| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
College of Life Sciences,3 Peking University, Beijing 100871, China
Shenzhen Graduate School,4 Peking University, Shenzhen 518055, China
Beijing Laboratory Animal Research Center,5 Beijing 100083, China
Institute of Zoology,6 Chinese Academy of Science, Beijing 100871, China
ABSTRACT
Mature mouse oocytes currently can be generated in vitro from the primary oocytes of primordial follicles but not from premeiotic fetal germ cells. In this study we established a simple, efficient method that can be used to obtain mature oocytes from the premeiotic germ cells of a fetal mouse 12.5 days postcoitum (dpc). Mouse 12.5-dpc fetal ovaries were transplanted under the kidney capsule of recipient mice to initiate oocyte growth from the premeiotic germ cell stage, and they were recovered after 14 days. Subsequently, the primary and early secondary follicles generated in the ovarian grafts were isolated and cultured for 16 days in vitro. The mature oocytes ovulated from these follicles were able to fertilize in vitro to produce live offspring. We further show that the in vitro fertilization offspring were normal and able to successfully mate with both females and males, and the patterns of the methylated sites of the in vitro mature oocytes were similar to those of normal mice. This is the first report describing premeiotic fetal germ cells able to enter a second meiosis and support embryonic development to term by a combination of in vivo transplantation and in vitro culture. In addition, we have shown that the whole process of oogenesis, from premeiotic germ cells to germinal vesicle (GV)-stage oocytes, can be carried out under the kidney capsule.
developmental biology, fetal germ cells, follicular development, meiosis, mouse oogenesis, oocyte, ovary, primordial germ cell development
INTRODUCTION
As it is the basis of fertilization and zygote development, oogenesis is a very important process during development and has been studied extensively in the mouse, zebra fish, and Drosophila [1, 2]. In the mouse, undifferentiated germ cell precursors known as primordial germ cells (PGCs) start to colonize the gonadal ridge at approximately 10.5 days postcoitum (dpc), and they also begin to form oogonia in the ovary [3–7]. Oogonia, the diploid germ cells, can develop into haploid gametes after the completion of two meiotic divisions. Oogonia enter the first meiosis after 12.5
13 dpc, and most germ cells in the ovary have embarked on first meiosis by 14.5 dpc and then become arrested in the late diplotene stage around the time of birth [5, 8]. After birth, mouse ovarian follicles continue growing, and oocytes inside the follicles are able to mature into the metaphase in the second meiosis and become fertilized when the mouse becomes an adult.
Mouse oogenesis has been studied by means of in vitro culture technology over the last few years [9–13]. The development of primary oocytes from newborn mice has proven successful, and live offspring have been produced via in vitro fertilization using in vitro-matured oocytes [9, 11]. These studies have demonstrated that mouse primary oocytes in the primordial follicles of newborn mouse ovaries were able to enter the second meiosis and achieve complete development in vitro, although the frequency of preimplantation development was low. In vitro development of mouse oocytes from premeiotic fetal germ cells has been attempted, but the oocytes generated are immature and only undertake the first meiosis [4, 8, 10, 13–15]. Earlier studies by Chuma and Nakatsuji have shown that the in vitro-grown oocytes derived from 10.511.5 dpc fetal germ cells only undergo the first meiotic transition into the leptotene stage on feeder cells [8]. Klinger and de Felici [10] developed a multistep in vitro culture system for the development of mouse fetal oocytes. In their studies, 50 µm oocytes were obtained by direct stimulation with Kit ligand (KITL) or co-culture with granulosa cells, and the in vitro-grown oocytes were induced to resume meiosis by okadaic acid (OA) treatment but were unable to maintain chromosome condensation in the M phase. Recently, Obata et al. and Niwa et al. [12, 15] reported that mouse 12.5-dpc fetal ovaries could be cultured for 28 days in vitro, with the oocytes increasing in diameter to 63.9 µm. However, the oocytes that developed in vitro from premeiotic germ cells did not complete the second meiosis in vitro and did not develop to term.
In this study we developed a method to induce functional germ cells from 12.5-dpc fetal ovaries to differentiate and develop into mature oocytes by a combination of in vivo transplantation and in vitro culture, and we demonstrate that premeiotic 12.5-dpc fetal germ cells are able to enter the second meiosis and form mature oocytes in vitro that can be fertilized and developed to term.
MATERIALS AND METHODS
Transplantation Procedure of Fetal Ovaries
The Institutional Animal Care and Use Committees of Peking University approved all animal procedures. Fetal mouse ovaries of 12.5 dpc were obtained from pregnant CD-1 mice (Vital River, Beijing, China), as judged by the presence of a copulation plug (0.5 dpc).
Two to four 12.5-dpc fetal mouse ovaries without attached mesonephroses were transplanted under the kidney capsule of each 10- to 12-wk-old ovariectomized female CD-1 recipient mouse. Each recipient's own ovaries were removed at the top of the uterine horns.
Evaluation of Germ Cells Entering into Progression through Meiotic Prophase I
Germ cells were isolated from 12.5-dpc, 13.5-dpc, and 14.5-dpc CD-1 fetal mouse gonadal ridges following the EDTA puncturing method [16], collected directly in phosphate-buffered saline (PBS) medium and dispersed on the 4-well plate (BD Biosciences, Bedford, MA) coated with poly-L-lysine (Sigma, St. Louis, MO). To check whether the isolated germ cells entered into meiosis, the germ cells were stained with anti-DDX4 antibody (kindly provided by Dr. Toshiaki Noce, Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan) and anti-SYCP3 antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Cells were fixed and treated as described in Chuma and Nakatsuji [8]. Briefly, after fixation cells were incubated overnight at 4°C with 1:200 anti-SYCP3 antibody [13]. TRITC-conjugated anti-rabbit immunoglobulin G (IgG) was used as a secondary antibody. The meiotic prophase stages were determined by characteristic patterns of SYCP3 immunostaining within the germ cells as described in Di Carlo et al. and Mahadevaiah et al. [17, 18]. Undifferentiated germ cells in the ovaries at different stages of 12.514.5 dpc were analyzed using anti-POU5F1 antibody (Santa Cruz Biotechnology).
Isolation and Culture of Follicles In Vitro
Experimental group 1: 14 days after transplantation ovarian grafts were recovered from the transplanted sites. The primary and early secondary follicles (90100 µm) were isolated from the transplanted ovaries and then cultured in vitro for 16 days using three different media [19]. Medium 1 consisted of 
-minimal essential medium (-MEM) with 1% heat-inactivated fetal calf serum (FCS) (Gibco-BRL, Carlsbad, CA), 100 mIU/ml of FSH, 10 mIU/ml of LH, and insulin-transferrin-selenium mix (ITS-mix; 5 mg/ml, 5 mg/ml, and 5 ng/ml, respectively) (Sigma). Medium 2 consisted of -MEM with the same components as medium 1, except for the concentration of FCS, which was used at 5%. Medium 3 consisted of -MEM supplemented with 1.5 IU/ml of hCG and 5 ng/ml of recombinant epidermal growth factor (rEGF) (Sigma).
Culture dishes (10-cm Petri dishes; BD Biosciences) containing sixty 10-µl culture droplets covered with 10 ml mineral oil (Sigma) were preincubated overnight at 37°C, 100% humidity, and 5% CO2 in air. Selected follicles were washed and placed individually in the culture droplets (medium 1). At Day 4 after culture, 10 µl of fresh medium 1 was added to each droplet. After Day 6, medium 1 was replaced with medium 2. From Day 8 onward, the depleted medium was refreshed by exchanging 10 µl of fresh medium 2 every other day. On Day 16, 10 µl of medium 2 was replaced with 10 µl of medium 3 to induce final oocyte maturation. After 1618 hours, cumulus expansion of the cumulus-oocyte complexes (COCs) was assessed. Oocyte maturation status was adjudged by germinal vesicle breakdown (GVBD) and extrusion of the first polar body.
Experimental group 2: 28 days after transplantation, ovarian grafts were recovered from the transplanted sites. COCs were isolated from the ovaries by puncturing the largest antral follicles with a needle, and the oocytes were then matured in vitro in compliance with the protocol described above.
Maturation of Oocytes in Controls
Control group 1: the primary and early secondary follicles were isolated from the ovaries of 8-day-old CD-1 mice and cultured in vitro for 16 days using the above-mentioned protocol.
Control group 2: COCs were isolated from the ovaries of 22-day-old CD-1 mice, and oocytes were matured and fertilized in vitro.
In Vitro Fertilization
The caudae epididymis were removed from 10- to 12-wk-old CD-1 male mice and placed into 1-ml drops of a mutant potassium simplex optimized medium (mKSOM) [20] supplemented with 0.4% (w/v) BSA (Sigma) in sperm dispersion dishes. The dispersion dishes were placed in a 5% CO2 incubator for 20 min to allow the sperm to disperse. Ten microliters of sperm suspension was added to 90-µl drops of mKSOM+BSA in the fertilization dishes. Capacitation was allowed to proceed for 4560 min at 37°C in the incubator. After capacitation, COCs were transferred to the 100-µl fertilization droplets. Incubation was allowed to proceed for 4 h at 37°C in a 5% CO2 atmosphere. Finally, approximately 10 inseminated oocytes were incubated in 20-µl drops of KSOM+BSA in culture dishes.
Embryo In Vitro Culture and Transfer
All of the zygotes fertilized in vitro in each experiment were separated into two groups. One group of zygotes was cultured in vitro for approximately 48 h. The medium then was replaced with mKSOM+BSA, and the culture continued for 48 h. By Day 4, embryo development was determined at the late morula-early blastocyst stages. The other zygotes were transferred to the oviducts of 0.5-dpc pseudopregnant mice.
Histology
Ovarian grafts were removed from 3 recipient mice after transplantation on Day 14. Normal ovaries were obtained from 8-day-old mice and processed for histological examination through a comparison with the grafted ovaries. The grafts and ovaries were fixed in Bouin fluid for 24 h and then embedded in paraffin. The grafts and ovaries serially sectioned at 5 µm were stained with hematoxylin and eosin (Sigma). The sections with follicles present were examined, and the developmental stages were determined.
Expression of Specific Genes in the Different Follicular Stages
RNA samples were isolated from the different stage of follicles (n = 10), and oligo-dT-primed cDNA was synthesized using superscript reverse transcriptase (RT) by the manufacturer's protocol (TaKaRa, Japan). One microliter of each RT reaction was used in each 25-µl PCR reaction primed with gene-specific oligonucleotides. The following primers were used to detect mouse specific genes in the different follicular stages: synaptonemal complex protein 3 (Sycp3): 5'-ATG ATG GAA ACT CAG CAG CAA GAG A-3' and 5'-TTG ACA CAA TCG TGG AGA GAA CAA C-3' (325 bp long); the mouse homolog of the yeast meiosis-specific homologous recombination gene, disrupted meiotic cDNA 1 homolog (Dmc1h): 5'-GGA CAT TGC TGA CCG CTT CAA CGT-3' and 5'-GGC GAT CCT CAG TTC TCC TCT TCC-3' (427 bp long); growth differentiation factor 9 (Gdf9): 5'-CCA GCA GAA GTC ACC TCT ACA A-3' and 5'-ACA TGG CCT CCT TTA CCA CA-3' (240 bp long); zona pellucida proteins Zp1, 5'-CTG AGG ATT GCC ACG GAT AA-3' and 5'-GGA GTC AAG GAG CAT GAA GGT-3' (324 bp long), Zp2, 5'-GCT ACA CAC ATG ACT CTC AC-3', 5'-GGT GAC TCA CAG TGG CAC TC-3' (380 bp long), and Zp3, 5'-TTG AGC AGA AGC AGT CCA GC-3' and 5'-CGG TTG CCT TGT GGA TGG TC-3' (441 bp long); luteinizing hormone/choriogonadotropin receptor (Lhcgr): 5'-AAT CTC TCC TTT GCA GAC TTT TG-3' and 5'-AGC ATA GGT GAT GGT GTG CCA-3' (216 bp long); Og2x (an oocyte-specific homeobox gene expressed in germ cell cysts and in primordial and growing oocytes): 5'-CCT TCA GTC ACA GTT TCC GTA T-3' and 5'-GGG AGG TTC TGG CAA GCA AT-3' (226 bp long); Ddx4 (a marker for postmigratory germ cells until postmeiotic stages): 5'-GGT CCA AAA GTG ACA TAT ATA CCC-3' and 5'-TTG GTT GAT CAG TTC TCG AGT-3' (420 bp long) [21–23]. Gapdh primers were used as a control supplied by the RT-PCR kit. The cycling conditions were as follows: 30 sec at 94°C, 30 sec at 60°C, and 1 min at 72°C (25 cycles).
DNA Methylation Analysis
Genomic DNA was prepared from the oocytes (n = 50) that developed in vitro from mouse fetal germ cells in experimental group 1, control group 1, and normal mice. DNA was treated with a sodium bisulfite solution as described previously [24, 25]. Region 2 of the insulin-like growth factor 2 receptor (Igf2r) gene and the 5' upstream region of paternally expressed 3 (Peg3) gene were amplified using Ex-Taq DNA polymerase (TaKaRa). The PCR primers were as follows: Igf2r, 5'-GAG GTT AAG GGT GAA AAG TTG TAT-3' and 5'-CAC TTT TAA ACT TAC CTC TCT TAC-3' (450 bp long); Peg3, 5'-TTT TGT AGA GGA TTT TGA TAA GGA GGT G-3' and 5'-CCC CAA ACA CCA TCT AAA CTC TAC AAA C-3' (288 bp long). The cycling conditions were as follows: 30 sec at 94°C, 30 sec at 60°C, and 1 min at 72°C (35 cycles). Amplified fragments were cloned into the plasmid vector pGEM-T Easy (Promega, Madison, WI), and 15 samples in each experiment sequenced using an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA).
Statistical Methods
Follicle survival data, oocyte and follicle diameter, and nuclear maturity were analyzed by chi-square test for likely trends. The proportion of oocytes that underwent GVBD and fertilization and late morula-early blastocyst formation were compared by contingency table analysis followed by chi-square test. P < 0.05 was considered statistically significant.
RESULTS
Experimental Process
As shown in Figure 1, we developed an efficient, simple two-step method to obtain mature oocytes derived from mouse 12.5-dpc premeiotic fetal germ cells. In experiment group 1 we first transplanted 12.5-dpc fetal mouse ovaries under mouse kidney capsules for 14 days to allow transitional development from the premeiotic germ cells to the primary oocytes. In the second step, we mechanically isolated the developed follicles from the ovarian grafts and subsequently cultured them in vitro for 16 days; mature oocytes were then obtained from the mature antral follicles. In experimental group 2 we transplanted 12.5-dpc fetal ovaries under the kidney capsule for 28 days to test whether the process of the development from premeiotic germ cells to mature oocytes could be accomplished entirely or partially in an ectopic site such as the kidney capsule. As controls for the development of oocytes in both experimental group 1 and group 2, the follicles developed in vivo entirely or partially in the mouse ovaries were studied in control groups. In control group 1 the first step was completed in mouse ovaries in vivo, and then the developed follicles from 8-day-old mouse ovaries were isolated and cultured in vitro for 16 days. In control group 2 both steps were completed in vivo in mouse ovaries, and the oocytes were isolated from the antral follicles of 22-day-old mouse ovaries.
|
Germ Cells of 12.5-dpc Fetal Ovaries Are in Premeiotic Stage
Mouse 12.5-dpc fetal ovaries (Fig. 2A) were obtained from the pregnant mice, and the germ cells of the fetal ovaries were analyzed using anti-DDX4 antibody (Fig. 2B). To test whether the germ cells derived from mouse 12.5-dpc ovaries are in premeiotic phase, we analyzed the mouse 12.5-, 13.5-, and 14.5-dpc female germ cells with antibody anti-SYCP3. SYCP3 is a meiosis-specific structural protein that appears at axial elements and lateral elements of the synaptonemal complex, which is an excellent marker for detection of the meiotic transition in mammals because its expression is required for the onset of the first meiotic division [22, 23]. Expression of SYCP3 protein was not found in female germ cells in the 12.5-dpc female embryos. However, the antigens recognized by the anti-SYCP3 antibody appeared around 13.5 dpc as fluorescent dots in more than 80% of female germ cells (Fig. 2C). On 14.5 dpc, a diffuse immunofluorescence staining of SYCP3 could be seen in more than 70% of oocytes (Fig. 2D); the immunostaining patterns of SYCP3 were similar to those described by Di Carlo et al. [17]. These results demonstrated that the female germ cells of 12.5-dpc fetal ovaries were in the premeiotic stage, and, subsequently, around 13.5 dpc germ cells entered into the early meiotic prophase.
|
Generating Developed Oocytes from 12.5 dpc Premeiotic Fetal Germ Cells using a Two-Step Method
In experimental group 1, mouse 12.5-dpc fetal ovaries were obtained from the pregnant mice, and histological examination of the 12.5-dpc fetal ovary revealed that no follicles were appeared (Fig. 2E). Fetal ovaries without attached mesonephroses were inserted into the left renal capsules of the recipient mice. In total, 201 fetal ovaries were transplanted into 76 recipients, and 146 grafts were recoverable from 58 recipients after 14 days of transplantation. Histological examination of the grafts recovered at Day 14 revealed that the follicles had proceeded to the primary and early secondary follicle stages, as shown in Figure 2, F and G, and the development of the follicles grown in the ovarian grafts was similar to that of the follicles derived from 8-day-old normal mice (Fig. 2H). Subsequently, 1827 follicles were isolated from the recovered grafts and cultured in vitro for 16 days. During this period follicles grew progressively from a spherical to a diffuse structure and then displayed an antrallike cavity formation (Fig. 2, I–N).
As shown in Figure 3A, 83% of the follicles survived during the 16-day culture period, and a high proportion (87.1%) of the surviving follicles were diffuse at Day 10. After Day 12, a first antrallike cavity formation was observed (Fig. 2L). Among the surviving follicles, 1012 (66.4%) of 1523 became antral follicles on Day 14, and COCs in the antrallike cavity were observed (Fig. 2N). In control group 1 we isolated 1164 follicles from 8-day-old mouse ovaries and then cultured them in vitro for 16 days. A total of 89% of the follicles survived during the culture period, 91% of these surviving follicles were diffused by Day 10 of culture, and, finally, 74.7% had became antral follicles at Day 14. The results in Figure 3A show that there was no significant difference for mouse follicular development between experimental group 1 and control group 1.
|
We measured the diameters of oocytes and follicles in experimental group 1 and compared them with control group 1 during the 16-day culture process (Fig. 3, B and C). The mean diameter of the oocytes in experimental group 1 (n = 224) at Day 16 was 75.2 ± 4.8 µm; the mean diameter of the oocytes in control group 1 (n = 170) at Day 16 was 76.7 ± 3.9 µm (Fig. 3B). The mean diameter of the follicles (n = 247) cultured for 16 days in vitro was 403 ± 37 µm in experimental group 1, whereas the mean diameter of the follicles (n = 256) in control group 1 was 421 ± 33 µm (Fig. 3C). As shown in Figure 3, B and C, there were no significant differences in the diameters of the oocyte and follicle between experimental group 1 and control group 1. We also measured the diameters of oocytes in experimental group 2 and compared them with control group 2. The mean diameter of the oocytes in experimental group 2 (n = 176) was 78.6 ± 4.3 µm; the mean diameter of the oocytes in control group 2 (n = 113) was 79.2 ± 4.1 µm. Compared with previously published studies [9, 11, 19, 20, 26], the sizes of the oocytes and follicles generated from premeiotic fetal germ cells by our culture method were similar to those of the follicles generated by in vitro culture of primordial follicles and primary follicles.
We further examined by RT-PCR the expression of follicle-specific genes (Ddx4, Zp1, Zp2, Zp3, Gdf9, Og2x, Lhcgr, Dmc1h, and Sycp3) in different stages of in vitro-cultured follicles. As shown in Figure 3D, expression of these follicle-specific genes was detected as early as the second day in our in vitro culture, which is about equal to Postnatal Day 10 of a mouse ovary in vivo (including 8 days of ovarian development under the mouse kidney capsule). In general, their expression patterns were similar to those of the follicles from various stages of postnatal mouse ovaries [19, 21–23].
Producing In Vitro Fertilization Embryos by Oocytes Derived from Premeiotic Germ Cells
To test the full-term developmental capacity of the oocytes derived from premeiotic germ cells, the COCs inside the antral follicles were matured in vitro using a medium supplemented with hCG and endothelial growth factor (EGF) for 1618 hours, and then fertilized in vitro. All of the zygotes fertilized in vitro in each experiment were separated into two groups. One group of zygotes was cultured in vitro and the other was transferred to the oviduct of pseudopregnant mice.
In experimental group 1 (a total of nine experiments were performed), 1194 oocytes were isolated from mature follicles (Fig. 4, A and B). Of these, 278 oocytes were matured in vitro, and the results are summarized in Table 1: 243 oocytes (87.4%) exhibited GVBD, and 32.1% of the GVBD oocytes progressed to the M2 stage (Fig. 4C and Table 1). The GVBD oocytes were fertilized, and 37.4% developed to zygotes in vitro. Subsequently, 69.2% of the embryos at the two-cell stage were maintained in culture, whereas 27.0% formed blastocysts at Day 4 (Fig. 4D).
|
|
In control group 1 (a total of seven experiments were performed), 871 oocytes were derived from 8-day-old mouse follicles. Of these, 227 oocytes were matured in vitro, and the results are summarized in Table 1: 83.3% oocytes exhibited GVBD, and 46% of the GVBD oocytes progressed to the M2 stage. A total of 189 oocytes were fertilized, 40.7% developed to zygotes, 66.2% developed into embryos at the two-cell stage, and 21.6% formed blastocysts. Comparing the results between experimental group 1 and control group 1, there was a clear difference in the M2 oocyte formation and blastocyst rates, whereas there were no major differences seen in the GVBD or insemination rate.
In experimental group 1 (a total of five experiments were performed), 113 fetal ovaries were transplanted into 37 recipients, and 87 grafts were isolated after 28 days of transplantation. A total of 410 oocytes were isolated from the transplanted fetal ovaries. Of these, 145 oocytes were matured in vitro, and the results are summarized in Table 1: 93.8% of the recovered oocytes underwent GVBD, 83.1% of the GVBD oocytes progressed to the M2 stage, 64% were fertilized, and 59.4% formed blastocysts from embryos in the two-cell stage.
In control group 2 (a total of three experiments were performed), 376 oocytes were isolated from 22-day-old mice and matured in vitro. Of these, 123 oocytes were cultured in vitro, and the results are summarized in Table 1: 96.7% of the recovered oocytes exhibited GVBD, whereas 86.6% of the GVBD oocytes progressed to the M2 stage, 52.1% were fertilized, and 64.8% formed blastocysts from embryos at the two-cell stage. There was no significant difference between experimental group 2 and control group 2.
Genetic Characterization of Offspring Produced by Oocytes Derived from Premeiotic Germ Cells
To detect embryonic development after implantation, some zygotes derived from in vitro-cultured oocytes were transferred to recipient mice to produce offspring. In experiment group 1, 241 embryos were produced from 916 oocytes, transferred to 12 recipient mice, and 9 embryos developed to term (summarized in Table 2). Seven (six females, one male) of the offspring survived (Fig. 4E). The reproductive ability of these seven mice was further evaluated via natural mating, and healthy offspring were produced by these seven mice, as shown in Figure 4F.
|
In control group 1, 209 embryos were produced from 644 oocytes, transferred to seven recipient mice, and nine living offspring (four females, five males) were produced (summarized in Table 2). Comparing the results in Tables 1 and 2, there were no significant differences in either the formation rate of GVBD oocytes or the insemination and the reproduction rates between experimental group 1 and control group 1, which indicated that the kidney capsule was able to functional equivalently to mouse ovaries, allowing premeiotic germ cells to differentiate into primary oocytes, which then could be fertilized and which supported development to term.
In experiment group 2, 160 embryos were produced from 265 oocytes. The results are summarized in Table 2: 18 embryos developed to term, and 16 (6 females, 10 males) offspring survived. Of 124 transferred embryos derived from 253 oocytes in control group 2, 30 developed to term and 29 (13 females, 16 males) offspring survived. The results summarized in Tables 1 and 2 demonstrate that the formation rate of M2 oocytes, the developmental rate of blastocysts, and the reproduction rate in experimental group 2 and control group 2 were similar and significantly higher than those in experimental group 1 and control group 1 (P < 0.01). These results indicate that the implanted kidney capsule functions equivalently to mouse ovaries in allowing premeiotic germ cells to differentiate to GV-stage oocytes, and the kidney capsule microenvironment proved to be better than the in vitro culture conditions for follicular development from the primary follicle to the antral follicle stage.
In experimental group 1, the mean body weight of the in vitro fertilization mice was 1.58 ± 0.28 g (n = 9), and in control group 1 it was 1.67 ± 0.31 g (n = 9). In experimental group 2 and control group 2 the weights were 1.51 ± 0.24 g (n = 18), and 1.54 ± 0.21 g (n = 30), respectively. No significant differences were observed among these four groups.
To check whether the methylated patterns of imprinted genes were completed during the oogenesis observed in our two-step method, we further examined the differentially methylated regions (DMRs) of the maternal imprinted gene Igf2r in the oocytes of experimental group 1. In mammals, parental imprinted memories persist in somatic cells after fertilization, and it is necessary for them to be erased and re-established during germ cell development to reflect the gender of the individual [27–29]. The erasure process of genomic imprinting memory proceeds in mouse PGCs from 10.5 dpc to 11.5 dpc [30, 31], and the maternal-specific imprints are established during oogenesis [32, 33]. The Igf2r gene, a maternally expressed imprinted gene, has been shown with a fully methylated pattern in the maternal allele and an unmethylated pattern in the paternal allele in somatic cells, and it begins to be methylated in mouse oocytes after 13.5 dpc [34]. As shown in Figure 4G, we analyzed the DNA methylation status in DMRs of Igf2r and in the control gene peg3, a paternally expressed imprinted gene. In experimental group 1, the Igf2r gene was methylated at DMR CpG sites in the oocytes developed in vitro, and the percentage of DNA clones with methylated sites accounted for more than 73.3% of all the CpG sites. Similar results were obtained in the normal oocytes and the control group 1.
DISCUSSION
In this study we show that premeiotic female germ cells derived from mouse fetuses as early as 12.5 dpc can be induced to efficiently complete meiosis by a combination of in vivo transplantation and in vitro culture, and these mature oocytes are highly competent in supporting development to term after in vitro fertilization. Furthermore, we confirmed that the erasure of genomic imprinting markers that had been imposed in the parental generation were imposed in the oocytes that differentiated from premeiotic germ cells. To our knowledge, this is the first time that living offspring have been produced by oocytes differentiated from premeiotic fetal germ cells.
In previous studies, mouse oocytes grown in vitro derived from premeiotic germ cells have been unable to resume meiosis and progress to the last stage of growth [8, 12, 14, 15]. In a female embryo, the germ cells at 13.5 dpc undergo one further round of DNA replication and then prophase of the first meiotic division in vivo. From fetal germ cells, however, the in vitro-grown oocytes were only able to undergo the first meiotic transition into the leptotene stage and were in a state of suspension [8, 14]. Earlier studies by De Felici et al. [16] and Pesce et al. [35] showed that the isolated PGCs recovered from 11.5- and 12.5-dpc gonads did not survive without somatic cell support and underwent rapid apoptotic degeneration. It was then realized that the interplay between the germ cells and the somatic cell lineages in the fetal ovary is essential for oocyte development [3, 5, 14]. When explanted into a cultured aggregate of embryonic lung tissue, the germ cells were found to be able to enter into meiosis at the same chronological time in the ovary [14]. If very few germ cells enter the female genital ridge, the granulosa cell differentiation is initiated, but the supporting cell lineage fails to differentiate, and the oocytes subsequently die or are eliminated [5]. Eppig [36] and Epifano and Dean [37] further demonstrated that the communication between oocytes and granulosa cells is bidirectional, and a complex interplay of regulatory factors governs the development of both types of cell, which is essential not only for oocyte development but also for follicular development, beginning with the initial assembly of the primordial follicle and continuing throughout ovulation. Most recently, Farini et al. [13] established a co-culture method on somatic cell monolayer with Kit ligand (KITL), fibroblast growth factor 2 (FGF2), and leukemia inhibitory factor (LIF) to allow purified mouse PGCs to reach diplotene stage, but no mature and fertilizable oocytes were obtained. In our initial experiments, the premeiotic germ cells were isolated from 12.5-dpc fetal mouse ovaries, and aggregated to the ovarylike cell clusters with the 12.5-dpc fetal ovarian somatic cells. Furthermore, by in vitro culture or the transplantation under the mouse kidney capsule method, only a few primordial follicles formed in the aggregated cell clusters (data not shown). Therefore, we changed our strategy to use premeiotic germ cells preserved in the natural fetal ovary rather than isolated PGCs as the starting materials, then performed culture in vitro or transplant in vivo for 28 days. Using this modified approach, a large number of follicles formed, and the oocytes of the follicles were able to further develop into more than 75-µm GV-stage oocytes. These results suggest that somatic cell support in the context of the intact 12.5-dpc fetal mouse ovary is most favorable for the development of germ cells in vitro or at an ectopic site in vivo.
Obata et al. [15] and Niwa et al. [12] also used intact mouse 12.5-dpc fetal ovaries to culture for 28 days in vitro, but the oocytes could not complete the secondary meiosis by their in vitro culture method. Several possible factors could account for the incompetence of oogenesis in vitro. One is an absence of synchrony between nuclear and cytoplasm maturation, which is necessary for the acquisition of the competence to resume meiosis [8, 14, 38]. Another is that abnormal gap junctions between the follicular granulosa cells and oocytes could influence the development of oocytes and block meiosis [10, 36, 39, 40]. Yet another is that the oocytes generated in vitro are unable to grow to adequate size, since the oocyte volume needs to reach 80% of its maximal volume to resume meiosis [1, 9]. Efforts to overcome these problems with the in vitro culture method have been limited by the fact that so little is known about the mechanisms underlying the initiation of oocyte growth, the formation of gap junctions in the follicle, and the entry into meiosis of the growing oocyte [10, 30]. To overcome the problems with the in vitro methods, in our study mouse fetal germ cells were transplanted under the microenvironment of the kidney capsule. It has been shown that the kidney capsule, which is highly vascularized and rich in angiogenic factors, enhances survival, revascularization, and reanastomosis of the engrafted ovaries, and therefore provides an ectopic ideal site for the development of the ovary [20, 26, 41–53]. Previously, several studies showed that certain stages of oogenesis from the primordial follicle to the mature follicle could be completed under the kidney capsule [20, 26, 41–53]. In this study, however, we have shown that the whole process of oogenesis from the premeiotic germ cell to the GV-stage oocyte can in fact be carried out under the kidney capsule. This approach should prove to be a powerful tool to study the development of germ cells in ectopic sites, and it will advance the search for the critical factors required for the induction of meiosis in vitro.
To facilitate the survival and growth of ovarian grafts, we used ovariectomized recipient mice for the ectopic transplantation of the mouse ovary. Several earlier studies [41, 46, 47, 53] have shown the growth of ovarian grafts and antral follicular development to be retarded or inhibited after the transplantation of ovarian tissue to intact recipients. This problem can be solved by using ovariectomized recipients, which promotes the development of mouse ovarian grafts by providing an endocrine milieu with high concentrations of circulating gonadotrophins, and low concentrations of estrogen and inhibin [46, 53]. The ovariectomy of the recipient gives rise to two important physiologic changes: circulating gonadotrophins are elevated, and factors secreted by the intact recipient ovaries are eliminated [53]. High levels of gonadotrophins in ovariectomized recipients facilitate follicular survival and development in transplanted ovaries; the proportion of growing follicles in grafted newborn ovaries under kidney capsule for 14 days was higher than those in vivo 2-wk-old mouse ovaries [20, 26, 53].
We have demonstrated that the problem of incompetent oogenesis in vitro can be overcome by this two-step method using a combination of 14-day in vivo transplantation and 16-day in vitro culture. We initially tried to culture the premeiotic female germ cells derived from mouse fetuses as early as 12.5 dpc in vitro, or else transplant the fetal ovaries under the kidney capsule for 7 days and subsequently culture in vitro. However, no mature and fertilizable oocytes could be obtained (data not shown). It has been reported by Eppig and O'Brien [9] and O'Brien et al. [11] that the oocyte–granulosa cell complexes from mice younger than 6 days are unable to sustain cultures in vitro. However, when the ovaries of newborn mice were grown in organ culture for 8 days, and then the developing oocyte–granulosa cell complexes were isolated from the organ-cultured ovaries and cultured for an additional 14 days, these oocytes were fertilizable in vitro and supported the development to term. Lenie et al. [19] reported that the early secondary mouse follicles from 8-day-old mouse ovaries could be cultured for 18 days and mature in vitro. It seems that the ovarian environment is crucial for the early development of mouse follicles before postnatal 8 days. Therefore, we first transplanted the premeiotic female germ cells of a 12.5-dpc ovary under the mouse kidney capsule for 14 days (including 6 days fetal ovarian development and 8 days postnatal ovarian development), and then the early secondary follicles of the grafted fetal ovary were isolated as soon as possible and subsequently cultured in vitro for 16 days, following the method of previous studies [9, 11, 19, 20, 26]. As shown in the results summarized in Tables 1 and 2, premeiotic germ cells were able to complete two meiotic divisions and generate a large number of mature oocytes, which were highly competent in supporting development to term. The oocytes generated by this method ultimately reached approximately 75.2 µm in diameter (Fig. 3B), which is similar to the size of oocytes grown normally in vivo. In future experiments, we will further optimize this two-step method and attempt to accomplish the whole process of mouse oogenesis in vitro from mouse PGCs.
ACKNOWLEDGMENTS
We thank Prof. Zhaoyi Wang, Dr. Yan Shi, and Dr. Yushan Guo for critical reading of the manuscript.
FOOTNOTES
1 Supported by grants from the Ministry of Science and Technology (2001CB510106), Science and Technology Plan of Beijing Municipal Government (H020220050290), National Nature Science Foundation of China for Outstanding Young Scientist Award (30125022) and for Creative Research Groups (30421004), and the Bill & Melinda Gates Foundation (37871) to H.D., and the Chinese Postdoctoral Science Foundation (2005037030) to W.S. 
2 Correspondence. FAX: 86 10 6275 6954; hongkui_deng{at}pku.edu.cn
Received: 9 February 2006.
First decision: 7 March 2006.
Accepted: 15 May 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  J C Sadeu, T Adriaenssens, and J Smitz Expression of growth differentiation factor 9, bone morphogenetic protein 15, and anti-Mullerian hormone in cultured mouse primary follicles Reproduction, August 1, 2008; 136(2): 195 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Qing, H. Liu, W. Wei, X. Ye, W. Shen, D. Zhang, Z. Song, W. Yang, M. Ding, and H. Deng Mature oocytes derived from purified mouse fetal germ cells Hum. Reprod., January 1, 2008; 23(1): 54 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Linher, D. Wu, and J. Li Glial Cell Line-Derived Neurotrophic Factor: An Intraovarian Factor that Enhances Oocyte Developmental Competence in Vitro Endocrinology, September 1, 2007; 148(9): 4292 - 4301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Shen, L. Li, Z. Bai, Q. Pan, M. Ding, and H. Deng In vitro development of mouse fetal germ cells into mature oocytes Reproduction, August 1, 2007; 134(2): 223 - 231. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
