Oocyte-like cells induced from mouse spermatogonial stem cells
© Wang et al.; licensee BioMed Central Ltd. 2012
Received: 13 July 2012
Accepted: 16 July 2012
Published: 6 August 2012
Skip to main content
© Wang et al.; licensee BioMed Central Ltd. 2012
Received: 13 July 2012
Accepted: 16 July 2012
Published: 6 August 2012
During normal development primordial germ cells (PGCs) derived from the epiblast are the precursors of spermatogonia and oogonia. In culture, PGCs can be induced to dedifferentiate to pluripotent embryonic germ (EG) cells in the presence of various growth factors. Several recent studies have now demonstrated that spermatogonial stem cells (SSCs) can also revert back to pluripotency as embryonic stem (ES)-like cells under certain culture conditions. However, the potential dedifferentiation of SSCs into PGCs or the potential generation of oocytes from SSCs has not been demonstrated before.
We report that mouse male SSCs can be converted into oocyte-like cells in culture. These SSCs-derived oocytes (SSC-Oocs) were similar in size to normal mouse mature oocytes. They expressed oocyte-specific markers and gave rise to embryos through parthenogenesis. Interestingly, the Y- and X-linked testis-specific genes in these SSC-Oocs were significantly down-regulated or turned off, while oocyte-specific X-linked genes were activated. The gene expression profile appeared to switch to that of the oocyte across the X chromosome. Furthermore, these oocyte-like cells lost paternal imprinting but acquired maternal imprinting.
Our data demonstrate that SSCs might maintain the potential to be reprogrammed into oocytes with corresponding epigenetic reversals. This study provides not only further evidence for the remarkable plasticity of SSCs but also a potential system for dissecting molecular and epigenetic regulations in germ cell fate determination and imprinting establishment during gametogenesis.
Despite the different genotypes of germ cells in males with XY cells and females with XX cells, both types of germ cells share the same progenitors, namely, primordial germ cells (PGCs). The differentiation of PGCs into either the male or female phenotype takes place in the sex glands at later stages of embryonic development, and sexual differentiation of the germ cells is controlled by the somatic environment of the gonad rather than the sex chromosome constitution of the germ cells themselves [1–3]. Somatic mutation of sex-determining genes contributes to the sex reversal of XY germ cells to oogonia during gonad development; thus, the fate of XY male germ cells varies in response to environmental signaling in the gonad . A few recent studies have demonstrated that spermatogonial stem cells (SSCs), which are the progeny of PGCs/gonocytes, can be reprogrammed into embryonic stem-like cells in vitro without transgene manipulation [5–9], indicating that SSCs retain remarkable plasticity. In addition, XY embryonic stem cells (ESCs) can differentiate into oocytes in culture . Therefore, it is interesting to know whether SSCs can be reprogrammed into female germ cells. Here, we report that SSCs can be converted into oocyte-like cells in culture.
To further confirm the genotype of these SSC-Oocs, we carried out a PCR analysis of Sry, which is located only on the Y chromosome, in the GFP-expressing larger SSC-Oocytes. We found that Sry was not present (Figure 3C), indicating that these SSC-oocytes should be of the XO karyotype and that YO cells died while growing due to the lack of whole X-linked genes.
The activity of X chromosome-linked genes in male germ cells is different from that of female germ cells. The Y chromosome genes have been reported to be essential for spermatogenesis but not for oogenesis. Therefore, we addressed the gene activity status of sex chromosomes in the SSC-Oocs. We examined the expression of sex-dependent X- and Y-linked genes [29, 30] and found that the X-linked testis specific genes were significantly down-regulated or turned off (Figure 5C and Additional file 4: Table S2), while oocyte specific genes including GDF9, X-linked BMP15 and Usp9x were turned on (Figure 5D). The Y-linked genes were silenced (Figure 5C and Additional file 4: Table S2). The loss of the expression of Y-linked gene could be the result of the absence of the Y chromosome in the SSC-Oocytes since YO cells might die. These data indicate that the gene expression pattern of the X chromosome was changed in favor of the formation of oocytes from SSCs. Thus, following the process of SSCs dedifferentiation back to gonocytes/PGCs and transdifferentiation into oocytes, the X chromosome might be subjected to large scale changes in gene expression and epigenetic modifications. This hypothesis was supported by the reversal expression of Usp9x (Figure 5D and Additional file 2: Table S2), which is X-linked and expressed in both male and female embryonic germ cells, but turned off in male germ cells after birth . Therefore, along with the morphological changes during the conversion of SSCs into oocytes, the epigenetic network was converted into the female germ cell form.
In mice, cellular pluripotency reprogramming mostly relies on the extrinsic signaling of LIF and the intrinsic factor Oct4; LIF signaling is sufficient in reprogramming of epiblast cells, in which Oct4 is not expressed, into pluripotent ESCs [32, 33]. Oct4 in a defined culture condition can reprogram somatic cells into pluripotent cells . It has also been revealed that a reversible path from stem cells to differentiation in the germ cell lineage exists [35–37]. In addition, a few studies have demonstrated that a small fraction of mouse SSCs can be reprogrammed back to embryonic stem-like cells [5–9]. Based on these findings, we thought as in the reprogramming epiblast cells into ES-like cells [32, 33], the LIF signaling might trigger the dedifferentiation of SSCs. In the present study, we found that SSCs can dedifferentiate back into PGCs and transdifferentiate into oocyte-like cells when cultured in KO-DMEM medium containing 1% FBS, LIF plus 2i followed by DMEM/F12 medium supplemented with 15% FBS, LIF, FSH, EGF, B27, and ITS. This observation is consistent with earlier findings concerning the reprogramming capability of the LIF signaling in the presence of 2i [13, 14] and further indicates the remarkable plasticity of SSCs in culture.
In our study, SSCs from 8 day-old testes were isolated and characterized as GFRa1+/PLZF+ but c-kit-. C-kit is a key marker of PGCs, and lack of c-kit expression excluded the presence of PGCs in the isolated cells. Thus, the isolated SSCs were unipotent. We used SSCs from 8 day-old OG2 male mice carrying EGFP transgene under an Oct4 promoter to trace the dedifferentiation. In situ, we can only observe green cells in testes of OG2 mice before postnatal day 6. We cultured SSCs from 8 days old testes, and found that ~20% off total SSCs were induced to express EGFP within one week, thereby indicating the dedifferentiation of SSCs. Despite of the presence of EGFP green cells that indicated they were positive of Oct4, we did not observe any ESC-like colonies formed from these cells. Furthermore, we found that E-cadherin expression was absent in them and teratoma could not form them in nude mice, indicating they were not ESCs but more like PGCs. This was further confirmed by the expression of Blimp1 and Nanog. These cells continued to grow with increasing size, expressed Stella, Nobox, and GDF9, and demonstrated morphology resembling that of oocytes. More interestingly, at around day 25 in culture, we observed ~2% of them developed into large cells with a polar body like MII oocytes. With our characterization of gene expression and morphology, we have clearly demonstrated that oocyte-like cells can be derived from SSCs via PGCs intermediates.
The imprinting patterns are established during gametogenesis with paternal imprints occurring during spermatogenesis and maternal imprints occurring during oogenesis. Defects in imprinting in gametes can give rise to severe problem in embryogenesis and predispose affected individuals to associated diseases after birth. To address if there were imprinting reversals in the conversion of SSCs into oocyte-like cells, we examined three key imprinting events including H19/Igf2, Snrpn, and Dik1-Gtl2/Meg3. It turned out that all three of them were switched to maternal status following the induction of oocyte-like cells from SSCs. Thus, in contrast to early report of germline-derived pluripotent stem (gPS) cells, which retain their original imprinting status , the reprogramming from SSCs to oocyte-like cells was accompanied by imprinting reversal. Therefore, SSCs possess both cellular and epigenetic plasticity and even give rise to oocyte-like cells in vitro in a manner similar to cases of sex reversal in vivo. Finally, we also demonstrated that a small number of SSC-Oocs were capable of being fertilized by sperm in vitro.
Our study has demonstrated that SSCs possess the potential to be reprogrammed into oocyte-like cells in culture. If culture conditions are further optimized, for example, using 3D culture system with ideal supportive cell feeder , to develop oocytes of high quality from SSCs, such a germ cell fate switch system should provide a useful in vitro model to study epigenetic regulation in oogenesis and sex reversal, furthering our understanding of the mechanisms that establish imprinting during gametogenesis.
This study was conducted in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals with the approval (SYXK-2003-0026) of the Scientific Investigation Board of Shanghai Jiao Tong University School of Medicine, Shanghai, China. Mice were euthanized by CO2 inhalation to ameliorate any suffering throughout these experimental studies.
We used two enzymatic steps to isolate the adult ovarian germ cells from a 60-day old nursing BALB/c female mouse, whose male pups were used to isolate SSCs. The ovarian tissue was cut into small pieces followed by trypsin digestion for 5 minutes, then washed with DMEM/F12 (GIBCO, Grand Island, NY, USA) once followed by treatment with 0.1% bovine testicular hyaluronidase (Sigma Aldrich, St Louis, MO, USA) for 20 minutes. Cells were dispersed by pipette and suspended in DMEM/F12 medium containing 1% FBS (Biochrom AG, Berlin, Germany) and seeded onto 6 well plates. After 24 hours culture, the round cells, which were mostly on the top of plate-adhering cells, were collected by pipette and cultured in the DMEM/F12 medium containing 15% FBS, 1,500 units/ml LIF (ESGRO, Chemicon, Billerica, MA, USA), 0.5 IU/ml FSH, 10 ng/ml EGF (Invitrogen, Carlsbad, CA, USA), B27 (GIBCO), and ITS supplement (GIBCO).
SSCs were isolated from 8-day old OG2 mice using two-step enzymatic digestion followed by MACS (Miltenyi Biotech, BergischGladbach, Germany) using GFRa1 antibody, a goat anti-mouse antibody recognizing the C-terminus of the GFRa1 receptor (Santa Cruz Biotechnology, Santa Cruz, CA, USA), with a 1:200 dilution. Isolated SSCs were cultured in gelatin-coated 6-well plates with KO-DMEM medium containing 1% FBS,1,500 units/ml LIF, and SU5402 (2 μM, Calbiochem, La Jolla, CA, USA) plus CHIR99021(3 μM, Axon Medchem, Groningen, Netherland)-2i, for one week; After GFP positive cells appeared, DMEM/F-12 medium containing 15% FBS, 1,500 units/ml LIF, 100 microM 2-mercaptoethanol (GIBCO), FSH (0.5 IU/ml), EGF (10 ng/ml), B27, and ITS supplement was used for further culture.
For PCR and RT-PCR, we used the primers listed in Additional file 5: Table S1. E 3.0 Embryos were collected from pregnant mice for analysis of the expression of Trim43a and Hmgpi. For immunofluorescence, the following antibodies were used: GFRa1 antibody (goat polyclonal ,Santa Cruz Biotechnology); PLZF antibody (rabbit polyclonal, Abcam, Cambridge, MA, USA); Anti-p57 Kip2 (Cell Signaling, Inc. Danvers, MA, USA); Nobox antibody (rabbit polyclonal, Abcam); Stella antibody (rabbit polyclonal, Abcam); Mos antibody (rabbit polyclonal, Santa Cruz Biotechnology); Nanog antibody (rabbit polyclonal, Abcam); Gamma-tubulin antibody (mouse monoclonal, Sigma-Aldrich); We utilized specific markers for nuclear envelope (lamin B1) and nucleic acids (YoYo1) to demonstrate the presence of polar bodies in oocytes, Lamin B1 antibody(1:100, Rabbit polyclonal, Abcam), and YOYO-1 (Molecular Probes, YOYO®-1 Iodide (491/509)), diluted in phosphate-buffered saline to make 2.4nM to use.
SSC-Oocs were rinsed thoroughly and kept in Hepes-CZB in original culture dish before injection. Adult OG2 mice were used as the donor for Oct4/GFP-carrying sperm. To retrieve sperm, seminiferous tubules were collected and put in Hepes-CZB. They were then cut into small pieces with a pair of fine scissors. A drop of the medium with tubule fragments was mixed with the same volume of Hepes-CZB containing 12% (w: v) polyvinylpyrrolidone and pipetted vigorously to release spermatozoa. Sperm were collected and injected into SSC-Oocs. Injected oocytes were activated by 30-min treatment with Ca2+-free CZB containing 5 mM SrCl2. Embryos were cultured in KSOM at 37°C in 5% CO2.
Genomic DNA was isolated from SSCs and SSC-Oocs. Bisulfite conversion was performed on a thermocycler using the QiagenEpiTect Kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions, with two additional cycles (5 min at 99°C and 3 h at 60°C) at the end. Converted DNA was eluted in 40 μl of elution buffer, and a 5-μl DNA sample was then amplified with the following primer sets: Snrpn, AATTTGTGTGATGTTTGTAATTATTTGG and ATAAAATACACTTTCACTACTAAAA TCCACAA; Igf2-H19-DMR, GGAATATTTGTGTTTTTGGAGGG and TTAAACCCCAACCTCTACTTTT ATAAC; Dlk1-Meg3/Gtl2-DMR, GGTTTGGTATATATGGATGT ATTGTAATATAGG and ATAAAACACCAAATCTATACCAAAATATACC. PCR was performed in 25-μl reactions using 2.5 units of ExTaq under the following conditions (38 cycles): 1) 96°C for 15 seconds, 60°C for 30 seconds, and 72°C for 30 seconds for Snrpn; 2) 96°C for 15 seconds (hot start), 55°C for 30 seconds, and 72°C for 1 minute for Igf2-H19-DMR and Dlk1-Meg3/Gtl2-DMR. The amplified fragments were cloned into the pMD19-T vector (TaKaRa Biotech Co., Ltd) and then sequenced.
All experiments were performed 4 times, and data were expressed as means ± SE and analyzed by one-way ANOVA analysis. A value of P < 0.05 was considered significant.
We thank Jinsong Li of Institute of Biochemistry and Cell Biology/Shanghai Institutes for Biological Sciences/Chinese Academy of Sciences for providing OG2 mice. We thank Haifan Lin of Yale University for their critical and useful comments on the manuscript.
This work was partially supported by grants from the National Key Basic Research and Development Program of China (2009CB941103) and from the National Natural Science Foundation of China (30870956) awarded to Dr. Feng; from the National Natural Science Foundation of China (81000049) awarded to Dr. Yu; and from the 211 program of Shanghai Jiao Tong University School of Medicine and Shanghai Leading Academic Discipline Project (S30201).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.