Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2002 Mar 5; 99(5): 2890–2894.
Published online 2002 Feb 26. doi:  10.1073/pnas.052658699
PMCID: PMC122443
Developmental Biology

The mammalian oocyte orchestrates the rate of ovarian follicular development


The development of both the mammalian oocyte and the somatic cell compartments of the ovarian follicle is highly coordinated; this coordination ensures that the ovulated oocyte is ready to undergo fertilization and subsequent embryogenesis. Disruption of this synchrony results in oocyte developmental failure. Communication between the oocyte and companion somatic cells is essential for successful development of both follicular compartments. However, it was not previously known whether one cell type, either the somatic or the germ cell compartment, determines the overall rate of follicular development. To test the hypothesis that the oocyte orchestrates the rate of follicle development, mid-sized oocytes isolated from secondary follicles were transferred back to primordial follicles, the earliest stage of follicular development. This transfer doubled the rate of follicular development and the differentiation of follicular somatic cells. Oocyte development in these accelerated follicles appeared normal; recovered oocytes were competent to undergo fertilization and embryonic development. These results demonstrate that oocytes orchestrate and coordinate the development of mammalian ovarian follicles and that the rate of follicular development is based on a developmental program intrinsic to the oocyte.

Complex cell-to-cell interactions coordinate the development of ovarian follicles. The pathways of cellular communication include endocrine, autocrine, and paracrine regulators, and gap junctions. Coordination of the development of oocyte and somatic follicular compartments ensures that the ovulated oocyte is ready to undergo fertilization and subsequent embryogenesis. Disruption of this synchrony by inappropriately timed administration of exogenous gonadotropins can produce oocyte developmental failure (1). Communication between the oocyte and companion somatic cells is essential for successful development of both follicular compartments (2). The oocyte depends on its association with companion somatic granulosa cells to support its growth and development and to regulate the progression of meiosis. Likewise, oocytes promote granulosa cell proliferation, differentiation, and function. The communication between granulosa cells and oocytes is, therefore, bidirectional and occurs throughout follicular development (2, 3). In fact, follicular formation itself appears coordinated by a transcription factor, factor in the germline α (FIGα), expressed by the oocyte (4). Early follicular development depends on oocyte-secreted members of the transforming growth factor β family, growth differentiation factor (GDF)-9, and bone morphogenic protein (BMP)-15 (57). These oocyte-derived paracrine factors also promote follicular somatic cell proliferation and steroidogenesis (810) and locally regulate gene expression in granulosa cells (1013). Thus, communication between the oocyte and companion somatic cells is crucial for the development of both cell types, but how this complex interaction is coordinated was not previously known.

At birth, mouse ovaries contain only primordial follicles. A cohort of these earliest follicles begins development shortly after birth, and within 10–12 days reaches the secondary follicle stage where the oocyte is in mid-growth stage and is surrounded by two layers of granulosa cells. Follicles develop to the large antral stage containing fully grown oocytes by 18–24 days after birth (Fig. (Fig.11a). To test the hypothesis that oocytes control the rate of follicular development, oocytes isolated from the secondary follicles of 12-day-old mice were combined with the somatic cells of newborn mouse ovaries to produce reaggregated ovaries (hereafter referred to as 12/0 ovaries). Primordial oocytes from newborn ovaries were reaggregated with the somatic cells of ovaries from mice of the same age as controls (hereafter referred to as 0/0 ovaries).

Figure 1
Diagrammatic representation of ovarian follicular development in the mouse. The normal progression of follicular development in neonatal mice (a) from the primordial to the secondary stage requires 10–12 days; development to the large antral follicle ...

Materials and Methods

Reaggregated Ovaries.

The 12/0 ovaries were prepared by isolation of oocytes from the ovaries of 12-day-old mice by using collagenase and DNase in calcium- and magnesium-free PBS as described (14). The zonae pellucidae were removed by brief treatment with acid Tyrode's solution. Newborn ovaries were dissociated to single cells and the somatic cells separated from the germ cells by differential adhesion as described in detail (15). Oocytes, from either 12-day-old or newborn (C57BL/6J × SJL/J)F1 mice, were mixed with the somatic cells of newborn mice in a microcentrifuge tube (0.5 ml capacity; USA/Scientific, Ocala, FL; catalog no. 1405-0000) with 200 μl of culture medium (M199 supplemented with 10% FBS). Phytohemagglutinin was added to a final concentration of 35 μg/ml to facilitate cell-to-cell adhesion, and the mixture of cells was incubated at 37°C for 10 min. Samples were then subjected to centrifugation in an Eppendorf 5415C centrifuge at 10,000 rpm for approximately 10 s. The tubes were then rotated 180° and centrifuged again for approximately 30 s. This double centrifugation protocol promoted a more even distribution of oocytes among the somatic cells. Pellets of reaggregated cells were gently removed from the tubes and “organ” cultured overnight as described (15). The reaggregated ovaries were surgically implanted beneath the renal capsules of bilaterally ovariectomized host females, ≈3 months old, of the same genetic background.

Histology and in Situ Hybridization.

Ovarian samples were prepared for histological examination by fixation for 3–5 h in 2.5% glutaraldehyde and 2.5% paraformaldehyde in 0.083 M sodium cacodylate buffer, pH 7.2, at 4°C. After washing for 24 h in 0.1 M sodium cacodylate buffer, the samples were embedded in JB4 (glycol methacrylate) plastic, and 2-μm sections were stained with periodic acid/Schiff reagent and hematoxylin. In situ hybridization was carried out by using the protocols described by Manova et al. (16). Ovaries were fixed in 4% paraformaldehyde overnight, then washed, dehydrated, and embedded in paraffin wax. Sections were cut at 4-μm thickness and fixed again with 4% paraformaldehyde. The samples were treated with proteinase K, which had been acetylated with 0.25% acetic anhydride in 0.1 M triethanolamine. Sense and antisense probes for luteinizing hormone receptor (LHR) mRNA expression were prepared as described (11). The probes incorporating [α-33P]CTP (NEN Life Science Products) were made with SP6 and T7 RNA polymerases, respectively, by using MAXIscript kits (Ambion, Austin, TX). After probe preparation, slides were hybridized overnight at 65°C and washed after a 30-min ribonuclease (RNase) treatment at 37°C (1:40 dilution of RNase mixture; Ambion). Washing steps included immersion in 50% formamide/2× standard saline citrate (SSC) at 65°C for 20 min and immersion in 0.1× SSC at room temperature for 1 h. After washing, slides were dipped in NTB2 emulsion (Kodak, New Haven, CT) and exposed 3–4 days before being developed and stained with hematoxylin and eosin.

Oocyte Maturation, Fertilization, and Preimplantation Embryo Culture.

Protocols for oocyte maturation, fertilization, and preimplantation embryo development in vitro were carried out exactly as described (17). Briefly, oocyte–cumulus cell complexes were isolated from ovaries by puncturing the large antral follicles with a 26-gauge syringe needle. Cumulus cell-enclosed, germinal vesicle-stage oocytes were matured for 15 h in Waymouth MB752/1 medium supplemented with 0.23 mM sodium pyruvate, 5% FBS, and 0.5 international units of follicle-stimulating hormone (FSH; human recombinant FSH obtained from the National Hormone and Peptide Program of the National Institute of Diabetes and Digestive and Kidney Diseases) at 37°C by using an atmosphere of 5% O2/5% CO2/90% N2, hereafter referred to as 5–5–90 gas. Fertilization was carried out in 0.5-ml drops of fertilization medium (MEM prepared with Earle's balanced salt solution, essential and nonessential amino acids, 3 mg/ml BSA, and 0.23 mM sodium pyruvate) under washed paraffin oil. Eggs were inseminated for 4 h at 37°C under 5–5–90 gas. After insemination, eggs were removed from fertilization drops, washed, and cultured in1 ml of fertilization medium, gassed with 5–5–90 gas for 24 h. After washing, 2-cell stage embryos were cultured in KSOM/AA (18) for 4 more days at 37°C under 5–5–90 gas to produce blastocysts.


Three days after grafting of 12/0 ovaries, the oocytes were surrounded by one or two layers of granulosa cells typical of the transition from primary to secondary follicles (Fig. (Fig.22a). Nine days after grafting, the 12/0 ovaries contained mostly large antral follicles that included prominent layers of theca cells (Fig. (Fig.22b). In contrast, at 9 days after grafting, control 0/0 ovaries did not contain follicles advanced beyond the secondary follicle stage (Fig. (Fig.22c). Large antral-stage follicles were not evident in 0/0 ovaries until 19–20 days after grafting (15). Morphological aspects of follicular development were, therefore, dramatically accelerated in the 12/0 ovaries (Fig. (Fig.11b), whereas follicles in 0/0 ovaries developed at essentially normal rates.

Figure 2
Follicular development in reaggregated ovaries. (a) 12/0 ovary 3 days after grafting to the renal bursa. Note that ≈1–2 layers of granulosa cells surround the oocyte. (b) 12/0 ovary 9 days after grafting. Three large antral ...

During normal follicular development, granulosa cells in large antral follicles differ in their pattern of gene expression and function according to their location within the follicle. The granulosa cells lining the follicular wall are referred to as mural granulosa cells, and those encompassing the oocyte are called cumulus granulosa cells. Mural granulosa cells in large antral follicles express receptors for luteinizing hormone (LH), but in mice the cumulus granulosa cells do not. This same pattern of LHR mRNA expression was observed by in situ hybridization in all of the 12/0 ovaries 9 days after grafting (Fig. (Fig.33 a–d), suggesting that the pattern of granulosa cell differentiation in large antral follicles reflect normal patterns despite accelerated development.

Figure 3
Differentiation of mural granulosa cells in 12/0 follicles. Expression of LHR mRNA shown by in situ hybridization. (a and b) Normal in vivo-grown follicles in ovary of 22-day-old mouse 44 h after stimulating follicular development by administration ...

To further test whether granulosa cell function in large 12/0 follicles is normal, the function of the cumulus cells associated with the oocyte was assessed. After the preovulatory LH surge, cumulus cells, but not mural granulosa cells, undergo a process referred to as cumulus expansion, or mucification. Cumulus expansion requires the synthesis and secretion of hyaluronic acid by the cumulus cells. Hyaluronic acid synthesis and cumulus expansion can be stimulated in vitro by treatment with FSH (19, 20). Granulosa cells of secondary or very early antral follicles do not undergo expansion when stimulated by FSH (21). Cumulus expansion, however, did occur when the oocyte-cumulus cell complexes were isolated from the large antral follicles of 12/0 ovaries just 9 days after grafting (Fig. (Fig.44 a and b). Therefore, in addition to accelerated morphological aspects of follicular development, the granulosa cells of 12/0 ovaries exhibit an accelerated pattern of differentiation and function.

Figure 4
Functional differentiation of the cumulus oophorus. Oocyte–cumulus complexes isolated from 12/0 follicle 9 days after grafting and cultured in control medium (a) or medium containing FSH (b; 0.5 international units/ml human recombinant ...

The function of oocytes that developed in accelerated follicles was also tested. Oocyte–cumulus cell complexes were isolated from large antral follicles 9 days after grafting and cultured to determine whether they had developed competence to resume and complete meiosis and undergo fertilization and subsequent preimplantation development. As shown in Fig. Fig.5,5, 68% of the oocytes isolated from 12/0 ovaries 9 days after grafting were competent to resume meiosis, compared with >95% of the control oocytes. Similarly, 53% of the 12/0 oocytes matured in vitro underwent fertilization and cleaved to the 2-cell stage compared with about 70% of the control oocytes. The frequency of embryonic development from the 2-cell stage to the blastocyst stage was approximately 75% in all groups of oocytes (Fig. (Fig.5).5). Oocyte development was, therefore, slightly retarded in the 12/0 ovaries, but many oocytes competent to resume meiosis and undergo fertilization as well as complete preimplantation development were produced in these accelerated follicles.

Figure 5
Competence of oocytes to resume meiosis and undergo fertilization and preimplantation development. Oocyte–cumulus cell complexes were isolated from the large antral follicles of three groups of ovaries: 12/0 ovaries 9 days after grafting ...


The results presented here show that the rate of follicular development was approximately doubled by transfer of mid-growth stage oocytes to the somatic environment of primordial follicles. This accelerated follicular development includes precocious differentiation and function of both the mural and cumulus granulosa cells. We conclude that oocytes orchestrate and coordinate the development of mammalian ovarian follicles and that the rate of follicular development is essentially based on a developmental program intrinsic to the oocyte.

The reciprocal experiment, the transfer of an early stage of oocyte development to a more advanced follicular stage, is not feasible. The ovary of a 12-day-old mouse contains a range of follicles from the primordial to the secondary (preantral) stage of follicular development, thus the follicular origin of somatic cells produced by ovarian dissociation cannot be specified. Moreover, discrete primordial or primary follicles did not form when primordial oocytes were reaggregated with granulosa cells from secondary follicles. Rather, a single mass of granulosa cells containing scattered oocytes was produced (results not shown). Nevertheless, if the transfer of earlier stages of oocytes to latter stage follicles could have been achieved, the anticipated result would be a prolongation of follicular development.

Oocytes in secondary follicles are in mid-growth stage and are incompetent to resume meiosis or undergo fertilization and embryogenesis. In contrast, oocytes from large antral follicles are competent to resume meiosis, progress to metaphase II, and then to undergo fertilization and embryonic development. Therefore, growth and development of oocytes occurs in coordination with follicular development and depends on interaction with companion granulosa cells. Oocytes that developed in the 12/0 constructs for 9 days were competent to undergo maturation, fertilization, and embryonic development. However, the incidence of successful development was less than that for control oocytes. This difference is most likely caused by a lag in oocyte development for a few days while follicular formation and function was reestablished after ovarian reaggregation. Because more than half of the oocytes of 12/0 follicles completed preimplantation development, it may be inferred that granulosa cell function in the 12/0 follicles was eventually coordinated with oocyte development.

Clearly, many factors contribute to follicular development: gonadotropic hormones as well as inter- and intra-follicular signaling molecules. Moreover, it is obvious from years of experimental and clinical experience that administration of exogenous hormones can accelerate at least the final stages of follicular development. However, dramatic stimulatory protocols upset the synchrony of the development of the oocyte and somatic cell components of follicles and produce oocytes whose developmental potential is compromised (1, 22). The development of follicular components must, therefore, be tightly coordinated. The experiments described here indicate that the oocyte is the coordinator of an oocyte–granulosa cell regulatory loop whose function is essential for the normal development of both the oocyte itself and the entire follicle as a structural and functional unit. This finding may be a model for other systems involving complex interactions between different cell types in which one cell type may orchestrate and coordinate the formation of a developmental or functional physiological unit.

It seems remarkable that oocytes in mid-growth stage from secondary follicles have retained the ability to organize a primordial follicle. FIGα is an oocyte-specific transcription factor that probably functions to coordinate the expression of the three zona pellucida genes during oocyte growth (23). However, expression of Figla was detected in oocytes as early as embryonic day 13 (4), and targeted mutagenesis of Figla resulted in failure of primordial follicle formation. These results suggest that FIGα not only coordinates expression of zona pellucida genes in growing oocytes, but also regulates the expression of genes associated with early follicular assembly (4). The results presented here indicate that growing oocytes isolated from secondary follicles are able to organize early follicles. This finding suggests that gene products regulated by FIGα, which are required for early follicle formation, may still be produced by oocytes beyond the stage at which their function is actually required for follicle formation. It is also possible that isolation of growing oocytes from their companion granulosa cells up-regulates the FIGα-dependent pathway in oocytes required for follicle formation. Moreover, FIGα-dependent gene products, other than zona pellucida proteins, may be required during later follicular development and early follicle formation.

It is possible that the primary function of the oocyte in orchestrating the overall pace of follicular development occurs during preantral follicular development. The influence of the oocyte on the granulosa cells of preantral follicles could establish the program for antral follicular development when gonadotropic hormones can exert their most dramatic effects. During antral follicular development, the oocyte may predominantly influence the development and function of the granulosa cells most closely associated with it, the cumulus cells. Nevertheless, it is also possible that the influence of the oocyte on follicular development extends, either directly or indirectly, beyond the cumulus cells of antral follicles and affects even the most distant cells.

The mechanism by which the oocyte orchestrates follicular development awaits resolution, but over the last decade factors secreted by oocytes have been identified that probably play crucial roles, including GDF-9, BMP-15, and oocyte secreted protein (OOSP)-1. Mutation of the Gdf9 or Bmp15 gene has profound effects on follicular development, although the phenotype is species-specific (57, 24, 25). Null mutation of Gdf9, but not Bmp15, prevents development beyond the primary follicle stage in mice (5, 6, 25). Failure to produce BMP-15 produces a similar early follicular arrest in sheep ovaries (7). The function of OOSP1 is not yet known, although the expression pattern is essentially the same as Gdf9 and Bmp15 in mouse oocytes (26).

There are probably many paracrine regulatory factors, in addition to GDF-9, BMP-15, and OOSP1, secreted by oocytes that affect follicular development. Their identification and characterization will be crucial to resolving the mechanisms by which the oocyte orchestrates follicular development. These regulators may serve as novel candidate targets for fertility control, or they may facilitate the regulation of follicular development in vitro, to provide oocytes for clinical and agricultural applications, and to enable the rescue of endangered species.


We thank Drs. Mary Ann Handel, Tim O'Brien, John Schimenti, and Maria Viveiros for their helpful suggestions in the preparation of this paper. This research was supported by National Institutes of Child Health and Human Development Grant HD23839. We are also grateful to the National Hormone and Peptide Program of the National Institute of Diabetes and Digestive and Kidney Diseases for providing the FSH used in these studies.


bone morphogenic protein 15
growth differentiation factor 9
germinal vesicle
GV breakdown
follicle-stimulating hormone
factor in the germline α
luteinizing hormone
LH receptor
oocyte secreted protein 1
transforming growth factor β


This paper was submitted directly (Track II) to the PNAS office.


1. Hunter R H F, Cook B, Baker T G. Nature (London) 1976;260:156–157. [PubMed]
2. Eppig J J. BioEssays. 1991;13:569–574. [PubMed]
3. Eppig J J. Reproduction. 2001;122:829–838. [PubMed]
4. Soyal S M, Amleh A, Dean J. Development (Cambridge, UK) 2000;127:4645–4654. [PubMed]
5. Dong J W, Albertini D F, Nishimori K, Kumar T R, Lu N F, Matzuk M M. Nature (London) 1996;383:531–535. [PubMed]
6. Carabatsos M J, Elvin J, Matzuk M M, Albertini D F. Dev Biol. 1998;204:373–384. [PubMed]
7. Galloway S M, McNatty K P, Cambridge L M, Laitinen M P E, Juengel J L, Jokiranta T S, McLaren R J, Luiro K, Dodds K G, Montgomery G, et al. Nat Genet. 2000;25:279–283. [PubMed]
8. Hayashi M, McGee E A, Min G, Klein C, Rose U M, van Duin M, Hsueh A J. Endocrinology. 1999;140:1236–1244. [PubMed]
9. Solovyeva E V, Hayashi M, Margi K, Barkats C, Klein C, Amsterdam A, Hsueh A J W, Tsafriri A. Biol Reprod. 2000;63:1214–1218. [PubMed]
10. Elvin J A, Clark A T, Wang P, Wolfman N M, Matzuk M M. Mol Endocrinol. 1999;13:1035–1048. [PubMed]
11. Eppig J J, Wigglesworth K, Pendola F L, Hirao Y. Biol Reprod. 1997;56:976–984. [PubMed]
12. Joyce I M, Pendola F L, Wigglesworth K, Eppig J J. Dev Biol. 1999;214:342–353. [PubMed]
13. Joyce I M, Pendola F L, O'Brien M, Eppig J J. Endocrinology. 2001;142:3187–3197. [PubMed]
14. Eppig J J, O'Brien M J. Biol Reprod. 1996;54:197–207. [PubMed]
15. Eppig J J, Wigglesworth K, Hirao Y. Dev Biol. 2000;224:60–68. [PubMed]
16. Manova K, Nocka K, Besmer P, Bachvarova R F. Development (Cambridge, UK) 1990;110:1057–1069. [PubMed]
17. Eppig J J. In: A Comparative Methods Approach to the Study of Oocytes and Embryos. Richter J D, editor. Oxford: Oxford Univ. Press; 1999. pp. 3–9.
18. Ho Y, Wigglesworth K, Eppig J J, Schultz R M. Mol Reprod Dev. 1995;41:232–238. [PubMed]
19. Eppig J J. Nature (London) 1979;281:483–484. [PubMed]
20. Eppig J J. J Exp Zool. 1979;208:345–353. [PubMed]
21. Vanderhyden B C, Caron P J, Buccione R, Eppig J J. Dev Biol. 1990;140:307–317. [PubMed]
22. Eppig J J, Schroeder A C, O'Brien M J. J Reprod Fert. 1992;95:119–127. [PubMed]
23. Liang L, Soyal S M, Dean J. Development (Cambridge, UK) 1997;124:4939–4947. [PubMed]
24. Elvin J A, Yan C N, Wang P, Nishimori K, Matzuk M M. Mol Endo. 1999;13:1018–1034. [PubMed]
25. Yan C, Wang P, DeMayo J, Elvin J A, Carino C, Prasad S V, Skinner S S, Dunbar B S, Dube J L, Celeste A J, Matzuk M M. Mol Endocrinol. 2001;15:854–866. [PubMed]
26. Yan C, Pendola F L, Jacob R, Lau A L, Eppig J J, Matzuk M M. Genesis. 2001;31:105–110. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...