SEARCH

SEARCH BY CITATION

Keywords:

  • Germline;
  • Pluripotent stem cells;
  • Reprogramming;
  • Microarray

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Germ cells are similar to pluripotent stem cells in terms of gene expression patterns and the capacity to convert to pluripotent stem cells in culture. The factors involved in germ cell development are also able to reprogram somatic cells. This suggests that germ cells are useful tools for investigating the mechanisms responsible for somatic cell reprograming. In this study, the expression of reprograming factors in primordial germ cells (PGCs) was analyzed. PGCs expressed Oct3/4, Sox2, and c-Myc but not Klf4. However, Klf2, Klf5, Essrb, or Essrg, which were expressed in PGCs, could compensate for Klf4 during somatic cell reprograming. Furthermore, PGCs could be converted to a pluripotent state by infection with any of the known reprogramming factors (Oct3/4, Sox2, Klf4, and c-Myc). These cells were designated as multipotent PGCs (mPGCs). Contrary to differences in the origins of somatic cells in somatic cell reprogramming, we hypothesized that the gene expression levels of the reprogramming factors would vary in mPGCs. Candidate genes involved in the regulation of tumorigenicity and/or reprogramming efficiency were identified by comparing the gene expression profiles of mPGCs generated by the exogenous expression of c-Myc or L-Myc. STEM CELLS2013;31:479–487


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Yamanaka and colleagues first showed that somatic cell reprogramming could be accomplished by the expression of four defined factors [1]. When the transcription factors Oct3/4, Sox2, Klf4, and c-Myc were introduced into mouse fibroblasts, the cells converted to a pluripotent embryonic stem cell (ESC)-like state. This induced pluripotent stem (iPS) cell technology has since been applied to human somatic cells [2]. Because the generation of iPS cells circumvents issues regarding ethical concerns and immune rejection, this technology holds great promise for use in regenerative medicine [3]. During reprogramming, gene expression in somatic cells is repressed, and pluripotency-associated genes are activated [4]. However, the mechanisms that govern the reprogramming process remain unclear. Reprogrammed iPS cells are heterogeneous, and not all cells convert to iPS cells when the four factors are expressed [5, 6]. Therefore, it is important to investigate the mechanisms underlying somatic cell reprogramming in more detail.

Of the four reprogramming factors, c-Myc plays an important role in somatic cell gene repression, although it is dispensable for reprogramming [7]. c-Myc also contributes to the reprogramming process by affecting cell cycle progression. However, c-Myc not only enhances the reprogramming of iPS cells but also increases their tumorigenicity. It was recently reported that the Myc family member, L-Myc, enhances reprogramming efficiency but not tumorigenicity [8]. The tumorigenicity of iPS cells is a major consideration for their clinical application. Therefore, it is important to understand the differences between the effects of c-Myc and L-Myc during somatic cell reprogramming.

Germ cells are unipotent cells that differentiate only into gametes and acquire totipotency through fertilization. Thus, germ cells are the only cells that undergo reprogramming under physiological conditions. However, germ cells can be induced to adopt a pluripotent state in vitro; therefore, they have the potential to become pluripotent [9, 10]. It is conceivable that similarities exist between germ cell development and somatic cell reprogramming. We showed previously that somatic cells can be reprogrammed via the induction of factors that are important for the acquisition of germ cell characteristics [11]. This suggests that germ cells may be a useful tool for studying the mechanisms involved in the acquisition of pluripotency in somatic cells.

Here, we analyzed the endogenous expression of reprogramming factors in primordial germ cells (PGCs). PGCs expressed Oct3/4, Sox2, and c-Myc but not Klf4. However, PGCs expressed Klf2, Klf5, Essrb, and Essrg, each of which could compensate for Klf4 during somatic cell reprogramming. In addition, PGCs could be converted to pluripotent cells by exogenous expression of any of the four known reprogramming factors. These results demonstrate that the expression level of reprogramming factors is important for the acquisition of pluripotency. Finally, we identified candidate genes that may regulate tumorigenicity and/or reprogramming efficiency.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Mice

The Nanog-green fluorescent protien (GFP) internal ribosome entry sites-(IRES)-puro transgenic mouse strain (RBRC02290) was described previously [11, 12]. ICR, C57BL/6, and BALB/c nude mice were purchased from Japan SLC (Shizuoka, Japan). Animal care was performed in accordance with the guidelines established by Keio University for animal and recombinant DNA experimentation. PGCs were prepared by crossing Nanog-GFP mice with ICR mice. Nanog-GFP ESCs were generated by crossing TG mice with C57BL/6 mice.

Plasmids

Retroviral plasmids for iPS cell induction were described previously [11]. pMXs-L-Myc was obtained from Addgene (plasmid 26023). To detect virus integration, probes were generated against the sequence of the pMXs backbone with the following primers: 5′-gtttgggaccgaagccgcgccgcg-3′ and 5′-ccatataagatctcatatggccat-3′. The amplified fragment was cloned into the pGEM-T easy plasmid (Promega, Madison, WI, http://www.promega.com).

Isolation and Culture of PGCs

Based on GFP fluorescence, PGCs were obtained from Nanog-GFP-IRES-puro transgenic embryos at E11.5 using a FACS Vantage and FACS AriaII Cell Sorter (BD Biosciences, San Diego, CA, http://www.bdbiosciences.com). Sorted PGCs were cultured on Sl/Sl4-m220 feeder cells in Knockout Dulbecco's modified Eagle's medium (DMEM) supplemented with 15% serum, 2 mM glutamine, 1 mM nonessential amino acids, and 2-mercaptoethanol (2-ME) in the presence of leukemia inhibitory factor (LIF). Sl/Sl4-m220 cells were prepared 1-day before use by treatment with 5 μg/ml mitomycin C (MMC) for 1 hour and plated at a density of 4 × 105 cells per 1.9 cm2. Virus infection was performed at the beginning of the culture period, with a multiplicity of infection of 10.408 ± 0.5785 (mean ± SD). After 48 hours, the cells were collected and reseeded on STO feeder cells. Established multipotent PGCs (mPGCs) were cultured on MMC-treated STO cells in the same medium. mPGCs were separated from STO cells by sorting based on GFP fluorescence. Collected cells were used for RNA analysis. For 2i + LIF culture, 1 μM mitogen-activated protien kinase/extracellular signal-regulated kinase kinase (MEK) inhibitor (PD325901) and 3 μM glycogen synthase kinase (GSK)-3β inhibitor (CHIR99021) were added to knockout DMEM supplemented with 15% knockout serum replacement (KSR), 2 mM glutamine, 1 mM nonessential amino acids, and 2-ME in the presence of LIF. For culture in basic fibroblast growth factor (bFGF) + Activin, 12 ng/ml bFGF and 2 ng/ml Activin were added to knockout DMEM/F12 supplemented with N2 and B27, in the presence of 20% KSR, 2 mM glutamine, 1 mM nonessential amino acids, 1 mM sodium pyruvate, and 2-ME.

Teratoma Formation

One million mPGCs were suspended in BD Matrigel (BD Biosciences) and injected subcutaneously into nude mice. Three to four weeks later, tumors were removed and fixed in phosphate buffered saline (PBS) containing 4% paraformaldehyde (PFA), embedded in paraffin, sectioned, and stained with hematoxylin and eosin.

Generation of Chimeric Mice

Blastocysts from an institute of cancer research (ICR) background were used to generate chimeric mice from mPGCs. mPGCs are mixed background of dilute brown non-agouti (DBA), C57BL/6, 129S4, and ICR. The chimerism was estimated from the coat color. The numbers of embryo used for the analysis and obtained chimeras were summarized in Supporting Information Table S1.

Gene Expression Analysis

Primers and probes for Oct3/4, Sox2, Klf4, and c-Myc, which were designed to distinguish between endogenous and viral transcripts, were described previously [13]. Transcript levels were determined using the 7500 Fast Real-Time polymerase chain reaction (PCR) system (Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com). Data represent the mean ± SD. Statistical significance was determined using Tukey's multiple comparison test.

Microarray Data Analysis

Expression profiles were analyzed using the 3D-Gene Mouse Oligo chip 24K (Toray Industries, Tokyo, Japan). Fluorescence intensities were detected using the Scan-Array Life Scanner (Perkin-Elmer, Waltham, MA), and photo-multiplier tube (PMT) levels were adjusted to achieve 0.1%–0.5% pixel saturation. Each TIFF image was analyzed using GenePix Pro (version 6.0) software (Molecular Devices, Sunnyvale, CA, http://www.moleculardevices.com). The data were filtered to remove low-confidence measurements and globally normalized per array such that the median signal intensity was set at 50. To analyze the gene expression levels, the 25,369 probes were collapsed into 14,538 genes with Entrez gene ID by selecting the probe with the highest intensity for each gene. Genes showing a twofold change in expression were categorized as being significantly upregulated or downregulated and were used for comparisons. To characterize the molecular backgrounds of these genes, enrichment analysis for Gene Ontology (GO) was performed by the GO Term Finder (http://go.princeton.edu/cgi-bin/GOTermFinder) [14]. Thereafter, the false discovery rate (FDR) was estimated using the Benjamini-Hochberg procedure. In this study, we selected enriched GO terms with a FDR <5%.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Endogenous Expression of Reprogramming Factors in PGCs

PGCs have the capacity to convert to a pluripotent state through controlled in vitro culture. Germ cells are closely related to pluripotent cells, however, they are unipotent and are committed to differentiating specifically into gametes. To examine the relationship between germ cell development and somatic cell reprogramming, the expression of reprogramming factors in PGCs was analyzed. PGCs strongly expressed the germ cell-specific gene, Mvh, on embryonic day (E) 11.5, but not the ESC-specific genes, ECAT1 and Eras (Fig. 1A–1C). However, both ESCs and PGCs expressed a high level of the pluripotency-associated gene, Nanog (Fig. 1D). Interestingly, PGCs expressed Oct3/4, Sox2, and c-Myc but not Klf4 (Fig. 1E–1H). Other members of the Klf family (Klf2 and Klf5) and the Esrr family (Esrrb and Esrrg) can participate in the reprogramming of somatic cells instead of Klf4 [7, 15]. Therefore, the expression of these factors in PGCs was analyzed (Fig. 1I–1L). Although all four of these transcripts were detected in PGCs, indicating that PGCs contain the factors necessary to attain pluripotency, these cells maintain germ lineage restriction in vivo.

thumbnail image

Figure 1. Gene expression profile of E11.5 PGCs. (A–D): Gene expression of the germ cell marker, Mvh (A) and the pluripotent cell markers, ECAT1 (B), Eras (C), and Nanog (D). (E–L): Endogenous expression of the four factors capable of inducing induced pluripotent stem cells (E–H) and the four Klf4 compensation factors (I–L). The expression in E11.5 PGCs was compared with that in ESCs (TT2), MEFs, and STO cells. Relative gene expression is shown (mean ± SD; three independent experiments, n = 9 (A–D and I–L), or two independent experiments, n = 4 (E–H)). *, p < .01. Abbreviations: ESCs, embryonic stem cells; MEF, mouse embryonic fibroblasts; PGCs, primordial germ cells; STO.

Download figure to PowerPoint

Reprogramming of PGCs by Oct3/4, Sox2, Klf4, or C-Myc

To further investigate the relationship between germ cells and defined reprogramming factors in somatic cells, pluripotency was induced in PGCs using an established induction method (Supporting Information Fig. S1). First, PGCs were collected from the gonads of E11.5 Nanog-GFP transgenic embryos [12]. The PGCs were then cultured on membrane-bound SCF-expressing m220 feeder cells in the presence of LIF and infected with different combinations of the four iPS cell induction factors. The emergence of ESC-like packed colonies was assessed on Day 10 after infection. Because the PGCs were not cultured in the presence of bFGF, PGCs did not convert to pluripotent embryonic germ (EG) cells without the addition of induction factors [16]. These results are shown in Figure 2 and Supporting Information Figure S2. ESC-like colonies were observed when PGCs were infected with all four factors (Supporting Information Fig. S2). As observed previously for somatic cell reprogramming, c-Myc was dispensable for the reprogramming of PGCs (Supporting Information Fig. S2) [7]. On the other hand, unlike somatic cells, PGCs generated colonies following exogenous expression of any two of the other three factors (Supporting Information Fig. S2). Furthermore, the addition of even a single factor was sufficient for induction (Supporting Information Fig. S2; Fig. 2A). The typical morphology of an ESC-like colony is shown in Figure 2B, 2C. When 2,500 PGCs were seeded per dish, the mean number of EG colonies formed was 23 per dish (±5.71) in the presence of bFGF, whereas it was 4.81 per dish (±2.28) when a single factor was used. Therefore, approximately 4.78-fold fewer ESC-like colonies were formed following induction of PGCs with a single transcription factor than after conventional EG cell induction in the presence of bFGF.

thumbnail image

Figure 2. Induction and differentiation capacity of mPGCs. (A–C): Induction of mPGCs. (A): The number of ESC-like colonies observed 10 days after 2,500 PGCs were induced with the indicated factor. The results from four independent experiments are shown. (B, C): Typical morphology of established mPGCs observed by phase contrast microscopy (B) and Nanog-GFP fluorescence (C). Scale bar = 500 μM. (D–O): Teratoma formation by mPGCs induced with a single factor. D–F, Klf4; G–I, Sox2; J–L, Oct3/4; M–O, c-Myc. Ectoderm (D, G, J, M), mesoderm (E, H, K, N), and endoderm (F, I, L, O) are shown. Scale bar = 100 μM. Abbreviations: bFGF, basic fibroblast growth factor; ESCs, embryonic stem cells; mPGC, multipotent PGCs, PGCs, primordial germ cells.

Download figure to PowerPoint

Characterization of mPGCs Derived from PGCs Induced with a Single Factor

Next, the pluripotency of the induced cells was examined at the molecular and functional levels. To confirm that mPGCs were able to differentiate, their capacity to form teratomas was assessed. One million mPGCs, which were induced from PGCs by any one of the four transcription factors, were injected subcutaneously into nude mice. Teratomas containing derivatives from all three germ layers were formed in all cases (Fig. 2D–2O). These results demonstrated that any one of the transcription factors could convert PGCs to a pluripotent state. To further confirm the pluripotency of mPGCs, their capacity to form embryoid bodies (EB) was analyzed in vitro (Supporting Information Fig. S3). When pluripotent stem cells were cultured under nonadherent conditions, they formed EBs, whereas PGCs did not. Furthermore, we compared the culture conditions for mPGCs with those of epiblast stem cells (EpiSCs), which are also able to form three germ layers [17]. Whereas EpiSCs are maintained in the presence of Activin and bFGF, ESCs require LIF (but not bFGF) [18]. mPGCs cultured in the presence of LIF and 2i (an inhibitor of MEK and GSK-3β) showed a typical dome-shaped morphology (Supporting Information Fig. S4). mPGCs did not form colonies when cultured in the presence of Activin and bFGF (Supporting Information Fig. S4). We next performed chimeric analysis of mPGCs and found that chimeric mice could be generated using mPGCs (Supporting Information Fig. S5 and Supporting Information Table S1). Taken together, these results indicate that mPGCs are ESC-like, not EpiSC-like, pluripotent cells. Although we demonstrated that mPGCs have chimera forming activity, germline transmission of mPGCs might be low for the following reason. It is reported that pluripotent cells established from PGCs at E12.5 does not have the capacity of germline transmission through chimera [19]. The reason is considered as due to the erasure of imprinting. In addition, even in the earlier stage, PGC-derived pluripotent cells lost parental imprinting patterns according to in vitro culture [20].

The gene expression patterns of mPGCs were then compared with those of parental PGCs from E11.5 embryos. PGCs expressed the germ cell-specific markers, Mvh and Blimp-1, whereas mPGCs did not (Fig. 3A, 3B) [21, 22]. However, mPGCs expressed the ESC-specific markers, ECAT-1 and Eras (Fig. 3C, 3D). Furthermore, mPGCs expressed Oct3/4, Sox2, Klf4, and Nanog, which are involved in the core transcription network required to maintain pluripotency (Fig. 3E–3H). Expression of Nanog was also analyzed by examining the promoter activity of Nanog-GFP. According to fluorescence-activated cell sorting (FACS) analysis, similar levels of GFP were expressed in mPGCs and PGCs (Supporting Information Fig. S6).

thumbnail image

Figure 3. Gene expression in mPGCs. (A, B): Gene expression profiles of germ cell markers in mPGCs compared with those of parental E11.5 PGCs. Relative expression of the Mvh (A) and Blimp-1 (B) genes in mPGCs. (C–H): Gene expression profiles of pluripotent cell markers in mPGCs compared with those in ESCs (TT2). Relative expression of the ECAT1 (C), Eras (D), Klf4 (E), Sox2 (F), Oct3/4 (G), and Nanog (H) genes in mPGCs. Factors used to induce mPGCs (shown along the horizontal axis) are designated as follows: O, Oct3/4; K, Klf4; S, Sox2; M, c-Myc. (mean ± SD; three independent experiments, n = 6). **, p < .01; *, p < .05. Abbreviations: ESCs, embryonic stem cells; mPGCs, multipotent primordial germ cells.

Download figure to PowerPoint

The integration of virus vectors into the genome of mPGCs was also examined using Southern blot analysis. Surprisingly, integrated retroviruses were not detected in the majority of clones, and when they were detected, the copy number was one (Supporting Information Fig. S7). Integrated clones were detected following expression of Klf4 and Oct3/4, or Sox2 and Oct3/4. To analyze whether virus mRNAs were expressed or not, we collected pluripotent candidate cells based on Nanog-GFP fluorescence and SSEA-1 expression [23]. Virus mRNAs were detected with sets of virus-specific primers and probes [13] (Supporting Information Fig. S8). Furthermore, when PGCs were infected with a DsRed expression virus, DsRed fluorescence was detected in Nanog-GFP- and SSEA-1-positive pluripotent candidate cells (Supporting Information Fig. S9). Therefore, transient expression of the transcription factor appears to be sufficient to generate mPGCs.

Each of the Factors Induces Klf4 Expression and Suppresses Expression of Germ Cell Markers

To examine the dynamics of the induction of mPGCs, we analyzed gene expression patterns 2 and 3 days after infection of PGCs with a single transcription factor. As a control, we analyzed EG cells cultured under standard conditions in the presence of bFGF. To purify pluripotent candidate cells, we sorted the cells based on Nanog-GFP fluorescence and SSEA-1 expression [23]. Under control conditions, the germ cell markers, Blimp-1 and Mvh, were downregulated (Fig. 4A, 4B). Because Blimp-1 is not expressed in pluripotent cells, and ectopic expression of Blimp-1 in ESCs induces growth retardation, it is conceivable that suppression of characteristic germ cell genes is important for cells to become pluripotent [11]. Blimp-1 was also downregulated following induction of PGCs with each reprogramming factor (Fig. 4A). However, the repression of Blimp-1 was weaker than that observed in the presence of bFGF, especially at Day 3. The pluripotent cell marker, ECAT1, was upregulated in control cells and following induction with each reprogramming factor (Fig. 4C). Interestingly, Klf4 was upregulated in all cells induced with a single factor to a similar level to that observed in EG cells cultured in the presence of bFGF (Fig. 4D). The introduction of Klf4 in EpiSCs induces their conversion to pluripotent cells [24]. Since EpiSCs and PGCs both express Oct3/4, Sox2, and Nanog, upregulation of Klf4 might be necessary for the induction of pluripotency in PGCs.

thumbnail image

Figure 4. Gene expression in PGCs at 2 and 3 days after infection with various factors. (A–D): Blimp-1 (A), Mvh (B), ECAT1 (C), and Klf4 (D) expression in PGCs at 2 and 3 days after infection with various factors compared with their expression in parental E11.5 PGCs and bFGF-treated embryonic germ cells. Cells were sorted by fluorescence-activated cell sorting based on the expression of Nanog-GFP and SSEA-1. Relative gene expression is shown (mean ± SD; three independent experiments, n = 6). **, p < .01; *, p < .05. Abbreviations: bFGF, basic fibroblast growth factor; PGCs, primordial germ cells.

Download figure to PowerPoint

Microarray Analysis of mPGCs Induced with a Single Factor

During somatic cell reprogramming of iPS cells, four specific factors induce many different types of cells to become pluripotent, and it has been reported that iPS cells retain the characteristics of the original cell [25]. PGCs infected with any of the four reprogramming factors were converted to mPGCs. Since iPS cells and mPGCs are both pluripotent stem cells, contrary to differences between the origins of somatic cells in iPS cells, mPGCs could be used to reveal how each reprogramming factor contributes to somatic cell reprogramming. To examine the contribution of each factor, transcriptome analysis of mPGCs induced with a single reprogramming factor was performed using DNA microarray analysis. Whole transcription patterns of mPGCs were compared with those of parental PGCs, mouse embryonic fibroblasts (MEFs), and other pluripotent cells such as ESCs and iPS cells.

First, cells were categorized according to their gene expression pattern (Fig. 5A). In this analysis, mPGCs were categorized in the same cluster as pluripotent stem cells such as ESCs and iPS cells but were not in the same cluster as PGCs and MEFs. This indicated that mPGCs have a similar gene expression profile to those of known pluripotent stem cells, irrespective of the factor that was used to induce their pluripotency.

thumbnail image

Figure 5. Microarray analysis of mPGCs. (A): Standardized Euclidean distance is shown according to the gene expression profiles of mPGCs, PGCs, pluripotent cells, and MEFs. Embryonic germ (EG) indicates pluripotent cells derived from PGCs by basic fibroblast growth factor. Nanog-ES indicates embryonic stem cells (ESCs) derived from Nanog-GFP transgenic blastocysts. MEF3F 3.33-9 and 3F 3.332-9 represent induced pluripotent stem (iPS) cells derived from MEFs induced with three factors, while MEF4F 4.47-9 represents iPS cells derived from MEFs induced with four factors. (B, C): Comparison of gene expression profiles of mPGCs induced with a single factor. The gene expression profiles of mPGCs induced by Klf4, Sox2, Oct3/4, c-Myc, or L-Myc were compared with that of EG cells, and the number of upregulated (B) and downregulated (C) genes are shown in a Venn diagram. The genes in each cluster are listed in Supporting Information Table S2. (D, E): The gene expression profiles of mPGCs induced by c-Myc and L-Myc were compared to iPSCs induced by L-Myc from hDFs (Nakagawa et al.,7). The numbers of upregulated (D) and downregulated (E) genes in mPGCs and hDFs induced with L-Myc (compared with those in c-Myc-induced cells) are shown in a Venn diagram. The common genes in each cluster are listed in Tables 1 and 2, and a complete list of the genes in the Venn diagrams is presented in Supporting Information Table S3. Abbreviations: hDFs, human dermal fibroblasts; mPGCs, multipotent primordial germ cells.

Download figure to PowerPoint

For identification of the differences between each induction factor, we tried to include the cells generated by L-Myc. It is known that c-Myc and L-Myc are closely related with regard to tumorigenecity and reprogramming efficiency [8]. To analyze the characteristics of L-Myc in reprogramming, mPGCs were induced with L-Myc. mPGCs induced by L-Myc showed a typical colony morphology, formed EBs, and showed a gene expression pattern similar to that of pluripotent cells (but not PGCs) (Supporting Information Figs. S10 and S11). To characterize the contribution of each factor to the induction of pluripotency, the gene expression profiles of mPGCs induced with each factor were compared with the gene expression profile of EG cells, which are pluripotent stem cells induced from PGCs without exogenous transgene expression. The number of upregulated and downregulated genes are shown in Figure 5B, 5C, and these genes are summarized in Supporting Information Table S2. mPGCs induced by c-Myc and L-Myc had significantly more upregulated and downregulated genes than mPGCs induced by the other reprogramming factors. This suggests that Myc family proteins play a unique role in the acquisition of pluripotency. Next, GO analysis was performed (Supporting information Table S4). Genes showing a statistically significant change in expression are presented in Supporting Information Table S5. No genes showing a statistically significant alteration in Klf4-induced mPGCs. Interestingly, L-Myc-induced mPGCs had many more genes that showed a statistically significant alteration in expression than mPGCs induced with any of the other factors. GO analysis of L-Myc-induced mPGCs showed that genes related to DNA repair and the DNA damage response were significantly upregulated. The generation of iPS cells requires multiple cell divisions, which increases the risk of mutation [13]. The low tumorigenicity of L-Myc-induced iPS cells may be due to upregulation of these DNA protective factors.

Finally, the gene expression profiles of mPGCs induced with c-Myc or L-Myc were compared. The upregulated and downregulated genes in these mPGCs are listed in Supporting Information Table S3. In mPGCs induced with L-Myc, 2,199 genes were upregulated and 783 genes were downregulated compared with their levels in mPGCs induced with c-Myc. These data were compared with previously reported [7] data obtained from human dermal fibroblasts (hDFs) (Fig. 5). Myc-induced mPGCs and hDF cells had 22 upregulated genes and 5 downregulated genes in common (Tables 1 and 2). These genes represent potential regulators of tumorigenicity and/or reprogramming efficiency.

Table 1. Upregulated genes in L-Myc-induced multipoint priordial germ cells and L-Myc-induced induced pluripotent stem cells (human dermal fibroblasts)
inline image
Table 2. Downregulated genes in L-Myc-induced multipotent primordial germ cells and L-Myc-induced induced pluripitent stem cells (human dermal fibroblasts)
inline image

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

To gain new insights into the reprogramming process, the mechanisms by which PGCs are converted to pluripotent cells by a single reprogramming factor must be better understood. We demonstrated that PGCs can be converted to a pluripotent state by expression of any one of the known reprogramming factors Oct3/4, Sox2, Klf4, c-Myc, and L-Myc. Although PGCs expressed Oct3/4, Sox2, and c-Myc endogenously, they did not express Klf4. Sox2 is dispensable for the reprogramming process in neural stem cells due to functional compensation by other reprogramming transcription factors [26]. Similarly, other endogenous factors were expected to compensate for the absence of Klf4 in PGCs. Although PGCs endogenously express Oct3/4, Sox2, and c-Myc, they required the exogenous expression of any one of these factors to attain pluripotency. This indicates that the expression levels of the reprogramming factors are critical for cells to become pluripotent. c-Myc induces proliferation of somatic cells. Originally, pluripotent cells derived from PGCs (EG cells) were established through the search for growth induction factors [9]. Therefore, a similar proliferative signal may induce PGCs to become pluripotent. Sox2 is important for ESC pluripotency, as it is involved in the maintenance of Oct3/4 expression and overexpression of Oct3/4 can rescue Sox2 depletion in ESCs [27]. Control of Oct3/4 expression is essential for maintaining ESCs in an undifferentiated state, and both overexpression and downregulation of Oct3/4 can induce ESC differentiation [28]. It is likely that tight regulation of Oct3/4 expression controls the acquisition of pluripotency in PGCs. Forced activation of Oct3/4 facilitates somatic cell reprogramming [29, 30].

Southern blot and quantitative PCR (qPCR) analyses revealed that transient expression of reprogramming factors is sufficient to generate mPGCs (Supporting Information Figs. S6–S8). Transient expression increased the total amount of these factors (Supporting Information Fig. S12; Fig. 4). It is probable that reprogramming factors are critical during the early phase of mPGC induction. Consistent with this, bFGF signaling is necessary during the first 24 hours of conventional culture of EG cells [16]. Moreover, transient expression of Zscan4 improves the reprogramming efficiency of somatic cells [31]. Thus, it is important to analyze gene expression during the early stages of the induction of pluripotency.

Although PGCs endogenously express all of the Yamanaka reprogramming factors except Klf4, they are committed to a germ cell lineage in vivo. The mechanisms by which PGCs maintain their germ cell characteristics are still unclear, however, this may depend on factors that are exclusively expressed in PGCs. PGC marker genes, such as Blimp-1 and Mvh, were downregulated in pluripotent cells (Fig. 3A, 3B). Furthermore, overexpression of Blimp-1 in ESCs results in growth retardation [11]. This implies that germ cell-specific factors repress not only somatic cell programming but also the acquisition of pluripotency. In future, it will be important to elucidate the function of germ cell-specific genes such as Blimp-1 in pluripotency.

The same factors induce various somatic cells to become iPS cells, and the iPS cells generated retain the characteristics of the somatic cells from which they were derived. By contrast, mPGCs were induced by different reprogramming factors from PGCs of a single origin. Transcriptome analysis of mPGCs induced by each factor showed similar gene expression profiles to those of pluripotent stem cells. However, more detailed analysis revealed differences in gene expression among mPGCs induced by each of the factors. When specific upregulated and downregulated genes were identified, differences in gene expression were particularly evident between c-Myc- and L-Myc-induced mPGCs. Comparison of our data with previously reported data from hDFs [8] identified several candidate genes that may regulate tumorigenicity and/or reprogramming efficiency. Although this analysis may not provide significant mechanistic insights, the data provide valuable information as a database. Because pluripotent stem cells are a heterogeneous population [32, 33], humoral factors are thought to be critical for regulating tumorigenicity. One of the genes listed in Table 1 is the humoral factor, Rspo1. Rspo1 activates and sustains the Wnt signaling pathway [34], which promotes somatic cells to become pluripotent and maintain a pluripotent state [35, 36]. Although the function of Rspo1 in pluripotent stem cells is not fully understood, Rspo1 regulates the undifferentiated state of intestinal stem cells through the target gene, Lgr5 [37]. It will be interesting to analyze the function of Rspo1 in pluripotent stem cells in future studies.

To further understand the mechanisms underlying reprogramming, it will be necessary to analyze sequential gene expression patterns during the conversion of PGCs to pluripotent cells at the clonal level. Germ cells present a good model for analyzing sequential gene expression patterns during the conversion of PGCs into mPGCs and may also provide insight into how PGCs maintain germ cell lineage restriction in vivo.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We showed that PGCs endogenously expressed Oct3/4, Sox2, and c-Myc but not Klf4. We showed that PGCs expressed Klf2, Klf5, Esrrb, and Esrrg, each of which could compensate for the absence of Klf4 during reprogramming. Furthermore, PGCs acquired pluripotency following infection with any of the four known reprogramming factors. Although PGCs expressed these reprogramming factors endogenously, PGCs required exogenous expression of any one of the factors for pluripotency. This indicates that the expression levels of the reprogramming factors are critical for cells to become pluripotent. Finally, we identified putative candidate genes for regulating tumorigenicity and/or reprogramming efficiency.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We thank Dr. T. Kitamura for the Plat-E cells; N. Yamasaki for culturing the mPGCs and A. Nagamachi for transplantation of mPGCs for chimeric analysis; N. Tago for cell sorting and M. Dahl for proofreading this paper. This study was supported by PRESTO of the Japan Science and Technology Agency and Scientific Research (C) and by a grant from the Project for Realization of Regenerative Medicine. Support for the core institutes for iPS cell research was provided by MEXT, a Grant-in-Aid for the Global century COE program from MEXT to Keio University and the Keio University Medical Science Fund.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
sc-12-0279_sm_SupplFigure1.tif614KSupplementary Figure S1. The culture method used to induce iPS cells from PGCs. The gonads of E11.5 Nanog-GFP transgenic embryos were collected and single cell suspensions were generated by trypsin treatment. PGCs were isolated by FACS sorting based on GFP fluorescence. The collected PGCs were seeded on SCF-expressing m220 feeder cells and infected with viruses containing various combinations of iPS cell induction factors.
sc-12-0279_sm_SupplFigure2.tif276KSupplementary Figure S2. Induction of mPGCs by factors used to induce iPS cells. The number of ES cell-like colonies observed 10 days after 2,500 PGCs were induced with the indicated factor(s). The results from two independent experiments are shown. Factors used to induce mPGCs (shown along the horizontal axis) are designated as follows: O, Oct3/4; K, Klf4; S, Sox2; M, c-Myc.
sc-12-0279_sm_SupplFigure3.pdf47KSupplementary Figure S3. EB formation from mPGCs. EBs generated from mPGCs induced with the indicated factors under non-adherent culture conditions.
sc-12-0279_sm_SupplFigure4.pdf89KSupplementary Figure S4. Morphology of mPGCs cultured in 2i +LIF or bFGF + Activin. (Upper) mPGCs induced with the indicated factors and cultured in the presence of 2i (MEK inhibitor and GSK-3β inhibitor) + LIF or (lower) bFGF + Activin, observed by phase contrast microscopy. Scale bars, 100μM.
sc-12-0279_sm_SupplFigure5.tif1157KSupplementary Figure S5. Generation of chimeric mice from mPGCs. mPGCs were injected into blastocysts of ICR mice. The contribution of chimeric mice was estimated from the coat color of the new-born mice. Arrowhead indicated a colored coat, which were due to mPGCs.
sc-12-0279_sm_SupplFigure6.tif396KSupplementary Figure S6. Nanog-GFP expression in mPGCs. GFP expression driven by the Nanog promoter in mPGCs induced with various factors was compared with that in E11.5 PGCs.
sc-12-0279_sm_SupplFigure7.tif2080KSupplementary Figure S7. Southern hybridization of mPGCs. Southern hybridization to monitor virus integration in mPGCs infected with viruses containing various induction factors. Genomic DNA was digested with EcoRV (a) or XhoI (b) and detected with a virus vector-specific probe. Factors used to induce mPGCs are designated as follows: O, Oct3/4; K, Klf4; S, Sox2; M, c-Myc. 4F iPS cells (TTF) induced from mouse tail tip fibroblasts by all four factors were used as a positive control.
sc-12-0279_sm_SupplFigure8.tif322KSupplementary Figure S8 Expression of virally-introduced factors in pluripotent candidate cells. PGCs were infected with viruses containing the indicated factors. Two days after infection, the pluripotent candidate cells were purified based on expression of Nanog-GFP and SSEA-1. The expression of virally-introduced factors were analyzed using specific sets of primers and probes (mean ± SD; three independent experiments, n = 6). **P<0.01.
sc-12-0279_sm_SupplFigure9.tif297KSupplementary Figure S9 Fluorescence of pluripotent candidate cells following infection with a virus expressing DsRed. PGCs were infected with a virus containing DsRed-Express. Two days after infection, the pluripotent candidate cells were purified based on expression of Nanog-GFP and SSEA-1. DsRed fluorescence was analyzed by flow cytometry.
sc-12-0279_sm_SupplFigure10.pdf78KSupplementary Figure S10. Induction of mPGCs with L-Myc. a. The number of ES cell-like colonies observed 10 days after infection of 2,500 PGCs with a virus containing L-Myc. The results from two independent experiments are shown. b, c. Typical morphology of established mPGCs observed by phase contrast microscopy (b) and Nanog-GFP fluorescence (c). Scale bars, 100μM. d, e. Gene expression profiles of germ cell markers in mPGCs compared with those of parental E11.5 PGCs. Relative gene expression in mPGCs is shown for Mvh (d) and Blimp-1 (e). f-k. Gene expression profiles of pluripotent cell markers in mPGCs compared with those of ES cells (TT2). Relative gene expression in mPGCs is shown for ECAT1 (f), Eras (g), Klf4 (h), Sox2 (i), Oct3/4 (j), and Nanog (k).
sc-12-0279_sm_SupplFigure11.pdf68KSupplementary Figure S11. Differentiation capacity and Nanog-GFP expression of mPGCs induced with L-Myc. a-c. Teratoma formation by mPGCs induced with L- Myc. Ectoderm (a), mesoderm (b), and endoderm (c) are shown. Scale bars, 200μM. d. EBs formed from mPGCs induced by L-Myc under non-adherent culture conditions. e. GFP expression driven by the Nanog promoter in mPGCs generated by L-Myc was compared with that in E11.5 PGCs.
sc-12-0279_sm_SupplFigure12.tif469KSupplementary Figure S12. Total expression levels of the reprogramming factors in pluripotent candidate cells at 2 days after infection. a-c. Total expression levels (endogenous and virus-derived) of Sox2, Oct3/4, and c-Myc in PGCs at 2 days after infection with viruses containing various factors compared with their expression levels in parental E11.5 PGCs and bFGF-treated EG cells. Cells were sorted by FACS based on the expression of Nanog-GFP and SSEA-1. Relative gene expression is shown (mean ± SD; three independent experiments, n = 6). **P<0.01, *P<0.05.
sc-12-0279_sm_SupplTable1.pdf27KSupplementary Table 1
sc-12-0279_sm_SupplTable2.xls585KSupplementary Table 2
sc-12-0279_sm_SupplTable3.xls426KSupplementary Table 3
sc-12-0279_sm_SupplTable4.xls7452KSupplementary Table 4
sc-12-0279_sm_SupplTable5.pdf34KSupplementary Table 5
sc-12-0279_sm_SupplTable6.pdf28KSupplementary Table 6

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.