Comparison of different in vitro differentiation conditions for murine female germline stem cells

Abstract Objectives In vitro differentiation of oocytes from female germline stem cells (FGSCs) has exciting potential applications for reproductive medicine. Some researchers have attempted to reveal the in vitro differentiation capacity of FGSCs. However, no systematic comparative study of in vitro differentiation conditions has been performed for murine FGSCs (mFGSCs). Materials and Methods mFGSCs line was cultured under five different conditions for in vitro differentiation. RT‐PCR was performed to detect the expression of Oct4, Fragilis, Blimp1, Mvh, Scp3 and Zp3. Immunofluorescence was carried out to test the expression of Mvh, Fragilis and Zp3. Two‐photon laser‐scanning microscope was used to analyze nucleus‐plasma ratio, and the proportion of chromatin of GV oocytes differentiated from mFGSCs in vitro (IVD‐GVO), GV oocytes from in vivo (GVO) and mFGSCs. Results RT‐PCR and immunofluorescence showed that mFGSC line expressed germ cell‐specific markers, but not a meiosis‐specific marker. By evaluating five different in vitro differentiation conditions, condition 5, which included a hanging drop procedure and co‐culture of mFGSCs with granulosa cells, was shown to be optimal. mFGSCs could be successfully differentiated into germinal vesicle (GV) ‐stage oocytes under this condition. 3D observation revealed that both the nucleus‐plasma ratio and proportion of chromatin were not significantly different between IVD‐GVO and GVO. Conclusion We evaluated five in vitro differentiation conditions for mFGSCs and successfully differentiate mFGSCs into GV‐stage oocytes using a three‐step differentiation process.

process of mammalian oogenesis are a longstanding focus of reproductive and developmental biology research.
Recently, researchers have reported the possibility of deriving female gametes from stem cells in vitro. 3 In 2003, Hubner et al first reported that mouse embryonic stem cells (mESCs) in culture could develop into oogonia that entered meiosis, recruited adjacent cells to form follicle-like structures, and later developed into blastocysts. 4 However, Novak et al found that mESC-derived oocytes did not progress through meiosis. 5 Interestingly, Lacham-Kaplan et al showed that male mESCs form ovarian-like structures containing putative oocytes when cultured in newborn mouse testicular cell-conditioned medium. 6 Qing et al found that ovarian granulosa cells induced mESCs to differentiate into oocyte-like cells, which expressed meiosis-and oocyte-specific genes. 7 Eguizabal et al achieved complete in vitro differentiation of human induced pluripotent stem cells (iPSCs) into post-meiotic cells. 8 12 FGSCs in long-term culture maintained their capacity to produce normal oocytes and fertile offspring after transplantation into ovaries. 12 Wu et al further traced and characterized the development of transplanted FGSCs in vivo. 13 Xie et al reported similar morphological and molecular signatures between female and male germline stem cells. 14 Li et al systematically identified and compared the expression profiles of lncRNAs and circRNAs in mFGSCs. 15 Moreover, Zhang et al performed integrative epigenomic analysis to reveal the unique epigenetic signatures involved in unipotency of mFGSCs. 16 In 2012, White et al reported ovaries of reproductive-age women possessed rare mitotically active germ cells that could be propagated in vitro. 17 Ding et al generated GV-stage oocytes from human FGSCs obtained from follicular aspirates. 18 Additionally, Zhou et al obtained FGSCs from female rat ovaries and developed a threestep system to differentiate rat FGSCs into GV-stage oocytes in vitro. 19 Many attempts have been made to reveal the differentiation capacity of FGSCs in vitro and in vivo. [20][21][22][23][24] However, no systematic comparative study of in vitro differentiation conditions for mFGSCs has been reported. Therefore, in this study, we evaluated five different in vitro differentiation conditions for mFGSCs. Results showed that a three-step procedure was the optimal differentiation condition for differentiating mFGSCs into GV-stage oocytes. Furthermore, we preliminarily studied the characteristics of these in vitro-differentiated cells using RT-PCR, fluorescent immunocytochemistry, and two-photon laserscanning microscope (TPLSM). Our study not only provides a tool to systematically identify factors and pathways involved in promoting oogenesis from FGSCs, but also has exciting potential applications for reproductive and regenerative medicine.

| Animals
Three-and six-week-old CD-1 wild-type female mice were used in this study. All procedures involving animals were approved by the Institutional Animal Care and Use Committee of Shanghai and were conducted in accordance with the National Research Council Guide for Care and Use of Laboratory Animals.

| Isolation and culture of mouse granulosa cells
Granulosa cells (GCs) were isolated and cultured as previously described with some modification. 18,25 Briefly, ovaries from 3-week-old CD1 female mice were dissected free of fat, bursa and oviduct. After washing with phosphate-buffered saline (PBS), GCs were released by manually puncturing ovaries with 25-gauge needles. Cell suspensions were then passed through a 40-μm nylon cell strainer and cen-

| Preparation of ovarian homogenate
Ovarian homogenate was prepared as previously described with some modification. 26 Ovaries from 5-6 wild-type adult mice were harvested and homogenized in 2 mL of D-Hanks buffer.
Homogenates were subsequently filtered through a 0.22-μm membrane to remove cell debris, aliquoted and stored at 4°C for further use.  Table 1. All cultures were maintained at 37°C in a 5% CO 2 atmosphere with morphological features monitored daily.

| In vitro differentiation
MEM-α Up to 10 mL Up to 10 mL Up to 10 mL Up to 10 mL Up to 10 mL Up to 10 mL Up to 10 mL

| BrdU labelling
BrdU (50 mg/mL; Sigma) was supplied into FGSCs culture medium to a final concentration of 50 mg/mL, and FGSCs was incubated for 5 hours with BrdU-containing medium before immunofluorescence assay.

| TUNEL staining
The protocol for TdT-mediated dUTP Nick-End Labelling (TUNEL) staining was performed based on a TdT-mediated dUTP Nick-End Labelling kit (Beyotime, China). Briefly, FGSCs after differentiation treatment were plated on poly-lysine coated 96-well dishes for 1 hour. Subsequently, the cells were fixed with 4% PFA and treated with 0.3% triton X-100 for permeation. After rinse with PBS, TdT enzyme contained TUNEL fluorescent staining medium was supplied to samples, and the samples were placed in the dark for 1 hour. Then, the medium was aspirated, and the samples were rinsed with PBS for twice, and DAPI was added for counterstaining. Finally, observed the sample under fluorescent microscope.

| Immunofluorescence
Culture media were discarded gently, and cells were washed carefully with PBS, followed by fixation using 4% paraformaldehyde at

| Image analysis
Consecutive optically sectioned images of GV oocytes, in vitro-differentiated FGSCs, and mFGSCs were preprocessed and analyzed using 3D reconstruction software Amira 5.2 (Visage Imaging, Berlin, Germany), as previously described. 27 Critical morphological feature parameters for the three types of cells included cell volume, nucleus volume and chromatin volume, with the coordinates of nuclear and chromatin centres quantified by Amira 5.2.

| Statistics
All data are presented as mean ± SEM. Statistical tests were performed with Student's t test using Statistical Package for the Social Sciences (SPSS) software (version 20.0; IBM). P < 0.05 was considered statistically significant, and P < 0.01 was considered highly significant.
Graph generation was carried out using SigmaPlot software (version 13.0).

| Characterization of mFGSC line
The mFGSC line reported in our previous study was subjected to in vitro-induced differentiation ( Figure 1A). After STO feeder cells were removed from mFGSC cultures by differential adherence,and the molecular signatures of the cultured mFGSC line were character-

| Characteristics of mFGSCs after in vitro differentiation using condition 1
To differentiate mFGSCs into oocytes in vitro, five differentiation conditions were evaluated (Figure 2). It is well recognized that

| Characteristics of mFGSCs after in vitro differentiation using condition 2
Ovaries

| Characteristics of mFGSCs after in vitro differentiation using condition 3
Based on the above observations, RA contributed to the differentiation of mFGSCs, and OH was conducive to maintaining the in vitro survival of mFGSCs. In condition 3, mFGSCs were subjected to in vitro differentiation in Medium 3, which was supplied with both RA and OH. The morphology of mFGSCs remained similar to those induced with RA only for the first 2 days.
There were 2-3 round cells per 40× field observed to be up to

| Characteristics of mFGSCs after in vitro differentiation with condition 4
Oestrogen and progesterone are important endocrine hormones that regulate the development of oocytes. 28 To evaluate the impact of hormones on differentiating mFGSCs, after 10 days of culture in differentiation Medium 3, cells were exposed to differentiation Medium 4 supplied with oestrogen and progesterone. The first day in Medium 4, no obvious morphological changes were observed ( Figure 5A). Two days later, the diameter of some round cells grews to nearly 45 μm ( Figure 5B), meanwhile some cells began to form vacuoles ( Figure 5C). At day 5, the size of round cells was mostly unchanged ( Figure 5D). On day 7, many round cells exhibited severe apoptosis ( Figure 5E). Differentiating cells in this condition were prone to apoptosis and could not be maintained longer in culture.

| Characteristics of mFGSCs after in vitro differentiation with condition 5
Granulosa cells transduce multiple signals to participate in the ini- These results suggested that this condition was optimal for in vitro differentiation of mFGSCs.

| Expression analysis of Scp3 and Zp3
The developmental stages of cells during differentiation in these conditions were examined. Expression of Scp3 (a meiosis-specific marker) and Zp3 (an oocyte-specific marker) was detected at days

| 3D observation and preliminarily quantitative analysis of GV oocytes differentiated from mFGSCs in vitro, GV oocytes from in vivo and mFGSCs
To further study the characteristics of in vitro-differentiated mFGSCs under condition 5, three types of cells including GV oocytes differentiated from mFGSCs in vitro (IVD-GVO), GV oocytes from in vivo (GVO), and mFGSCs, were collected for 3D observation by TPLSM. Images of the x-y plane for these three types of cells are shown in Figure 8A. There was a significant difference in diameter between IVD-GVO (53 ± 1.39 μm) and GVO (69 ± 2.09 μm), P < 0.01). However, the diameter of both IVD-GVO and GVO was much larger than mFGSCs (11.2569 ± 0.31 μm,

| D ISCUSS I ON
Successfully obtaining functional oocytes in vitro from stem cells is not only conducive to understanding the regulatory mechanisms of oogenesis, but also improving the fecundity of mammalian females.
FGSCs are capable of producing fully functional oocytes and fertile offspring after transplantation into ovaries. 12,13,19,29 They provide an alternative strategy to study mammalian oogenesis in vitro because We preliminarily evaluated five different in vitro differentiation conditions for mFGSCs. It is believed that retinoic acid (RA) causes germ cells in the ovary to enter meiosis and initiates oogenesis. 27,30 Thus, RA induction was first attempted to promote FGSC differentiation in our study (condition 1 Stem cells located within a specific microenvironment, known as a niche that has both anatomical and functional dimensions. 32 FGSCs exist in the ovarian cortex surface beneath the epithelium. 13 Ovaries, the female reproductive organ, are both gonads and endocrine glands. They are the site of production and periodical release of egg cells. In our experiments, the OH-supplied condition (condition 2) turned out to be insufficient for FGSC differentiation; instead, FGSCs retained better renewal and growth, suggesting that OH supports FGSC maintenance rather than initiation of differentiation. Therefore, we subsequently evaluated the synergistic effect of RA induction and OH conditioning (condition 3) and observed a more advanced differentiation status. Notably, differentiating more mature oocytes under condition 3. This may be ascribed to the fact that they failed to receive effective hormone stimulation.
Oestrogen (E2) and progesterone (P4) are important endocrine hormones that regulate oocyte maturation. 28 Thus, we next attempted to expose differentiating cells to condition 4 after culture with differentiation condition 3 for 10 days. When E2 and P4 stimulation was applied, differentiating FGSCs showed no significant changes compared with their unsupplemented counterparts. The differentiation process was further impeded by intensive cell apoptosis in prolonged culture, suggesting that E2 and P4 were not sufficient for the maturation of differentiating FGSCs; this might require more complex "niche" conditions. GV-stage oocytes in vitro, respectively. 18,19 This result indicates that this three-step system is the most optimal differentiation condition for murine, rat and human FGSCs reported to date. Moreover, the in vitro differentiation mechanism of FGSCs may be conserved among species.
In summary, we evaluated five different in vitro differentiation conditions for mFGSCs and successfully differentiated mFGSCs into GV-stage oocytes under a three-step differentiation condition. To our knowledge, this is the first observation of mouse GV oocytes derived from mFGSCs in vitro. While further investigation is needed to determine whether it is possible to produce fertilizable oocytes from FGSCs in vitro, our study provides both a valuable model for studying the mechanisms underlying mammalian oogenesis and an important alternative source of oocytes. Medicine-Engineering Fund (YG2017ZD11).

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest regarding the publication of this article.