movie 1). However, not all ovulated oocytes released first polar bodies after ovulation (Figure 2E and Table 2), and although ovulated oocytes could be fertilized, development ceased at the 4-cell stage (data not shown). To elucidate mechanisms by which dormant primordial follicles are activated in mouse ovaries, we detected primordial follicle activation in cultured tissues from transgenic mice carrying Oog1pro3.9 and R26-H2B-mCherry transgenes (Figure 4, Figure 5, and Supplementary Movie 2). Oocytes express very low levels of the Oog1 gene from the primordial follicle stage, and time-lapse imaging distinguished low AcGFP1-expressing primordial follicles from the high AcGFP1-expressing primary follicles. Primordial and primary follicles are present simultaneously in P4 mouse ovaries, whereas only primordial follicles are present in P0 ovaries (Figure 4A, B). Thus, we optimized time-lapse imaging conditions to detect only primary follicles in cultured P4 ovaries (Figure 4C). Subsequently, we captured images of cultured P0 ovaries for 10 days in the same conditions (Figure 4D–G). AcGFP1 can be used as an index of primary follicles in these transgenic mice and AcGFP1-positive primary follicles were detectable in cultured P0 ovaries after 30–40 h culture (Figure 4, Figure 5A-F, and Supplementary Movie 2). Primordial and primary follicles in cultured ovaries were also distinguished by mCherry-positive nuclei of granulosa cells (Figure 5B, E). Specifically, mCherry signals indicate forms of follicles and allow observation of follicle stages in cultured ovaries (Figure 5B, E, and H). Therefore, this culture system using ovaries from Oog1pro3.9/R26-H2B-mCherry mice revealed the process of early follicle development from primordial to secondary follicle stages. These methods will facilitate studies of the regulatory mechanisms of gonadotropin-independent follicle development.
Figure 1: Time course of medium changes. Medium A contains 5% FBS, 100-mU FSH, 10-mU LH, 100-U/mL penicillin, and 100 mg/mL streptomycin in MEM-alpha. Medium B contains 5% FBS, 100-mU FSH, 100-mU LH, 100-U/mL penicillin, and 100 mg/mL streptomycin in MEM-alpha. Medium B was used to produce LH surges every 4. Please click here to view a larger version of this figure.
Figure 2: Images of cultured ovarian tissues. Images were captured at 24 h intervals using confocal microscopy. (A) Culture day 1; (B) Culture day 5; (C) Culture day 10; (D) Culture day 13; (E) Magnified image of the region indicated by the square in (D); oocytes in D were ovulated from the follicles indicated by arrows in B, C, and D. (F) Oocyte releasing a polar body after ovulation; Scale bar = 200 µm. Please click here to view a larger version of this figure.
Figure 3: Measurements of follicle areas in cultured ovaries. (A) Image of cultured ovary; dotted lines represent the area measured by ImageJ. (B) Follicular areas in cultured ovary after 3 weeks; gray lines represent the culture period from the addition of 100-mM LH (LH surge). Lines show the development stages in each follicle in the cultured ovary; Scale bar = 200 µm. Please click here to view a larger version of this figure.
Figure 4: Hematoxylin and Eosin (H&E) staining images of P0 and P4 ovaries and time-lapse imaging. (A) H&E staining of P0 ovary; (B) H&E staining of P4 ovary; (C) Image of a P4 ovary from an Oog1pro3.6 transgenic mouse after 10 h culture. The arrow shows an AcGFP1-positive primary follicle. (D–G) Images of P0 ovaries from transgenic mice expressing Oog1pro3.9 and R26-H2B-mCherry; green and red signals in D and F represent AcGFP1 and mCherry, respectively. White signals in E and G represent AcGFP1 in D and F, respectively. D and E are images of cultured ovaries after 0.5 h culture. F and G show ovaries after 180 h culture; Scale bar = 100 µm. Please click here to view a larger version of this figure.
Figure 5: Images of primordial, primary, and secondary follicles in cultured ovaries from oog1pro3.9/R26-H2B-mCherry mice. (A–C) Images of primordial follicles in cultured P0 ovaries; (D–F) Images of primary follicles in P0 cultured ovary; (G–H) Images of secondary follicles in cultured 4-week-old mouse ovaries. A, D, and G are merged images of B and C, E and F, and H and I, respectively. Green, AcGFP1; red, mCherry; Scale bar = 100 µm. Please click here to view a larger version of this figure.
microscope | objective lens | laser | exposure (ms) | Gain | time-interval | z-stack | others | Figures and Movies |
CV1000 | X10 | 488 nm: 30, 561 nm: 20 | 488 nm: 700, 561 nm: 600 | 488 nm: 90, 561nm: 20 | 30 min | 5 mm | Figure 4C-G, Movie 1 and 2 | |
BZ-X700 | X20 (with collection collar) | 40% | auto | X4 | 30 min | 10 mm | binning 3×3 | Movie 3 |
LSM710 | X10 | 488 nm: 6%, 564 nm: 5% | camera speed: 4 | 488 nm: 850, 564 nm: 850 | 24 h | 10 mm | digital gain, 488 nm: 2, 564 nm:1 | Figure 2, 3 and 5 |
Table 1: Time-lapse imaging conditions. Time-lapse imaging condition for confocal and bright field microscopes.
Ovary no. | Number of total ovulated oocytes | Number of MII oocytes |
1 | 7 | 2 |
2 | 19 | 9 |
3 | 14 | 6 |
4 | 7 | 3 |
5 | 12 | 4 |
6 | 15 | 6 |
7 | 25 | 11 |
8 | 13 | 3 |
9 | 11 | 7 |
10 | 19 | 8 |
11 | 18 | 4 |
12 | 7 | 1 |
Table 2: Numbers of ovulated oocytes in cultured ovaries. Numbers of ovulated oocytes from twelve cultured ovaries; MII indicates the numbers of ovulated oocytes that released first polar bodies after ovulation.
Supplementary Movie 1: Time-lapse movie of a cultured ovary. Ovaries were collected from 4-week old mice and were sliced and cultured for 3 weeks. The movie spans a culture period of 0–200 h at 10 frames/s. Please click here to view this video. (Right-click to download.)
Supplementary Movie 2: Time-lapse movie of a cultured P0 ovary from a transgenic mouse. Green, AcGFP1; red, mCherry. The movie spans a culture period of 0–180 h at 10 frames/s. Please click here to view this video. (Right-click to download.)
Supplementary Movie 3: Time-lapse movie of a cultured ovary from a 4-week old mouse. Green, AcGFP1; red; mCherry; the movie spans a culture period of 9–13 culture days at 10 frames/s. The arrows in the first image indicate follicles in which atresia occurred during culture. Please click here to view this video. (Right-click to download.)
In this study, we developed two new methods for studying follicle development in mouse ovaries. The first method involves culture of sliced ovarian adult mice tissues followed by analyses of follicle development, and the second involves the use of time-lapse imaging to visualize early follicle development during the gonadotropin-independent stage. Previously, we used the present ovary tissue culture method to assess the effect of leukemia inhibitory factor and progesterone on follicle development8,9, and showed that their effects vary with concentration and follicle stage. In the present study, sliced ovarian tissues exhibited follicle dynamics, warranting further analyses of ovarian follicle development mechanism using this model.
Ovaries from mice of more than 5 weeks of age contain corpus lutea that hinders microscopic observations. Hence, 4-week-old mouse ovaries are preferred for observations of follicle development and ovulation, and subsequent late primary, secondary, and antral follicles are more easily distinguished using bright field microscopy. Our methods offer comparisons of the effects of growth factors and drugs in culture media, with discernable changes in follicle development between differentially treated ovaries (Figure 3).
Transgenic mice expressing Cre or fluorescein proteins in oocytes have been described in previous studies18,19,20. However, follicular stage classification depends on histological characteristics including granulosa cell shapes, numbers of granulosa cell layers, and whether theca cells are formed. Thus, distinctions between primordial and primary follicles may be limited from observations of oocytes alone. Herein, we used transgenic mice containing the Oog1pro3.9 and R26-H2B-mCherry transgenes11,17 and visualize early follicle development from primordial to secondary follicle stages in cultured ovaries. Oog1 expression becomes detectable in oocytes during the primordial follicle stage, and is markedly increased in primary follicles. Hence, the AcGFP1 signal intensities in oocytes are associated with follicle stage (Figure 5) and were used to observe primordial-primary follicle transitions (Figure 4, Supplementary Movie 2). Cell nuclei are stained red in R26-H2B-mCherry transgenic mice (Figure 4, Supplementary Movie 2, and Supplementary Movie 3), allowing observations of follicle stages according to numbers of granulosa cells (Figure 5). Thus, collectively the present methods allow analyses of follicle development from primordial through to ovulation using Oog1pro3.9/R26-H2B-mCherry transgenic mice. Furthermore, because apoptotic cell chromatin is condensed, mCherry signals of atretic follicles are stronger than those of normal follicles (Supplementary Movie 3).
Among methodological subtleties, correct thicknesses of sliced tissues are required for successful culturing, with increased cell death in ovary slices that are too thick, and losses of large antral follicles in ovary slices that are too thin. The follicle growth speed is slow; hence, it is easy to trace and analyze the process of each follicle via 24-h interval observation. The present confocal microscope setting, including laser intensity, interval time, and digital gain, are dependent on the microscope mode, but are critical to consider when obtaining time-lapse images. In particular, strong laser intensities and long exposure times are required to capture clear images, but can affect cultured cells. Hence, moderation of these settings to avoid phototoxicity is critical. Sliced ovarian tissues of 4-week-old mice can be cultured for approximately 4 weeks, whereas primordial and primary follicles remain present subsequently, they no longer grow. In contrast, P0 and P4 ovaries can be cultured for approximately 10 days, during which primordial and primary follicles develop into primary or secondary follicles, respectively. However, follicles in cultured P0 and P4 ovaries do not develop into antral follicles. Therefore, the excision of ovaries is required at various stages for studies of the regulatory mechanisms associated with whole ovary follicle development.
The authors have nothing to disclose.
We thank Dr. Naojiro Minami (Kyoto University) for providing the Oog1pro3.9 mice. This research was supported by the JSPS (KAKENHI # JP15H06275) and the Nitto Foundation.
Follicle stimulating hormone from human pituitary | SIGMA | F4021 | |
Lutenizing hormone from equine pituitary | SIGMA | L9773 | |
Penicillin-streptomycin solution | Wako Pure Chemical Industries | 168-23191 | |
MEM a, GlutaMax, no nucleotides | Thermo Fisher | 32561037 | |
Glass bottom dish | MatTek | P35G-0-10-C | 35mm dish, No. 0 coverslip, 10mm glass diameter |
Millicell cell culture insert | Merck Millipore | PICM0RG50 | Diameter: 315 mm, pore size: 0.4 mm, material: hydrophilic PTTE |
3.5cm cell culture dishes | greiner bio-one | 627160 | |
50ml / centrifuge tube with triple seal cap | IWAKI | 2345-050 | |
Low-profile disposable blades 819 | Leica | 14035838925 | |
LSM 710 | Carl Zeiss | Confocal microscope | |
CellVoyager, CV1000 | Yokogawa Electric Corporation | Time-lapse imaging | |
BZ-X700 | KEYENCE | Time-lapse imaging |