Get 20M+ Full-Text Papers For Less Than $1.50/day. Subscribe now for You or Your Team.

Learn More →

Two classes of ovarian primordial follicles exhibit distinct developmental dynamics and physiological functions

Two classes of ovarian primordial follicles exhibit distinct developmental dynamics and... Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 920–928 doi:10.1093/hmg/ddt486 Advance Access published on October 1, 2013 Two classes of ovarian primordial follicles exhibit distinct developmental dynamics and physiological functions { { Wenjing Zheng , Hua Zhang , Nagaraju Gorre, Sanjiv Risal, Yan Shen and Kui Liu Department of Chemistry and Molecular Biology, University of Gothenburg, Gothenburg SE-405 30, Sweden Received August 26, 2013; Revised September 24, 2013; Accepted September 25, 2013 Inthemammalianovary,progressiveactivationofprimordialfolliclesservesasthesourceoffertilizableova,and disorders in the development of primordial follicles lead to various ovarian diseases. However, very little is known about the developmental dynamics of primordial follicles under physiological conditions, and the fates of distinct populations of primordial follicles also remain unclear. In this study, by generating the Foxl2- T2 T2 CreER and Sohlh1-CreER inducible mouse models, we have specifically labeled and traced the in vivo devel- opment of two classes of primordial follicles, the first wave of simultaneously activated follicles after birth and the primordial follicles that are gradually activated in adulthood. Our results show that the first wave of follicles exists in the ovaries for ∼3 months and contributes to the onset of puberty and to early fertility. The primordial follicles at theovarian cortex gradually replace the first wave of follicles and dominate the ovary after 3 months of age, providing fertility until the end of reproductive life. Moreover, by tracing the time periods needed for prim- ordial follicles to reach various advanced stages in vivo, we were able to determine the exact developmental dy- namics of the two classes of primordial follicles. We have now revealed the lifelong developmental dynamics of ovarian primordial follicles under physiological conditions and have clearly shown that two classes of primor- dial follicles follow distinct, age-dependent developmental paths and play different roles in the mammalian reproductive lifespan. these primordial follicles in the medulla are synchronously acti- INTRODUCTION vated and become the first wave of activated follicles after birth, In mammals, the pool of primordial follicles serves as the source and they have long been considered not to contribute to fertility of developing follicles and fertilizable ova for the entire repro- (5,7). The supporting cells recruited in postnatal mouse ovaries ductive lifespan of the organism (1,2). Whereas the majority of form primordial follicles in the cortical region (5,6). The cortical primordial follicles remain in a dormant state, a limited primordial follicles are believed to be activated gradually as a number of primordial follicles are recruited from the resting fol- means of providing mature ova over the entire course of the licle reservoir into the growing follicle pool. In recent years, the animal’s reproductive life (8,9). molecular mechanisms controlling the activation of primordial In the current study, we have generated two tamoxifen- follicles have begun to be revealed mostly by using genetically inducible knock-in mouse models, the Forkhead box L2 (Foxl2)- T2 modified mouse models, which represent extreme situations in CreER mice and the Spermatogenesis and oogenesis-specific T2 which a particular gene is deleted (for reviews see references 3 basic helix-loop-helix 1 (Sohlh1)-CreER mice. These mouse and 4). Very little is known, however, about the development models have allowed us to specifically label the first wave of acti- of primordial follicles under physiological conditions through- vated follicles in the ovarian medulla and the primordial follicles out the entire female reproductive lifespan. in the cortex, respectively. We have traced the in vivo develop- Previous studies have shown that the supporting cells ment of the two classes of follicles over the lifetimes of the recruited in fetal mouse ovaries form the initial primordial folli- mice, and our results show that the first wave of medullary folli- cles in the medullary region of the ovary (5,6). Once formed, cles and the cortical primordial follicles are distinct populations To whom correspondence should be addressed. Tel: +46 317866585; Fax: +46 317863801; Email: kui.liu@gu.se The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. # The Author 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 921 in terms of their developmental dynamics, their contributions to the onset of puberty and their contributions to fertility during young adulthood and later in reproductive life. RESULTS T2 Generation of a Foxl2-CreER knock-in mouse model for labeling the first wave of activated follicles in postnatal ovaries Foxl2 is a forkhead transcription factor that is expressed in somatic cells of the early XX gonad and in pregranulosa and granulosa cells (10,11). To selectively label the first wave of acti- vated follicles that appear in the ovarian medulla after birth, we generated a tamoxifen-inducible knock-in mouse model by T2 placing an IRES-CreER cassette after the endogenous Foxl2 T2 exon to produce Foxl2-CreER mice (Supplementary Material, Fig. S1A and B). This mouse model allowed us to use tamoxifen to induce the labeling of granulosa cells of only the first wave of activated follicles after birth, but to avoid labeling the entire pool of follicles, as described later. T2 We crossed the Foxl2-CreER mice with the mT/mG reporter mice. As illustrated in Figure 1A, labeled pregranulosa cells show a switch from red fluorescence (mT) to green fluorescence (mG) (12). Throughout the manuscript, we refer to green T2 cells as ‘labeled’ and red cells as ‘unlabeled’. As shown in Figure 1. Labeling the first wave of activated follicles in Foxl2-CreER ;mT/mG mouse ovaries. (A) Illustration of the tamoxifen-induced labeling of pregranu- Figure 1B, when we gave a single injection of tamoxifen to preg- losa and granulosa cells of the first wave of activated follicles in postnatal Foxl2- nant females at Embryonic Day (E) 16.5 [15 mg/kg body weight T2 CreER ;mT/mG mice. In Foxl2-expressing pregranulosa and granulosa cells, T2 (BW), 600 mg per mouse] all activated follicles in Postnatal the CreER recombinase is not active and the cells express membrane-targeted Day (PD) 3 ovaries of the offspring showed labeling of the gran- Tomato (mT), a red fluorescent protein. Upon tamoxifen induction, the activated T2 CreER recombinase mediates the deletion of the mT region and switches on the ulosa cells with green fluorescence (Fig. 1B, arrows). The prim- expression of membrane-targeted green fluorescent protein (mG) resulting in ordial follicles in the cortical region were unlabeled and their follicles that are labeled with green pregranulosa and granulosa cells. (B and pregranulosa cells still expressed red fluorescence (Fig. 1B, C) Efficient labeling of the first wave of activated follicles by administration of arrowheads). Similarly, at PD13, almost all growing follicles tamoxifen to pregnant females. Pregnant females were given a single intraperito- in the ovarian medulla were labeled with green fluorescent gran- neal injection of tamoxifen at E16.5, and the ovaries of their pups at PD3 (B) and PD13 (C) were analyzed. Note that the first wave of activated follicles in the ulosa cells (Fig. 1C, arrows), whereas the primordial follicles in ovarian medulla was efficiently labeled (arrows) whereas primordial follicles the cortical region remained unlabeled and showed red fluores- in the ovarian cortex were not labeled (arrowheads). Although not all pregranu- cence (Fig. 1C, arrowheads). losa and granulosa cells were labeled by the low dosage of tamoxifen used, a fol- T2 As a comparison, when Foxl2-CreER ;mT/mG pups were licle was considered successfully labeled as long as one somaticcell turnedgreen. (D) Labeling of all follicles by tamoxifen injection at PD5. The female mice were injected with tamoxifen at PD5 (15 mg/kg BW, 60 mg per given a single intraperitoneal injection of tamoxifen at PD5 and were analyzed at pup), which is the time when the cortical primordial follicles PD13. This protocol of tamoxifen administration led to the labeling of all folli- are being formed (6), we could label both the granulosa cells cles, including the first wave of follicles (arrows) and the cortical primordial fol- of all growing follicles (Fig. 1D, arrows) and the pregranulosa licles (arrowheads). cells of all cortical primordial follicles (Fig. 1D, arrowheads) at PD13. As negative controls, no labeling of cells was seen in T2 ages and to determine the time periods that they remain in the Foxl2-CreER ;mT/mG mice injected with vehicle alone (Sup- mouse ovaries as well as their contributions to the onset of plementary Material, Fig. S2A) or in mT/mG mice injected puberty and to fertility. with tamoxifen (Supplementary Material, Fig. S2B). In addition, the development of ovarian follicles was not influenced by em- Tracing the first wave of activated follicles to the onset bryonic exposure to tamoxifen (see Supplementary Material, of puberty (PD23) Text and Supplementary Material, Fig. S3A–C). At PD23, the time when puberty begins, most of the growing These results clearly indicate that by giving tamoxifen to preg- follicles (including primary and further-developed follicles) nant females at E16.5, we have specifically labeled the first wave contained granulosa cells with green fluorescence (Fig. 2A, of activated follicles in the ovarian medulla, but have avoided la- arrows). This green fluorescence showed that these follicles beling the primordial follicles in the ovarian cortex. had developed from the labeled first wave of follicles that were activated after birth (Fig. 1B and C, arrows). Some of the first The first wave of activated follicles contribute to the onset wave follicles had even reached the antral stage by PD23 of puberty and to fertility in young adulthood (Fig. 2A, AF). The specific labeling of the first wave of activated follicles Some cortical primordial follicles also started to become acti- allowed us to follow their in vivo development to different vated at this age. As shown in Figure 2A (inset, arrowheads), Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 922 Human Molecular Genetics, 2014, Vol. 23, No. 4 T2 Figure 2. Tracing the development of the first wave of activated follicles in Foxl2-CreER ;mT/mG mouse ovaries. Tamoxifen was administered to pregnant females T2 at E16.5. The Foxl2-CreER ;mT/mG pups born to these females were sacrificed at different ages to analyze their ovaries. (A and B) Most of the growing follicles in ovaries collected from PD23 (A) and PD35 (B) mice show labeled green granulosa cells (arrows), indicating that they have developed from the first wave of activated follicles. The presence of labeled antral follicles (AF) indicates that follicles from the first wave contribute to the onset of puberty and to fertility during young adult- hood. A few unlabeled red growing follicles (arrowheads) can also be seen, indicating that they developed from the cortical primordial follicles. (C) At PD45, both labeled green (arrows) and unlabeled red (arrowheads) growing follicles are observed, and some unlabeled follicles have reached the antral stage (AF). (D) At PD90, the majority of the growing follicles, including antral follicles (AF), are unlabeled (arrowheads). The number of labeled follicles (arrow) has decreased, and most of these are at advanced stages or have differentiated into corpora lutea (CL). (E) At PD105, only green debris can be observed in the ovary (arrow). (F) At 4 months, no green fluorescent cells can be observed. (G) Quantification of labeled and unlabeled growing follicles at different ages. At PD105, the first wave of activated follicles becomes depleted, and the follicles grown from cortical primordial follicles become the only source of the growing follicle pool. (H) Quantification of labeled and unlabeled antral follicles and CL as a marker for their contributions to fertility. At PD105, no antral follicles or CL are from the first wave of follicles. In G and H, n ¼ 4 for PD13, n ¼ 4 for PD23, n ¼ 5 for PD35, n ¼ 5 for PD45, n ¼ 5 for PD60, n ¼ 5 for PD90, n ¼ 4 for PD105 and n ¼ 4 for 4 months. Values are means+ SD. Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 923 primary and secondary follicles showing red fluorescence were cortical primordial follicles served as the only source of seen in the vicinity of the cortical region indicating that these fol- growing follicles (Fig. 2G). licles developed from the unlabeled cortical primordial follicles To summarize, the first wave of follicles dominated the (Fig. 1 B and C, arrowheads). growing follicle pool until young adulthood (PD45) and facili- As quantified in Figure 2G, at PD23, the first wave of follicles tated the onset of puberty and the entry into sexual maturity. accounted for 89.8+ 5.1% (mean+ SD) of the growing follicle Then, the proportion of the first wave of follicles had dropped pool. Accordingly, 10.2+ 5.1% of the growing follicles exhib- to 20% of all growing follicles at PD60 and to 2% at PD90 ited red fluorescence showing that the cortical primordial folli- (Fig. 2G). These results suggest that the first wave of follicles cles had started to replace the first wave of follicles at PD23. provides fertility up to 3 months of age (Fig. 2H). Notably, al- Based on these results, we conclude that the majority of the though the activation of the first wave of follicles was synchron- growing follicles at the time of puberty onset originate from ous, their further development was asynchronous. Some of them the first wave of activated follicles and are the main follicular grew more rapidly and reached the antral stage at PD23 (Fig. 2A, population that contributes to the onset of puberty. AF) whereas others developed more slowly and ovulated around PD90 (Fig. 2D, CL). Tracing the first wave of activated follicles to the time of sexual maturity (PD35) T2 Generation of a Sohlh1-CreER knock-in mouse model At the time of sexual maturity (around PD35), 79.8+ 11.7% for labeling primordial follicles in adult mouse ovaries (Fig. 2G) of the growing follicles were still labeled with green granulosa cells (Fig. 2B, arrows), indicating that the first wave To trace the in vivo development of cortical primordial follicles of follicles was still dominant in the ovary. As shown in throughout adulthood, we generated another knock-in mouse T2 Figure 2B (AF) and quantified in Figure 2H, 100% of the antral- model in which CreER was driven by the endogenous pro- stage follicles in the ovary were labeled with green fluorescence moter of Sohlh1 (Supplementary Material, Fig. S4A and B). and thus had developed from the first wave of follicles. There- The Sohlh1 promoter is only active in the oocytes of primordial fore, the first wave of follicles contributed to early fertility at follicles (13), and by crossing these mice to a Rosa26 reporter the time of sexual maturity. (R26R) mouse line, we were able to exclusively label the Meanwhile, the proportion of unlabeled red growing follicles oocytes of primordial follicles (Fig. 3A). T2 (Fig. 2B, arrowheads) had increased to 20.2+ 11.7% (Fig. 2G), We injected the Sohlh1-CreER ;R26R mice with tamoxifen at indicating that the activated cortical primordial follicles were 3monthsofage(80 mg/kgBW, 2 mgpermouse,dailyfor3con- gradually replacing the first wave of follicles. secutive days). Five days after the first tamoxifen injection, prim- ordial follicles in the cortical region were specifically labeled as Tracing the first wave of activated follicles to young shown by the b-galactosidase staining in the oocytes (Fig. 3B). adulthood (PD45–90) These labeled primordial follicles were always found in the At PD45, 60.4+ 10.0% (Fig. 2G) of the growing follicles had ovaries as seen at 7 weeks (Fig. 3C) andat 14months (the originated from the first wave of follicles (Fig. 2C, arrows), longest time studied) (Fig. 3D) after tamoxifen injection. and 39.6+ 10.0% (Fig. 2G) of the growing follicles had origi- Notably, no oocytes within primary or further-developed folli- nated from cortical primordial follicles (Fig. 2C, arrowheads). cles (Fig. 3B, arrows) were labeled 5 days after the first tamoxifen At this point in development, some unlabeled red follicles had administration. This showed a stringent labeling specificity for developed as far as the antral stage (Fig. 2C, AF), and these primordial follicles that was in accordance with the reported made up 36.3+ 7.9% of all antral follicles (Fig. 2H). Sohlh1 promoter activity (13). Neither vehicle-injected Sohlh1- T2 At PD90, the percentage of labeled green follicles (Fig. 2D, CreER ;R26R mice (Supplementary Material, Fig. S5A) nor arrow) dropped to 2.4+ 0.7% (Fig. 2G). In contrast, the un- tamoxifen-injected R26R mice (Supplementary Material, labeled red growing follicles (Fig. 2D, arrowheads) represented Fig. S5B) had any labeled primordial follicles, and this confirmed T2 97.6+ 0.7% of all growing follicles and had become dominant the specificity of this inducible Sohlh1-CreER mouse model. In in the ovary (Fig. 2G). As quantified in Figure 2H, 20.9+ 8.2% addition, the development of ovarian follicles was not influenced of all antral follicles and corpora lutea (CL) were labeled with by tamoxifen injection at adulthood (see Supplementary Material, green fluorescent cells at PD90 (Fig. 2D, arrow and CL), Text and Supplementary Material, Fig. S6 A and B). whereas 79.1+ 8.2% of all antral follicles and CL contained We found that 30% of the primordial follicles in the adult only red fluorescent cells at this age. mouse ovaries could be labeled by tamoxifen administration These results showed that the first wave of follicles was still (Fig. 3B). This level of efficiency, however, did not affect our actively contributing to fertility in young adulthood, but follicles ability to trace the follicular development because we used suf- that developed from the cortical primordial follicles were grad- ficient numbers of experimental animals as described later. ually but steadily becoming the major follicular resource between PD45 and PD90. Tracing the developmental dynamics of primordial follicles in the adult mouse ovary Tracing the first wave of activated follicles up to 4 months of age Only a small amount of green cellular debris was observed in the To understand the nature of primordial follicular development in ovaries of PD105 mice (Fig. 2E, arrow). At 4 months of age, all adult life, we traced the in vivo development of labeled primor- T2 follicles and CL were composed of unlabeled red somatic cells dial follicles to different ages in the Sohlh1-CreER ;R26R (Fig. 2F). Therefore, the first wave of follicles was present in mice. The classification of follicles was based on well- the mouse ovary up to 3 months after birth. After that, the established standards (14). For this work, a total of 140 mice Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 924 Human Molecular Genetics, 2014, Vol. 23, No. 4 Figure 3. Labeling of primordial follicles in adult mouse ovaries using the T2 Sohlh1-CreER ;R26R mouse model. (A) Illustration of the tamoxifen-induced T2 labeling of adult primordial follicles in the Sohlh1-CreER ;R26R mice. In the T2 oocytes of primordial follicles, the CreER mediated by the Sohlh1 promoter is not active and no b-galactosidase is expressed before tamoxifen is given. T2 Upon tamoxifen administration, the CreER recombinase becomes activated and mediates the removal of the STOP sequence that is in front of the lacZ (the gene encoding b-galactosidase) cDNA. This results in the expression of lacZ in the oocytes and the production of blue-colored b-galactosidase staining. (B–D) Labeling of only primordial follicles in adult mouse ovaries. T2 Three-month-old Sohlh1-CreER ;R26R females were given intraperitoneal injections of tamoxifen for 3 consecutive days. (B) The labeled primordial folli- cles (shown by the blue dots in the oocytes) can be seen 5 days after the injection. Arrows indicate the unlabeled growing follicles. (C) Labeled primordial follicles are still seen in ovaries 7 weeks after tamoxifen injection. (D) Labeled primordial follicles are still seen in the ovaries 14 months after tamoxifen injection. were sacrificed between 5 and 57 days after the first tamoxifen injection at 3 months of age. Tracing the development of adult primordial follicles to the primary stage We first studied the transition from primordial to primary follicles. As shown in Figure 4A, labeled primary follicles were first observed in one out of four mice (25%) sacrificed 7 days after the first tamoxifen injection. Two out of four (50%) mice showed labeled primary follicles 8 days after tamoxifen injection, and from 9 days onward, labeled primary follicles were seen in all sacrificed mice (100%). We conclude, therefore, that it takes 7–9 Figure 4. Tracing the development of primordial follicles in adult Sohlh1- days for primordial follicles to develop into primary stage follicles. T2 T2 CreER ;R26R mice. Sohlh1-CreER ;R26R adult females were injected with The fact that labeled primary follicles were always observed tamoxifen and sacrificed at different time points. The Y-axis is the percentage in all 124 mice sacrificed from 9 to 57 days after tamoxifen injec- of mice showing labeled follicles. The numbers of mice used for each time point are shown above the bars. Representative labeled follicles are shown in tion (Fig. 4A) indicated that the activation of primordial follicles the right panels. (A) A minimum of 7 days is needed for labeled primordial fol- might take longer than 9 days and that such activation occurs in a licles to reach the primary stage. Labeled primary follicles can be found in all progressive manner as previously hypothesized (3,15). mice sacrificed from 9 to 57 days after tamoxifen injection. (B) A minimum of 23 days is needed for labeled primordial follicles to develop into secondary fol- licles. (C) A minimum of 37 days is needed for labeled primordial follicles to Tracing of adult primordial follicles to the secondary stage develop into the early antral stage. (D) A minimum of 47 days is needed for We next looked into how the primordial follicles reached the sec- labeled primordial follicles to develop into the antral stage. ondary stage. As shown in Figure 4B, labeled secondary follicles 24 days after tamoxifen injection, and in almost all mice sacri- were seen in one out of five (20%) mice sacrificed 23 days after ficed from 25 to 57 days after tamoxifen injection. These tamoxifen injection, four out of five (80%) mice sacrificed Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 925 results indicated that a minimum of 23–24 days was needed for the primordial follicles to reach the secondary stage. If 7 days are assumed to be the minimum time needed to develop from primordial to primary follicles, the shortest dwell- ing time of primary follicles in mouse ovaries is 16 days. Tracing of adult primordial follicles to the early antral stage As shown in Figure 4C, labeled early antral follicles were found in three out of four (75%) mice sacrificed 37 days after tamoxifen injection, 11 out of 15 (73%) mice sacrificed from 38 to 43 days after tamoxifen injection, and in all mice sacrificed from 44 to 57 days after tamoxifen injection. Thus, the development from primordial follicles to the early antral stage takes at least 37 days. If a minimum of 23 days is assumed for primordial follicles to reach the secondary stage, the time for secondary follicles to develop into early antral follicles in mouse ovaries is at least 14 days. Tracing of adult primordial follicles to the antral stage No labeled antral follicles were seen in 87 mice sacrificed from 5 to 45 days after tamoxifen injection or in 7 mice sacrificed 46 days after tamoxifen injection (Fig. 4D). After that, a stable pro- Figure 5. Labeled primordial follicles persisted to the end of the reproductive T2 T2 portion of mice (30%–50%) sacrificed from 47 to 57 days after lifespan in Sohlh1-CreER ;R26R mouse ovaries. Sohlh1-CreER ;R26R adult females were given tamoxifen and sacrificed 10 months later. (A) A representa- tamoxifen injection contained labeled antral follicles (Fig. 4D). tive labeled primary follicle with a labeled primordial follicle in the same field. Thus, it takes a minimum of 47 days for a primordial follicle to go (B) A representative labeled secondary follicle with a labeled primordial follicle through the different developmental stages and become an antral in the same field. (C) A representative labeled early antral follicle. (D) A repre- follicle in adult mouse ovaries. sentative labeled antral follicle. The minimum dwelling time for an early antral follicle can be calculated as 10 days because it takes at least 37 days to develop mice over a period of 2 to 14 months after tamoxifen injection. from primordial to early antral follicles. Labeled primordial, primary, secondary and early antral follicles were always observed in all of the mice sacrificed at various Comparison of the developmental dynamics of the postnatal times during this period (Fig. 3D and Fig. 5A–C). In addition, and adult primordial follicles labeled antral follicles were found in mice sacrificed 10 The developmental dynamics of adult primordial follicles in the months after tamoxifen injection (13 months old) (Fig. 5D). ovarian cortex are distinct from those of the first wave of acti- This result indicated that the primordial follicles labeled in vated primordial follicles in the ovarian medulla. We have sum- early adulthood were competent for activation after a long marized the age-dependent minimal times needed for their period of dormancy and could contribute to ovulation. This respective development in Table 1. finding is in accordance with the notion that a fixed pool of prim- ordial follicles serves as the only source of developing follicles throughout the reproductive lifespan of mice (16–18). Long-term tracing of primordial follicles To trace the development of primordial follicles throughout the entire mouse reproductive lifespan, we labeled the adult cortical DISCUSSION primordial follicles by administering tamoxifen to 17 Sohlh1- T2 In the mammalian ovary, progressive activation of primordial CreER ;R26R mice at 3 months of age and sacrificed the follicles serves as the source of fertilizable ova. However, very little is known about the developmental dynamics of primordial follicles under physiological conditions. In this study, we have Table 1. The distinct age-dependent minimal developmental times of the two T2 classes of primordial follicles generated tamoxifen-inducible Foxl2-CreER and Sohlh1- T2 CreER knock-in mouse models to specifically label the first wave of postnatally activated primordial follicles and the adult Primordial Primordial to Primordial to Primordial to primary secondary early antral to antral cortical primordial follicles, respectively. We showed that the first wave of activated primordial follicles First wave of Right after 8 days 13 days 23 days remain in the mouse ovaries for 3 months after birth and con- activated formation tribute to the onset of puberty and to fertility during early adult- primordial follicles hood. During this period, primordial follicles that are formed Adult cortical 7–9 days 23–24 days 37 days 47 days after birth and located in the ovarian cortex gradually replace primordial the first wave of follicles until they become the sole source of follicles follicles from 3 months until the end of the reproductive life. Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 926 Human Molecular Genetics, 2014, Vol. 23, No. 4 T2 We have also determined the age-dependent time periods Generation of a Foxl2-CreER knock-in mouse model to required for the two classes of primordial follicles to develop label the first postnatal wave of activated follicles into advanced stages and have shown that the developmental dy- The mouse Foxl2 gene contains one exon, and we aimed to keep namics were completely different for the two types of primordial the original open reading frame of Foxl2 intact. An 8 kb region follicles (Table 1). Moreover, our follicular tracing data showed used to construct the targeting vector was first sub-cloned from a that primordial follicles that are labeled early in life can persist in C57BL/6 BAC clone that contained the full-length Foxl2 allele the ovaries until the end of reproductive life and that these folli- (RP23: 142D12). As shown in Supplementary Material, Figure cles were competent to develop into antral stage. S1A, the long arm extended 5.5 kb from the 5 end to the site The visual in vivo tracing of follicles performed in this work of the cassette insertion after the coding region of Foxl2. The provides a precise picture of long-term follicular development short arm extended 2.5 kb from the site of the cassette insertion starting from the primordial stage. Such accurate tracing has T2 to the 3 end. An IRES-CreER -FRT-Neo-FRT cassette was not been possible by autoradiographic labeling of follicles inserted at 259 bp downstream of the STOP codon of Foxl2 (19,20). We believe that our results provide a relatively conclu- using Red/ET recombineering technology (Gene Bridges, sive picture of how the two distinct classes of primordial follicles Germany). develop under normal physiological conditions. Such an under- The targeting constructwas sub-cloned into a backbone vector standing is valuable for studying ovarian organogenesis and pSP72 (Promega, US) containing an ampicillin selection cas- physiology as well as ovarian diseases that are related to sette for re-transformation of the construct prior to electropor- altered follicular development. ation. Ten micrograms of the targeting vector were linearized It has been generally believed that the first wave of activated by NotI and then transfected by electroporation of BA1 follicles are anovulatory (5,7) even though oocytes from the (C57BL/6 × 129/SvEv) (Hybrid) embryonic stem (ES) cells. first wave of follicles can mature in vitro and generate live After selection with G418 antibiotic, surviving clones were pups (21,22). Our data clearly show, however, that the first selected by PCR and Southern blot analysis following standard wave of follicles plays an active role in providing mature eggs procedures (Supplementary Material, Fig. S1B). The positive until 3 months of age. Thus, it is likely that the fast-growing clone was microinjected into C57BL/6 blastocysts following first wave of follicles facilitates the establishment of the hypo- standard procedures. Resulting chimeric mice with a high per- thalamic–pituitary–ovarian axis and thereby plays a key role centage agouti coat color were mated to wild-type C57BL/6 in the onset of puberty and the initiation of reproductive life. mice to generate F1 heterozygous offspring. The heterozygous Given the fact that the growth pattern of the first wave of acti- mice were healthy and fertile. The neomycin expression cassette vated follicles is conserved among mammals (including T2 was excised in vivo by crossing F1 Foxl2-CreER mice with humans) (23,24), it can be postulated that the fertility of transgenic mice expressing flippase (FLP) (003800, Jackson La- women from puberty onset to young adulthood might rely on T2 boratory, US). Foxl2-CreER males were mated with mT/mG the first wave of follicles that are already activated at the fetal females (007576, Jackson Laboratory) to generate Foxl2- stage. In contrast to this first wave of activated follicles, we T2 CreER ;mT/mG mice. have shown that the relatively slow-growing adult primordial follicles contribute to continuous ovulation throughout the middle and late stages of reproductive life, and the gradual but T2 Generation of Sohlh1-CreER knock-in mice to label steady loss of the cortical primordial follicles leads to reproduct- primordial follicles in adulthood ive aging. It is possible that the developmental dynamics of adult primordial follicles in mice might be similar to adult primordial A 13 kb region used to construct the targeting vector was first follicle development in humans. sub-cloned from a positively identified C57BL/6 (RP23: 57O11) Taken together, this study presents several lines of in vivo evi- BAC clone that contained the full-length Sohlh1 allele. The dence that clarify the distinct developmental dynamics of two region was designed such that the short arm extended 2.3 kb 5 classes of primordial follicles under physiological conditions to the ATG of exon 1. The long arm was located on the 3 side T2 and provides a solid foundation for studying how primordial fol- of exon 8 and was 6.5 kb long. The CreER -FRT-Neo-FRT cas- licles develop in humans. sette replaced 4.2 kb of the gene, including exons 1–8 at the ATG site (Supplementary Material, Fig. S4A). The targeting construct was sub-cloned into a 2.4 kb pSP72 MATERIALS AND METHODS backbone vector. Ten micrograms of targeting vector were line- arized by SwaI. The electroporation, ES cell characterization Reagents (Supplementary Material, Fig. S4B), blastocyst injection and Restriction enzymes were purchased from New England neomycin cassette removal were performed with standard proce- Biolabs, UK. Antibiotics were purchased from Invitrogen, US. dures as described earlier. The heterozygous mice were found to T2 Tamoxifen and corn oil were purchased from Sigma-Aldrich, be healthy and fertile. Sohlh1-CreER males were mated with Germany. Reagents for b-galactosidase staining [paraformalde- R26R females (003474, Jackson Laboratory) to generate the T2 hyde, magnesium chloride, ethylene glycol tetraacetic acid, Sohlh1-CreER ;R26R mice. sodium deoxycholate, NP40, potassium ferricyanide, potassium All mice were crossed with C57BL/6 mice (Charles River, ferrocyanide, 5-Bromo-4-chloro-3-indolyl b-D-galactoside US) for six generations to obtain identical genetic backgrounds. (X-gal) and 0.1% nuclear fast red solution] were all purchased Mice were housed in controlled environmental conditions with from Sigma–Aldrich. Paraffin and hematoxylin were purchased free access to water and food. Illumination was on between from Histolab, Sweden. 0600 and 1800. Experimental protocols were approved by the Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 927 regional ethical committee of the University of Gothenburg, mice. Briefly, ovaries were fixed in 4% paraformaldehyde for 1 h Sweden. at 48C and then rinsed 3 times for 15 min each in a buffer consist- ing of 1 × PBS (pH 7.4), 2 mM magnesium chloride, 5 mM ethylene glycol tetraacetic acid, 0.01% sodium deoxycholate T2 Tamoxifen administration in Foxl2-CreER ;mT/mG mice and 0.02% NP40 at room temperature. The ovaries were then incubated with a staining solution consisting of 1 × PBS (pH Tamoxifen was resuspended at 100 mg/mL in 95% ethanol and 7.4), 2 mM magnesium chloride, 0.01% sodium deoxycholate, further diluted with corn oil to a final concentration of 20 mg/mL 0.02% NP40, 5 mM potassium ferricyanide, 5 mM potassium (25). For embryonic injection, mT/mG females were first T2 ferrocyanide and 1 mg/mL X-gal at 378C overnight. plugged by Foxl2-CreER males. A single intraperitoneal in- Following staining, the ovaries were re-fixed in 4% parafor- jection of tamoxifen (15 mg/kg BW, 600 mg per mouse) was maldehyde for 8 h, embedded in paraffin, serially cut into given to pregnant females at E16.5 (the day the vaginal plug 8-mm sections and counterstained with 0.1% nuclear fast red so- appeared was counted as E0.5). Because the tamoxifen injection lution. The sections were examined under a Zeiss Axio Scope A1 compromised the ability of the mice to have spontaneous vaginal upright microscope. A follicle was considered to be labeled if a delivery, pups were delivered by Cesarean section at E20. Surro- blue dot was observed in the cytoplasm of the oocyte. gate mothers were used to nurture the pups. For postnatal injec- T2 tion, Foxl2-CreER ;mT/mG females were given a single intraperitoneal injection of tamoxifen at PD5 at a dose of Quantification of ovarian follicles for evaluating the 15 mg/kg BW. influence of tamoxifen on follicular development For quantification of ovarian follicles and the morphological Histological analysis and quantification of ovarian studies shown in Supplementary Material, Figs S3 and S6, T2 follicles from Foxl2-CreER ;mT/mG mice ovaries were fixed in 4% paraformaldehyde, dehydrated and em- T2 bedded in paraffin. Paraffin-embedded ovaries were serially cut To observe the ovarian morphology of Foxl2-CreER ;mT/mG into 8-mm sections and stained with hematoxylin for morpho- mice, ovaries were fixed in 4% paraformaldehyde, dehydrated logical observation. Ovarian follicles at different stages of devel- and embedded in paraffin. The paraffin-embedded ovaries opment, including primordial (type 2), primary (type 3), were serially cut into 8-mm sections and rehydrated. The fluores- secondary (type 4 and 5), early antral (type 6) and antral (type cent images were taken with a Zeiss Axio Scope A1 upright 7) follicles were counted in all sections of an ovary based on microscope installed with filter sets for mT (tdTomato, 554/ the well-accepted standards established by Pedersen and Peters 581 nm) and mG (EGFP, 488/507 nm) and merged with the (14). Follicles that contained oocytes with clearly visible Zeiss AxioVision software. nuclei were scored in each section, as previously reported (28). For quantification of follicles, the fluorescent images from Judged from careful morphological analysis, the incidence of every five sections per ovary were used to count the follicles. counting the same follicle twice or of missing a follicle was low. The classification of follicular development was based on well- established standards by Pedersen and Peters (14). The numbers of growing follicles with green (designated as labeled) or red Statistical analysis (designated as unlabeled) granulosa cells in each image were All experiments were repeated at least 3 times. In Supplementary counted. Material, Figures S3 and S6, the differences between the numbers of ovarian follicles in tamoxifen- and vehicle-injected T2 mice were calculated with Student’s t-test, and the difference Tamoxifen administration in Sohlh1-CreER ;R26R mice was considered to be significant if P , 0.05. To reach the optimal labeling efficiency, 3-month-old adult T2 Sohlh1-CreER ;R26R female mice were given an intraperito- neal injection of 80 mg/kg BW (2 mg per mouse) tamoxifen SUPPLEMENTARY MATERIAL T2 per day for three consecutive days. The CreER -mediated Supplementary Material is available at HMG online. DNA recombination can take place within 24 h after tamoxifen injection (26,27), so the day of the first tamoxifen injection was counted as day 0. The labeled primordial follicles can be clearly Conflict of Interest statement. None declared. visualized by b-galactosidase staining 5 days after the first tam- oxifen injection. For tracing the follicular development, 140 mice were sacrificed between 5 and 57 days after the first tamoxi- FUNDING fen injection. For long-term tracing of primordial follicles, 17 mice were sacrificed between 2 to 14 months after the first tam- This work was supported by grants (to K. L.) from the Young Re- oxifen injection. searcher Award (Umea˚ University), the Jane and Dan Olssons Foundation, the LUA/ALF-medel Va¨stra Go¨talandsregionen, AFA Insurance, the Swedish Research Council, the Swedish T2 b-galactosidase staining of Sohlh1-CreER ;R26R ovaries Cancer Foundation, the Faculty of Natural Science of the Uni- Whole-mountb-galactosidase staining was performed to visual- versity of Gothenburg and the Novo Nordisk Foundation T2 ize the labeled follicles in the ovaries of Sohlh1-CreER ;R26R (Denmark). Funding to pay the Open Access publication Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 928 Human Molecular Genetics, 2014, Vol. 23, No. 4 14. Pedersen, T.and Peters, H. (1968)Proposalfor a classification of oocytesand charges for this article was provided by the Swedish Research follicles in the mouse ovary. J. Reprod. Fertil., 17, 555–557. Council. 15. McGee, E.A. and Hsueh, A.J. (2000) Initial and cyclic recruitment of ovarian follicles. Endocr. Rev., 21, 200–214. 16. Zuckerman, S. (1951) The number of oocytes in the mature ovary. Recent REFERENCES Prog. Horm. Res., 6, 63–109. 1. Eppig, J.J., Bivens, C.M., Viveiros, M.M. and de la Fuente, R. (2003) 17. Lei, L. and Spradling, A.C. (2013) Female mice lack adult germ-line stem Regulation of mammalian oocyte maturation. In Peter, C.K.L. and Eli, Y.A. cells but sustain oogenesis using stable primordial follicles. Proc. Natl. (eds), The Ovary, 2nd edn. Academic Press, San Diego, USA, pp. 113–129. Acad. Sci. USA, 110, 8585–8590. 2. Matzuk, M.M., Burns, K.H., Viveiros, M.M. and Eppig, J.J. (2002) 18. Zhang, H., Zheng, W., Shen, Y., Adhikari, D., Ueno, H. and Liu, K. (2012) Intercellular communication in the mammalian ovary: oocytes carry the Experimental evidence showing that no mitotically active female germline conversation. Science, 296, 2178–2180. progenitors exist in postnatal mouse ovaries. Proc. Natl. Acad. Sci. USA, 3. Adhikari, D. and Liu, K. (2009) Molecular mechanisms underlying the 109, 12580–12585. activation of mammalian primordial follicles. Endocr. Rev., 30, 438–464. 19. Oakberg, E.F. (1979) Timing of oocyte maturation in the mouse and its 4. Reddy, P., Zheng, W. and Liu, K. (2010) Mechanisms maintaining the relevance to radiation-induced cell killing and mutational sensitivity. Mutat. dormancy and survival of mammalian primordial follicles. Trends Res., 59, 39–48. Endocrinol. Metab., 21, 96–103. 20. Pedersen, T. (1970) Follicle kinetics in the ovary of the cyclic mouse. Acta. 5. Hirshfield, A.N. (1992) Heterogeneity of cell populations that contribute to Endocrinol. (Copenh), 64, 304–323. the formation of primordial follicles in rats. Biol. Reprod., 47, 466–472. 21. Eppig, J.J., O’Brien, M.J., Wigglesworth, K., Nicholson, A., Zhang, W. and 6. Mork, L., Maatouk, D.M., McMahon, J.A., Guo, J.J., Zhang, P., McMahon, King, B.A. (2009) Effect of in vitro maturation of mouse oocytes on the A.P. and Capel,B. (2012) Temporal differences in granulosa cell specification health and lifespan of adult offspring. Hum. Reprod., 24, 922–928. in the ovary reflect distinct follicle fates in mice. Biol. Reprod., 86,37. 22. O’Brien, M.J., Pendola, J.K. and Eppig, J.J. (2003) A revised protocol for in 7. Eppig, J.J. and Handel, M.A. (2012) Origins of granulosa cells clarified and vitro development of mouse oocytes from primordial follicles dramatically complexified by waves. Biol. Reprod., 86, 34. improves their developmental competence. Biol. Reprod., 68, 1682–1686. 8. Hirshfield, A.N. and DeSanti, A.M. (1995) Patterns of ovarian cell 23. Peters, H., Byskov, A.G., Himelstein-Braw, R. and Faber, M. (1975) proliferation in rats during the embryonic period and the first three weeks Follicular growth: the basic event in the mouse and human ovary. J. Reprod. postpartum. Biol. Reprod., 53, 1208–1221. Fertil., 45, 559–566. 9. Peters, H. (1969) The development of the mouse ovary from birth to 24. Lintern-Moore, S., Peters, H., Moore, G.P. and Faber, M. (1974) Follicular maturity. Acta. Endocrinol. (Copenh), 62, 98–116. development in the infant human ovary. J. Reprod. Fertil., 39, 53–64. 10. Schmidt,D.,Ovitt, C.E., Anlag, K.,Fehsenfeld,S., Gredsted, L., Treier, A.C. 25. John, G.B., Gallardo, T.D., Shirley, L.J. and Castrillon, D.H. (2008) Foxo3 is and Treier, M. (2004) The murine winged-helix transcription factor Foxl2 is a PI3K-dependent molecular switch controlling the initiation of oocyte required for granulosa cell differentiation and ovary maintenance. growth. Dev. Biol., 321, 197–204. Development, 131, 933–942. 26. Hayashi, S. and McMahon, A.P. (2002) Efficient recombination in diverse 11. Uda, M., Ottolenghi, C., Crisponi, L., Garcia, J.E., Deiana, M., Kimber, W., tissues by a tamoxifen-inducible form of Cre: a tool for temporally regulated Forabosco, A., Cao, A., Schlessinger, D. and Pilia, G. (2004) Foxl2 gene activation/inactivation in the mouse. Dev. Biol., 244, 305–318. disruption causes mouse ovarian failure by pervasive blockage of follicle 27. Feil, R., Brocard, J., Mascrez, B., LeMeur, M., Metzger, D. and Chambon, P. development. Hum. Mol. Genet., 13, 1171–1181. (1996) Ligand-activated site-specific recombination in mice. Proc. Natl. 12. Muzumdar, M.D., Tasic, B., Miyamichi, K., Li, L. and Luo, L. (2007) A Acad. Sci. USA, 93, 10887–10890. global double-fluorescent Cre reporter mouse. Genesis, 45, 593–605. 13. Pangas, S.A., Choi, Y., Ballow, D.J., Zhao, Y., Westphal, H., Matzuk, M.M. 28. Reddy, P., Liu, L., Adhikari, D., Jagarlamudi, K., Rajareddy, S., Shen, Y., and Rajkovic, A. (2006) Oogenesis requires germ cell-specific Du, C., Tang, W., Ha¨ma¨la¨inen, T., Peng, S.L. et al. (2008) Oocyte-specific transcriptional regulators Sohlh1 and Lhx8. Proc. Natl. Acad. Sci. USA, 103, deletion of Pten causes premature activation of the primordial follicle pool. 8090–8095. Science, 319, 611–613. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Human Molecular Genetics Oxford University Press

Two classes of ovarian primordial follicles exhibit distinct developmental dynamics and physiological functions

Loading next page...
 
/lp/oxford-university-press/two-classes-of-ovarian-primordial-follicles-exhibit-distinct-Lf0g5Olca0

References (29)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Oxford University Press
Subject
ARTICLES
ISSN
0964-6906
eISSN
1460-2083
DOI
10.1093/hmg/ddt486
pmid
24087793
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 920–928 doi:10.1093/hmg/ddt486 Advance Access published on October 1, 2013 Two classes of ovarian primordial follicles exhibit distinct developmental dynamics and physiological functions { { Wenjing Zheng , Hua Zhang , Nagaraju Gorre, Sanjiv Risal, Yan Shen and Kui Liu Department of Chemistry and Molecular Biology, University of Gothenburg, Gothenburg SE-405 30, Sweden Received August 26, 2013; Revised September 24, 2013; Accepted September 25, 2013 Inthemammalianovary,progressiveactivationofprimordialfolliclesservesasthesourceoffertilizableova,and disorders in the development of primordial follicles lead to various ovarian diseases. However, very little is known about the developmental dynamics of primordial follicles under physiological conditions, and the fates of distinct populations of primordial follicles also remain unclear. In this study, by generating the Foxl2- T2 T2 CreER and Sohlh1-CreER inducible mouse models, we have specifically labeled and traced the in vivo devel- opment of two classes of primordial follicles, the first wave of simultaneously activated follicles after birth and the primordial follicles that are gradually activated in adulthood. Our results show that the first wave of follicles exists in the ovaries for ∼3 months and contributes to the onset of puberty and to early fertility. The primordial follicles at theovarian cortex gradually replace the first wave of follicles and dominate the ovary after 3 months of age, providing fertility until the end of reproductive life. Moreover, by tracing the time periods needed for prim- ordial follicles to reach various advanced stages in vivo, we were able to determine the exact developmental dy- namics of the two classes of primordial follicles. We have now revealed the lifelong developmental dynamics of ovarian primordial follicles under physiological conditions and have clearly shown that two classes of primor- dial follicles follow distinct, age-dependent developmental paths and play different roles in the mammalian reproductive lifespan. these primordial follicles in the medulla are synchronously acti- INTRODUCTION vated and become the first wave of activated follicles after birth, In mammals, the pool of primordial follicles serves as the source and they have long been considered not to contribute to fertility of developing follicles and fertilizable ova for the entire repro- (5,7). The supporting cells recruited in postnatal mouse ovaries ductive lifespan of the organism (1,2). Whereas the majority of form primordial follicles in the cortical region (5,6). The cortical primordial follicles remain in a dormant state, a limited primordial follicles are believed to be activated gradually as a number of primordial follicles are recruited from the resting fol- means of providing mature ova over the entire course of the licle reservoir into the growing follicle pool. In recent years, the animal’s reproductive life (8,9). molecular mechanisms controlling the activation of primordial In the current study, we have generated two tamoxifen- follicles have begun to be revealed mostly by using genetically inducible knock-in mouse models, the Forkhead box L2 (Foxl2)- T2 modified mouse models, which represent extreme situations in CreER mice and the Spermatogenesis and oogenesis-specific T2 which a particular gene is deleted (for reviews see references 3 basic helix-loop-helix 1 (Sohlh1)-CreER mice. These mouse and 4). Very little is known, however, about the development models have allowed us to specifically label the first wave of acti- of primordial follicles under physiological conditions through- vated follicles in the ovarian medulla and the primordial follicles out the entire female reproductive lifespan. in the cortex, respectively. We have traced the in vivo develop- Previous studies have shown that the supporting cells ment of the two classes of follicles over the lifetimes of the recruited in fetal mouse ovaries form the initial primordial folli- mice, and our results show that the first wave of medullary folli- cles in the medullary region of the ovary (5,6). Once formed, cles and the cortical primordial follicles are distinct populations To whom correspondence should be addressed. Tel: +46 317866585; Fax: +46 317863801; Email: kui.liu@gu.se The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. # The Author 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 921 in terms of their developmental dynamics, their contributions to the onset of puberty and their contributions to fertility during young adulthood and later in reproductive life. RESULTS T2 Generation of a Foxl2-CreER knock-in mouse model for labeling the first wave of activated follicles in postnatal ovaries Foxl2 is a forkhead transcription factor that is expressed in somatic cells of the early XX gonad and in pregranulosa and granulosa cells (10,11). To selectively label the first wave of acti- vated follicles that appear in the ovarian medulla after birth, we generated a tamoxifen-inducible knock-in mouse model by T2 placing an IRES-CreER cassette after the endogenous Foxl2 T2 exon to produce Foxl2-CreER mice (Supplementary Material, Fig. S1A and B). This mouse model allowed us to use tamoxifen to induce the labeling of granulosa cells of only the first wave of activated follicles after birth, but to avoid labeling the entire pool of follicles, as described later. T2 We crossed the Foxl2-CreER mice with the mT/mG reporter mice. As illustrated in Figure 1A, labeled pregranulosa cells show a switch from red fluorescence (mT) to green fluorescence (mG) (12). Throughout the manuscript, we refer to green T2 cells as ‘labeled’ and red cells as ‘unlabeled’. As shown in Figure 1. Labeling the first wave of activated follicles in Foxl2-CreER ;mT/mG mouse ovaries. (A) Illustration of the tamoxifen-induced labeling of pregranu- Figure 1B, when we gave a single injection of tamoxifen to preg- losa and granulosa cells of the first wave of activated follicles in postnatal Foxl2- nant females at Embryonic Day (E) 16.5 [15 mg/kg body weight T2 CreER ;mT/mG mice. In Foxl2-expressing pregranulosa and granulosa cells, T2 (BW), 600 mg per mouse] all activated follicles in Postnatal the CreER recombinase is not active and the cells express membrane-targeted Day (PD) 3 ovaries of the offspring showed labeling of the gran- Tomato (mT), a red fluorescent protein. Upon tamoxifen induction, the activated T2 CreER recombinase mediates the deletion of the mT region and switches on the ulosa cells with green fluorescence (Fig. 1B, arrows). The prim- expression of membrane-targeted green fluorescent protein (mG) resulting in ordial follicles in the cortical region were unlabeled and their follicles that are labeled with green pregranulosa and granulosa cells. (B and pregranulosa cells still expressed red fluorescence (Fig. 1B, C) Efficient labeling of the first wave of activated follicles by administration of arrowheads). Similarly, at PD13, almost all growing follicles tamoxifen to pregnant females. Pregnant females were given a single intraperito- in the ovarian medulla were labeled with green fluorescent gran- neal injection of tamoxifen at E16.5, and the ovaries of their pups at PD3 (B) and PD13 (C) were analyzed. Note that the first wave of activated follicles in the ulosa cells (Fig. 1C, arrows), whereas the primordial follicles in ovarian medulla was efficiently labeled (arrows) whereas primordial follicles the cortical region remained unlabeled and showed red fluores- in the ovarian cortex were not labeled (arrowheads). Although not all pregranu- cence (Fig. 1C, arrowheads). losa and granulosa cells were labeled by the low dosage of tamoxifen used, a fol- T2 As a comparison, when Foxl2-CreER ;mT/mG pups were licle was considered successfully labeled as long as one somaticcell turnedgreen. (D) Labeling of all follicles by tamoxifen injection at PD5. The female mice were injected with tamoxifen at PD5 (15 mg/kg BW, 60 mg per given a single intraperitoneal injection of tamoxifen at PD5 and were analyzed at pup), which is the time when the cortical primordial follicles PD13. This protocol of tamoxifen administration led to the labeling of all folli- are being formed (6), we could label both the granulosa cells cles, including the first wave of follicles (arrows) and the cortical primordial fol- of all growing follicles (Fig. 1D, arrows) and the pregranulosa licles (arrowheads). cells of all cortical primordial follicles (Fig. 1D, arrowheads) at PD13. As negative controls, no labeling of cells was seen in T2 ages and to determine the time periods that they remain in the Foxl2-CreER ;mT/mG mice injected with vehicle alone (Sup- mouse ovaries as well as their contributions to the onset of plementary Material, Fig. S2A) or in mT/mG mice injected puberty and to fertility. with tamoxifen (Supplementary Material, Fig. S2B). In addition, the development of ovarian follicles was not influenced by em- Tracing the first wave of activated follicles to the onset bryonic exposure to tamoxifen (see Supplementary Material, of puberty (PD23) Text and Supplementary Material, Fig. S3A–C). At PD23, the time when puberty begins, most of the growing These results clearly indicate that by giving tamoxifen to preg- follicles (including primary and further-developed follicles) nant females at E16.5, we have specifically labeled the first wave contained granulosa cells with green fluorescence (Fig. 2A, of activated follicles in the ovarian medulla, but have avoided la- arrows). This green fluorescence showed that these follicles beling the primordial follicles in the ovarian cortex. had developed from the labeled first wave of follicles that were activated after birth (Fig. 1B and C, arrows). Some of the first The first wave of activated follicles contribute to the onset wave follicles had even reached the antral stage by PD23 of puberty and to fertility in young adulthood (Fig. 2A, AF). The specific labeling of the first wave of activated follicles Some cortical primordial follicles also started to become acti- allowed us to follow their in vivo development to different vated at this age. As shown in Figure 2A (inset, arrowheads), Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 922 Human Molecular Genetics, 2014, Vol. 23, No. 4 T2 Figure 2. Tracing the development of the first wave of activated follicles in Foxl2-CreER ;mT/mG mouse ovaries. Tamoxifen was administered to pregnant females T2 at E16.5. The Foxl2-CreER ;mT/mG pups born to these females were sacrificed at different ages to analyze their ovaries. (A and B) Most of the growing follicles in ovaries collected from PD23 (A) and PD35 (B) mice show labeled green granulosa cells (arrows), indicating that they have developed from the first wave of activated follicles. The presence of labeled antral follicles (AF) indicates that follicles from the first wave contribute to the onset of puberty and to fertility during young adult- hood. A few unlabeled red growing follicles (arrowheads) can also be seen, indicating that they developed from the cortical primordial follicles. (C) At PD45, both labeled green (arrows) and unlabeled red (arrowheads) growing follicles are observed, and some unlabeled follicles have reached the antral stage (AF). (D) At PD90, the majority of the growing follicles, including antral follicles (AF), are unlabeled (arrowheads). The number of labeled follicles (arrow) has decreased, and most of these are at advanced stages or have differentiated into corpora lutea (CL). (E) At PD105, only green debris can be observed in the ovary (arrow). (F) At 4 months, no green fluorescent cells can be observed. (G) Quantification of labeled and unlabeled growing follicles at different ages. At PD105, the first wave of activated follicles becomes depleted, and the follicles grown from cortical primordial follicles become the only source of the growing follicle pool. (H) Quantification of labeled and unlabeled antral follicles and CL as a marker for their contributions to fertility. At PD105, no antral follicles or CL are from the first wave of follicles. In G and H, n ¼ 4 for PD13, n ¼ 4 for PD23, n ¼ 5 for PD35, n ¼ 5 for PD45, n ¼ 5 for PD60, n ¼ 5 for PD90, n ¼ 4 for PD105 and n ¼ 4 for 4 months. Values are means+ SD. Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 923 primary and secondary follicles showing red fluorescence were cortical primordial follicles served as the only source of seen in the vicinity of the cortical region indicating that these fol- growing follicles (Fig. 2G). licles developed from the unlabeled cortical primordial follicles To summarize, the first wave of follicles dominated the (Fig. 1 B and C, arrowheads). growing follicle pool until young adulthood (PD45) and facili- As quantified in Figure 2G, at PD23, the first wave of follicles tated the onset of puberty and the entry into sexual maturity. accounted for 89.8+ 5.1% (mean+ SD) of the growing follicle Then, the proportion of the first wave of follicles had dropped pool. Accordingly, 10.2+ 5.1% of the growing follicles exhib- to 20% of all growing follicles at PD60 and to 2% at PD90 ited red fluorescence showing that the cortical primordial folli- (Fig. 2G). These results suggest that the first wave of follicles cles had started to replace the first wave of follicles at PD23. provides fertility up to 3 months of age (Fig. 2H). Notably, al- Based on these results, we conclude that the majority of the though the activation of the first wave of follicles was synchron- growing follicles at the time of puberty onset originate from ous, their further development was asynchronous. Some of them the first wave of activated follicles and are the main follicular grew more rapidly and reached the antral stage at PD23 (Fig. 2A, population that contributes to the onset of puberty. AF) whereas others developed more slowly and ovulated around PD90 (Fig. 2D, CL). Tracing the first wave of activated follicles to the time of sexual maturity (PD35) T2 Generation of a Sohlh1-CreER knock-in mouse model At the time of sexual maturity (around PD35), 79.8+ 11.7% for labeling primordial follicles in adult mouse ovaries (Fig. 2G) of the growing follicles were still labeled with green granulosa cells (Fig. 2B, arrows), indicating that the first wave To trace the in vivo development of cortical primordial follicles of follicles was still dominant in the ovary. As shown in throughout adulthood, we generated another knock-in mouse T2 Figure 2B (AF) and quantified in Figure 2H, 100% of the antral- model in which CreER was driven by the endogenous pro- stage follicles in the ovary were labeled with green fluorescence moter of Sohlh1 (Supplementary Material, Fig. S4A and B). and thus had developed from the first wave of follicles. There- The Sohlh1 promoter is only active in the oocytes of primordial fore, the first wave of follicles contributed to early fertility at follicles (13), and by crossing these mice to a Rosa26 reporter the time of sexual maturity. (R26R) mouse line, we were able to exclusively label the Meanwhile, the proportion of unlabeled red growing follicles oocytes of primordial follicles (Fig. 3A). T2 (Fig. 2B, arrowheads) had increased to 20.2+ 11.7% (Fig. 2G), We injected the Sohlh1-CreER ;R26R mice with tamoxifen at indicating that the activated cortical primordial follicles were 3monthsofage(80 mg/kgBW, 2 mgpermouse,dailyfor3con- gradually replacing the first wave of follicles. secutive days). Five days after the first tamoxifen injection, prim- ordial follicles in the cortical region were specifically labeled as Tracing the first wave of activated follicles to young shown by the b-galactosidase staining in the oocytes (Fig. 3B). adulthood (PD45–90) These labeled primordial follicles were always found in the At PD45, 60.4+ 10.0% (Fig. 2G) of the growing follicles had ovaries as seen at 7 weeks (Fig. 3C) andat 14months (the originated from the first wave of follicles (Fig. 2C, arrows), longest time studied) (Fig. 3D) after tamoxifen injection. and 39.6+ 10.0% (Fig. 2G) of the growing follicles had origi- Notably, no oocytes within primary or further-developed folli- nated from cortical primordial follicles (Fig. 2C, arrowheads). cles (Fig. 3B, arrows) were labeled 5 days after the first tamoxifen At this point in development, some unlabeled red follicles had administration. This showed a stringent labeling specificity for developed as far as the antral stage (Fig. 2C, AF), and these primordial follicles that was in accordance with the reported made up 36.3+ 7.9% of all antral follicles (Fig. 2H). Sohlh1 promoter activity (13). Neither vehicle-injected Sohlh1- T2 At PD90, the percentage of labeled green follicles (Fig. 2D, CreER ;R26R mice (Supplementary Material, Fig. S5A) nor arrow) dropped to 2.4+ 0.7% (Fig. 2G). In contrast, the un- tamoxifen-injected R26R mice (Supplementary Material, labeled red growing follicles (Fig. 2D, arrowheads) represented Fig. S5B) had any labeled primordial follicles, and this confirmed T2 97.6+ 0.7% of all growing follicles and had become dominant the specificity of this inducible Sohlh1-CreER mouse model. In in the ovary (Fig. 2G). As quantified in Figure 2H, 20.9+ 8.2% addition, the development of ovarian follicles was not influenced of all antral follicles and corpora lutea (CL) were labeled with by tamoxifen injection at adulthood (see Supplementary Material, green fluorescent cells at PD90 (Fig. 2D, arrow and CL), Text and Supplementary Material, Fig. S6 A and B). whereas 79.1+ 8.2% of all antral follicles and CL contained We found that 30% of the primordial follicles in the adult only red fluorescent cells at this age. mouse ovaries could be labeled by tamoxifen administration These results showed that the first wave of follicles was still (Fig. 3B). This level of efficiency, however, did not affect our actively contributing to fertility in young adulthood, but follicles ability to trace the follicular development because we used suf- that developed from the cortical primordial follicles were grad- ficient numbers of experimental animals as described later. ually but steadily becoming the major follicular resource between PD45 and PD90. Tracing the developmental dynamics of primordial follicles in the adult mouse ovary Tracing the first wave of activated follicles up to 4 months of age Only a small amount of green cellular debris was observed in the To understand the nature of primordial follicular development in ovaries of PD105 mice (Fig. 2E, arrow). At 4 months of age, all adult life, we traced the in vivo development of labeled primor- T2 follicles and CL were composed of unlabeled red somatic cells dial follicles to different ages in the Sohlh1-CreER ;R26R (Fig. 2F). Therefore, the first wave of follicles was present in mice. The classification of follicles was based on well- the mouse ovary up to 3 months after birth. After that, the established standards (14). For this work, a total of 140 mice Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 924 Human Molecular Genetics, 2014, Vol. 23, No. 4 Figure 3. Labeling of primordial follicles in adult mouse ovaries using the T2 Sohlh1-CreER ;R26R mouse model. (A) Illustration of the tamoxifen-induced T2 labeling of adult primordial follicles in the Sohlh1-CreER ;R26R mice. In the T2 oocytes of primordial follicles, the CreER mediated by the Sohlh1 promoter is not active and no b-galactosidase is expressed before tamoxifen is given. T2 Upon tamoxifen administration, the CreER recombinase becomes activated and mediates the removal of the STOP sequence that is in front of the lacZ (the gene encoding b-galactosidase) cDNA. This results in the expression of lacZ in the oocytes and the production of blue-colored b-galactosidase staining. (B–D) Labeling of only primordial follicles in adult mouse ovaries. T2 Three-month-old Sohlh1-CreER ;R26R females were given intraperitoneal injections of tamoxifen for 3 consecutive days. (B) The labeled primordial folli- cles (shown by the blue dots in the oocytes) can be seen 5 days after the injection. Arrows indicate the unlabeled growing follicles. (C) Labeled primordial follicles are still seen in ovaries 7 weeks after tamoxifen injection. (D) Labeled primordial follicles are still seen in the ovaries 14 months after tamoxifen injection. were sacrificed between 5 and 57 days after the first tamoxifen injection at 3 months of age. Tracing the development of adult primordial follicles to the primary stage We first studied the transition from primordial to primary follicles. As shown in Figure 4A, labeled primary follicles were first observed in one out of four mice (25%) sacrificed 7 days after the first tamoxifen injection. Two out of four (50%) mice showed labeled primary follicles 8 days after tamoxifen injection, and from 9 days onward, labeled primary follicles were seen in all sacrificed mice (100%). We conclude, therefore, that it takes 7–9 Figure 4. Tracing the development of primordial follicles in adult Sohlh1- days for primordial follicles to develop into primary stage follicles. T2 T2 CreER ;R26R mice. Sohlh1-CreER ;R26R adult females were injected with The fact that labeled primary follicles were always observed tamoxifen and sacrificed at different time points. The Y-axis is the percentage in all 124 mice sacrificed from 9 to 57 days after tamoxifen injec- of mice showing labeled follicles. The numbers of mice used for each time point are shown above the bars. Representative labeled follicles are shown in tion (Fig. 4A) indicated that the activation of primordial follicles the right panels. (A) A minimum of 7 days is needed for labeled primordial fol- might take longer than 9 days and that such activation occurs in a licles to reach the primary stage. Labeled primary follicles can be found in all progressive manner as previously hypothesized (3,15). mice sacrificed from 9 to 57 days after tamoxifen injection. (B) A minimum of 23 days is needed for labeled primordial follicles to develop into secondary fol- licles. (C) A minimum of 37 days is needed for labeled primordial follicles to Tracing of adult primordial follicles to the secondary stage develop into the early antral stage. (D) A minimum of 47 days is needed for We next looked into how the primordial follicles reached the sec- labeled primordial follicles to develop into the antral stage. ondary stage. As shown in Figure 4B, labeled secondary follicles 24 days after tamoxifen injection, and in almost all mice sacri- were seen in one out of five (20%) mice sacrificed 23 days after ficed from 25 to 57 days after tamoxifen injection. These tamoxifen injection, four out of five (80%) mice sacrificed Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 925 results indicated that a minimum of 23–24 days was needed for the primordial follicles to reach the secondary stage. If 7 days are assumed to be the minimum time needed to develop from primordial to primary follicles, the shortest dwell- ing time of primary follicles in mouse ovaries is 16 days. Tracing of adult primordial follicles to the early antral stage As shown in Figure 4C, labeled early antral follicles were found in three out of four (75%) mice sacrificed 37 days after tamoxifen injection, 11 out of 15 (73%) mice sacrificed from 38 to 43 days after tamoxifen injection, and in all mice sacrificed from 44 to 57 days after tamoxifen injection. Thus, the development from primordial follicles to the early antral stage takes at least 37 days. If a minimum of 23 days is assumed for primordial follicles to reach the secondary stage, the time for secondary follicles to develop into early antral follicles in mouse ovaries is at least 14 days. Tracing of adult primordial follicles to the antral stage No labeled antral follicles were seen in 87 mice sacrificed from 5 to 45 days after tamoxifen injection or in 7 mice sacrificed 46 days after tamoxifen injection (Fig. 4D). After that, a stable pro- Figure 5. Labeled primordial follicles persisted to the end of the reproductive T2 T2 portion of mice (30%–50%) sacrificed from 47 to 57 days after lifespan in Sohlh1-CreER ;R26R mouse ovaries. Sohlh1-CreER ;R26R adult females were given tamoxifen and sacrificed 10 months later. (A) A representa- tamoxifen injection contained labeled antral follicles (Fig. 4D). tive labeled primary follicle with a labeled primordial follicle in the same field. Thus, it takes a minimum of 47 days for a primordial follicle to go (B) A representative labeled secondary follicle with a labeled primordial follicle through the different developmental stages and become an antral in the same field. (C) A representative labeled early antral follicle. (D) A repre- follicle in adult mouse ovaries. sentative labeled antral follicle. The minimum dwelling time for an early antral follicle can be calculated as 10 days because it takes at least 37 days to develop mice over a period of 2 to 14 months after tamoxifen injection. from primordial to early antral follicles. Labeled primordial, primary, secondary and early antral follicles were always observed in all of the mice sacrificed at various Comparison of the developmental dynamics of the postnatal times during this period (Fig. 3D and Fig. 5A–C). In addition, and adult primordial follicles labeled antral follicles were found in mice sacrificed 10 The developmental dynamics of adult primordial follicles in the months after tamoxifen injection (13 months old) (Fig. 5D). ovarian cortex are distinct from those of the first wave of acti- This result indicated that the primordial follicles labeled in vated primordial follicles in the ovarian medulla. We have sum- early adulthood were competent for activation after a long marized the age-dependent minimal times needed for their period of dormancy and could contribute to ovulation. This respective development in Table 1. finding is in accordance with the notion that a fixed pool of prim- ordial follicles serves as the only source of developing follicles throughout the reproductive lifespan of mice (16–18). Long-term tracing of primordial follicles To trace the development of primordial follicles throughout the entire mouse reproductive lifespan, we labeled the adult cortical DISCUSSION primordial follicles by administering tamoxifen to 17 Sohlh1- T2 In the mammalian ovary, progressive activation of primordial CreER ;R26R mice at 3 months of age and sacrificed the follicles serves as the source of fertilizable ova. However, very little is known about the developmental dynamics of primordial follicles under physiological conditions. In this study, we have Table 1. The distinct age-dependent minimal developmental times of the two T2 classes of primordial follicles generated tamoxifen-inducible Foxl2-CreER and Sohlh1- T2 CreER knock-in mouse models to specifically label the first wave of postnatally activated primordial follicles and the adult Primordial Primordial to Primordial to Primordial to primary secondary early antral to antral cortical primordial follicles, respectively. We showed that the first wave of activated primordial follicles First wave of Right after 8 days 13 days 23 days remain in the mouse ovaries for 3 months after birth and con- activated formation tribute to the onset of puberty and to fertility during early adult- primordial follicles hood. During this period, primordial follicles that are formed Adult cortical 7–9 days 23–24 days 37 days 47 days after birth and located in the ovarian cortex gradually replace primordial the first wave of follicles until they become the sole source of follicles follicles from 3 months until the end of the reproductive life. Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 926 Human Molecular Genetics, 2014, Vol. 23, No. 4 T2 We have also determined the age-dependent time periods Generation of a Foxl2-CreER knock-in mouse model to required for the two classes of primordial follicles to develop label the first postnatal wave of activated follicles into advanced stages and have shown that the developmental dy- The mouse Foxl2 gene contains one exon, and we aimed to keep namics were completely different for the two types of primordial the original open reading frame of Foxl2 intact. An 8 kb region follicles (Table 1). Moreover, our follicular tracing data showed used to construct the targeting vector was first sub-cloned from a that primordial follicles that are labeled early in life can persist in C57BL/6 BAC clone that contained the full-length Foxl2 allele the ovaries until the end of reproductive life and that these folli- (RP23: 142D12). As shown in Supplementary Material, Figure cles were competent to develop into antral stage. S1A, the long arm extended 5.5 kb from the 5 end to the site The visual in vivo tracing of follicles performed in this work of the cassette insertion after the coding region of Foxl2. The provides a precise picture of long-term follicular development short arm extended 2.5 kb from the site of the cassette insertion starting from the primordial stage. Such accurate tracing has T2 to the 3 end. An IRES-CreER -FRT-Neo-FRT cassette was not been possible by autoradiographic labeling of follicles inserted at 259 bp downstream of the STOP codon of Foxl2 (19,20). We believe that our results provide a relatively conclu- using Red/ET recombineering technology (Gene Bridges, sive picture of how the two distinct classes of primordial follicles Germany). develop under normal physiological conditions. Such an under- The targeting constructwas sub-cloned into a backbone vector standing is valuable for studying ovarian organogenesis and pSP72 (Promega, US) containing an ampicillin selection cas- physiology as well as ovarian diseases that are related to sette for re-transformation of the construct prior to electropor- altered follicular development. ation. Ten micrograms of the targeting vector were linearized It has been generally believed that the first wave of activated by NotI and then transfected by electroporation of BA1 follicles are anovulatory (5,7) even though oocytes from the (C57BL/6 × 129/SvEv) (Hybrid) embryonic stem (ES) cells. first wave of follicles can mature in vitro and generate live After selection with G418 antibiotic, surviving clones were pups (21,22). Our data clearly show, however, that the first selected by PCR and Southern blot analysis following standard wave of follicles plays an active role in providing mature eggs procedures (Supplementary Material, Fig. S1B). The positive until 3 months of age. Thus, it is likely that the fast-growing clone was microinjected into C57BL/6 blastocysts following first wave of follicles facilitates the establishment of the hypo- standard procedures. Resulting chimeric mice with a high per- thalamic–pituitary–ovarian axis and thereby plays a key role centage agouti coat color were mated to wild-type C57BL/6 in the onset of puberty and the initiation of reproductive life. mice to generate F1 heterozygous offspring. The heterozygous Given the fact that the growth pattern of the first wave of acti- mice were healthy and fertile. The neomycin expression cassette vated follicles is conserved among mammals (including T2 was excised in vivo by crossing F1 Foxl2-CreER mice with humans) (23,24), it can be postulated that the fertility of transgenic mice expressing flippase (FLP) (003800, Jackson La- women from puberty onset to young adulthood might rely on T2 boratory, US). Foxl2-CreER males were mated with mT/mG the first wave of follicles that are already activated at the fetal females (007576, Jackson Laboratory) to generate Foxl2- stage. In contrast to this first wave of activated follicles, we T2 CreER ;mT/mG mice. have shown that the relatively slow-growing adult primordial follicles contribute to continuous ovulation throughout the middle and late stages of reproductive life, and the gradual but T2 Generation of Sohlh1-CreER knock-in mice to label steady loss of the cortical primordial follicles leads to reproduct- primordial follicles in adulthood ive aging. It is possible that the developmental dynamics of adult primordial follicles in mice might be similar to adult primordial A 13 kb region used to construct the targeting vector was first follicle development in humans. sub-cloned from a positively identified C57BL/6 (RP23: 57O11) Taken together, this study presents several lines of in vivo evi- BAC clone that contained the full-length Sohlh1 allele. The dence that clarify the distinct developmental dynamics of two region was designed such that the short arm extended 2.3 kb 5 classes of primordial follicles under physiological conditions to the ATG of exon 1. The long arm was located on the 3 side T2 and provides a solid foundation for studying how primordial fol- of exon 8 and was 6.5 kb long. The CreER -FRT-Neo-FRT cas- licles develop in humans. sette replaced 4.2 kb of the gene, including exons 1–8 at the ATG site (Supplementary Material, Fig. S4A). The targeting construct was sub-cloned into a 2.4 kb pSP72 MATERIALS AND METHODS backbone vector. Ten micrograms of targeting vector were line- arized by SwaI. The electroporation, ES cell characterization Reagents (Supplementary Material, Fig. S4B), blastocyst injection and Restriction enzymes were purchased from New England neomycin cassette removal were performed with standard proce- Biolabs, UK. Antibiotics were purchased from Invitrogen, US. dures as described earlier. The heterozygous mice were found to T2 Tamoxifen and corn oil were purchased from Sigma-Aldrich, be healthy and fertile. Sohlh1-CreER males were mated with Germany. Reagents for b-galactosidase staining [paraformalde- R26R females (003474, Jackson Laboratory) to generate the T2 hyde, magnesium chloride, ethylene glycol tetraacetic acid, Sohlh1-CreER ;R26R mice. sodium deoxycholate, NP40, potassium ferricyanide, potassium All mice were crossed with C57BL/6 mice (Charles River, ferrocyanide, 5-Bromo-4-chloro-3-indolyl b-D-galactoside US) for six generations to obtain identical genetic backgrounds. (X-gal) and 0.1% nuclear fast red solution] were all purchased Mice were housed in controlled environmental conditions with from Sigma–Aldrich. Paraffin and hematoxylin were purchased free access to water and food. Illumination was on between from Histolab, Sweden. 0600 and 1800. Experimental protocols were approved by the Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 Human Molecular Genetics, 2014, Vol. 23, No. 4 927 regional ethical committee of the University of Gothenburg, mice. Briefly, ovaries were fixed in 4% paraformaldehyde for 1 h Sweden. at 48C and then rinsed 3 times for 15 min each in a buffer consist- ing of 1 × PBS (pH 7.4), 2 mM magnesium chloride, 5 mM ethylene glycol tetraacetic acid, 0.01% sodium deoxycholate T2 Tamoxifen administration in Foxl2-CreER ;mT/mG mice and 0.02% NP40 at room temperature. The ovaries were then incubated with a staining solution consisting of 1 × PBS (pH Tamoxifen was resuspended at 100 mg/mL in 95% ethanol and 7.4), 2 mM magnesium chloride, 0.01% sodium deoxycholate, further diluted with corn oil to a final concentration of 20 mg/mL 0.02% NP40, 5 mM potassium ferricyanide, 5 mM potassium (25). For embryonic injection, mT/mG females were first T2 ferrocyanide and 1 mg/mL X-gal at 378C overnight. plugged by Foxl2-CreER males. A single intraperitoneal in- Following staining, the ovaries were re-fixed in 4% parafor- jection of tamoxifen (15 mg/kg BW, 600 mg per mouse) was maldehyde for 8 h, embedded in paraffin, serially cut into given to pregnant females at E16.5 (the day the vaginal plug 8-mm sections and counterstained with 0.1% nuclear fast red so- appeared was counted as E0.5). Because the tamoxifen injection lution. The sections were examined under a Zeiss Axio Scope A1 compromised the ability of the mice to have spontaneous vaginal upright microscope. A follicle was considered to be labeled if a delivery, pups were delivered by Cesarean section at E20. Surro- blue dot was observed in the cytoplasm of the oocyte. gate mothers were used to nurture the pups. For postnatal injec- T2 tion, Foxl2-CreER ;mT/mG females were given a single intraperitoneal injection of tamoxifen at PD5 at a dose of Quantification of ovarian follicles for evaluating the 15 mg/kg BW. influence of tamoxifen on follicular development For quantification of ovarian follicles and the morphological Histological analysis and quantification of ovarian studies shown in Supplementary Material, Figs S3 and S6, T2 follicles from Foxl2-CreER ;mT/mG mice ovaries were fixed in 4% paraformaldehyde, dehydrated and em- T2 bedded in paraffin. Paraffin-embedded ovaries were serially cut To observe the ovarian morphology of Foxl2-CreER ;mT/mG into 8-mm sections and stained with hematoxylin for morpho- mice, ovaries were fixed in 4% paraformaldehyde, dehydrated logical observation. Ovarian follicles at different stages of devel- and embedded in paraffin. The paraffin-embedded ovaries opment, including primordial (type 2), primary (type 3), were serially cut into 8-mm sections and rehydrated. The fluores- secondary (type 4 and 5), early antral (type 6) and antral (type cent images were taken with a Zeiss Axio Scope A1 upright 7) follicles were counted in all sections of an ovary based on microscope installed with filter sets for mT (tdTomato, 554/ the well-accepted standards established by Pedersen and Peters 581 nm) and mG (EGFP, 488/507 nm) and merged with the (14). Follicles that contained oocytes with clearly visible Zeiss AxioVision software. nuclei were scored in each section, as previously reported (28). For quantification of follicles, the fluorescent images from Judged from careful morphological analysis, the incidence of every five sections per ovary were used to count the follicles. counting the same follicle twice or of missing a follicle was low. The classification of follicular development was based on well- established standards by Pedersen and Peters (14). The numbers of growing follicles with green (designated as labeled) or red Statistical analysis (designated as unlabeled) granulosa cells in each image were All experiments were repeated at least 3 times. In Supplementary counted. Material, Figures S3 and S6, the differences between the numbers of ovarian follicles in tamoxifen- and vehicle-injected T2 mice were calculated with Student’s t-test, and the difference Tamoxifen administration in Sohlh1-CreER ;R26R mice was considered to be significant if P , 0.05. To reach the optimal labeling efficiency, 3-month-old adult T2 Sohlh1-CreER ;R26R female mice were given an intraperito- neal injection of 80 mg/kg BW (2 mg per mouse) tamoxifen SUPPLEMENTARY MATERIAL T2 per day for three consecutive days. The CreER -mediated Supplementary Material is available at HMG online. DNA recombination can take place within 24 h after tamoxifen injection (26,27), so the day of the first tamoxifen injection was counted as day 0. The labeled primordial follicles can be clearly Conflict of Interest statement. None declared. visualized by b-galactosidase staining 5 days after the first tam- oxifen injection. For tracing the follicular development, 140 mice were sacrificed between 5 and 57 days after the first tamoxi- FUNDING fen injection. For long-term tracing of primordial follicles, 17 mice were sacrificed between 2 to 14 months after the first tam- This work was supported by grants (to K. L.) from the Young Re- oxifen injection. searcher Award (Umea˚ University), the Jane and Dan Olssons Foundation, the LUA/ALF-medel Va¨stra Go¨talandsregionen, AFA Insurance, the Swedish Research Council, the Swedish T2 b-galactosidase staining of Sohlh1-CreER ;R26R ovaries Cancer Foundation, the Faculty of Natural Science of the Uni- Whole-mountb-galactosidase staining was performed to visual- versity of Gothenburg and the Novo Nordisk Foundation T2 ize the labeled follicles in the ovaries of Sohlh1-CreER ;R26R (Denmark). Funding to pay the Open Access publication Downloaded from https://academic.oup.com/hmg/article/23/4/920/635594 by DeepDyve user on 16 July 2022 928 Human Molecular Genetics, 2014, Vol. 23, No. 4 14. Pedersen, T.and Peters, H. (1968)Proposalfor a classification of oocytesand charges for this article was provided by the Swedish Research follicles in the mouse ovary. J. Reprod. Fertil., 17, 555–557. Council. 15. McGee, E.A. and Hsueh, A.J. (2000) Initial and cyclic recruitment of ovarian follicles. Endocr. Rev., 21, 200–214. 16. Zuckerman, S. (1951) The number of oocytes in the mature ovary. Recent REFERENCES Prog. Horm. Res., 6, 63–109. 1. Eppig, J.J., Bivens, C.M., Viveiros, M.M. and de la Fuente, R. (2003) 17. Lei, L. and Spradling, A.C. (2013) Female mice lack adult germ-line stem Regulation of mammalian oocyte maturation. In Peter, C.K.L. and Eli, Y.A. cells but sustain oogenesis using stable primordial follicles. Proc. Natl. (eds), The Ovary, 2nd edn. Academic Press, San Diego, USA, pp. 113–129. Acad. Sci. USA, 110, 8585–8590. 2. Matzuk, M.M., Burns, K.H., Viveiros, M.M. and Eppig, J.J. (2002) 18. Zhang, H., Zheng, W., Shen, Y., Adhikari, D., Ueno, H. and Liu, K. (2012) Intercellular communication in the mammalian ovary: oocytes carry the Experimental evidence showing that no mitotically active female germline conversation. Science, 296, 2178–2180. progenitors exist in postnatal mouse ovaries. Proc. Natl. Acad. Sci. USA, 3. Adhikari, D. and Liu, K. (2009) Molecular mechanisms underlying the 109, 12580–12585. activation of mammalian primordial follicles. Endocr. Rev., 30, 438–464. 19. Oakberg, E.F. (1979) Timing of oocyte maturation in the mouse and its 4. Reddy, P., Zheng, W. and Liu, K. (2010) Mechanisms maintaining the relevance to radiation-induced cell killing and mutational sensitivity. Mutat. dormancy and survival of mammalian primordial follicles. Trends Res., 59, 39–48. Endocrinol. Metab., 21, 96–103. 20. Pedersen, T. (1970) Follicle kinetics in the ovary of the cyclic mouse. Acta. 5. Hirshfield, A.N. (1992) Heterogeneity of cell populations that contribute to Endocrinol. (Copenh), 64, 304–323. the formation of primordial follicles in rats. Biol. Reprod., 47, 466–472. 21. Eppig, J.J., O’Brien, M.J., Wigglesworth, K., Nicholson, A., Zhang, W. and 6. Mork, L., Maatouk, D.M., McMahon, J.A., Guo, J.J., Zhang, P., McMahon, King, B.A. (2009) Effect of in vitro maturation of mouse oocytes on the A.P. and Capel,B. (2012) Temporal differences in granulosa cell specification health and lifespan of adult offspring. Hum. Reprod., 24, 922–928. in the ovary reflect distinct follicle fates in mice. Biol. Reprod., 86,37. 22. O’Brien, M.J., Pendola, J.K. and Eppig, J.J. (2003) A revised protocol for in 7. Eppig, J.J. and Handel, M.A. (2012) Origins of granulosa cells clarified and vitro development of mouse oocytes from primordial follicles dramatically complexified by waves. Biol. Reprod., 86, 34. improves their developmental competence. Biol. Reprod., 68, 1682–1686. 8. Hirshfield, A.N. and DeSanti, A.M. (1995) Patterns of ovarian cell 23. Peters, H., Byskov, A.G., Himelstein-Braw, R. and Faber, M. (1975) proliferation in rats during the embryonic period and the first three weeks Follicular growth: the basic event in the mouse and human ovary. J. Reprod. postpartum. Biol. Reprod., 53, 1208–1221. Fertil., 45, 559–566. 9. Peters, H. (1969) The development of the mouse ovary from birth to 24. Lintern-Moore, S., Peters, H., Moore, G.P. and Faber, M. (1974) Follicular maturity. Acta. Endocrinol. (Copenh), 62, 98–116. development in the infant human ovary. J. Reprod. Fertil., 39, 53–64. 10. Schmidt,D.,Ovitt, C.E., Anlag, K.,Fehsenfeld,S., Gredsted, L., Treier, A.C. 25. John, G.B., Gallardo, T.D., Shirley, L.J. and Castrillon, D.H. (2008) Foxo3 is and Treier, M. (2004) The murine winged-helix transcription factor Foxl2 is a PI3K-dependent molecular switch controlling the initiation of oocyte required for granulosa cell differentiation and ovary maintenance. growth. Dev. Biol., 321, 197–204. Development, 131, 933–942. 26. Hayashi, S. and McMahon, A.P. (2002) Efficient recombination in diverse 11. Uda, M., Ottolenghi, C., Crisponi, L., Garcia, J.E., Deiana, M., Kimber, W., tissues by a tamoxifen-inducible form of Cre: a tool for temporally regulated Forabosco, A., Cao, A., Schlessinger, D. and Pilia, G. (2004) Foxl2 gene activation/inactivation in the mouse. Dev. Biol., 244, 305–318. disruption causes mouse ovarian failure by pervasive blockage of follicle 27. Feil, R., Brocard, J., Mascrez, B., LeMeur, M., Metzger, D. and Chambon, P. development. Hum. Mol. Genet., 13, 1171–1181. (1996) Ligand-activated site-specific recombination in mice. Proc. Natl. 12. Muzumdar, M.D., Tasic, B., Miyamichi, K., Li, L. and Luo, L. (2007) A Acad. Sci. USA, 93, 10887–10890. global double-fluorescent Cre reporter mouse. Genesis, 45, 593–605. 13. Pangas, S.A., Choi, Y., Ballow, D.J., Zhao, Y., Westphal, H., Matzuk, M.M. 28. Reddy, P., Liu, L., Adhikari, D., Jagarlamudi, K., Rajareddy, S., Shen, Y., and Rajkovic, A. (2006) Oogenesis requires germ cell-specific Du, C., Tang, W., Ha¨ma¨la¨inen, T., Peng, S.L. et al. (2008) Oocyte-specific transcriptional regulators Sohlh1 and Lhx8. Proc. Natl. Acad. Sci. USA, 103, deletion of Pten causes premature activation of the primordial follicle pool. 8090–8095. Science, 319, 611–613.

Journal

Human Molecular GeneticsOxford University Press

Published: Feb 15, 2014

Keywords: hair follicle; mice; primordial follicles; foxl2 gene; tamoxifen

There are no references for this article.