| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Mammalian Cellular Dynamics, and
2 Experimental Animal Division, RIKEN BioResource Center, Tsukuba, Japan
3 Department of Urology, Yokohama City University Graduate School of Medicine, Yokohama, Japan
4 Stem Cell Engineering, Institute for Frontier Medical Sciences, Kyoto University, Kyoto, Japan
5 Department of Bioinformatics and Genomes, Graduate School of Information Science, Nara Institute of Science and Technology, Nara, Japan
6 Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Our micro-array analysis showed that the analyzed samples could be classified into two groups: one consisting of all the ESCs and most of EGCs, and the other containing PGC samples, iPGC, one type of female EGC and GS cells. We then identified "signature" genes for the two groups, and used them to characterize GS cells, EGC, and iPGCs, and revealed developmental status of each cell type. The relationships between PGCs and stem cells derived from embryos or germ cells are discussed in light of these findings.
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Interestingly, a capacity for pluripotency appears to be retained in germ-line cells, as exemplified by the generation of EGCs from PGCs. Germ-line stem (GS) cells are spermatogonial cell lines, isolated from mouse testes. These cells produced normal sperm and offspring when they were transplanted into the seminiferous tubules of infertile mice, and unlike pluripotent ESCs, did not form teratomas in the tubules, suggesting that GS cells are committed to the germ cell lineage (Brinster 2002). However, on rare occasions, pluripotent ESC-like cells can be established from GS cell cultures, and are designated "multipotent GS cells" or mGS (Kanatsu-Shinohara 2004). Recently, it was reported that multipotent adult germ-line stem cells (maGSCs), another pluripotent cell line, can be isolated efficiently from the spermatogonia of adult mouse testes (Guan et al. 2006). Such pluripotent stem cells could never be derived from the culture of somatic, differentiated cells without some genetic manipulation (Takahashi & Yamanaka 2006). These facts suggest that, although germ cells are committed cells with a limited developmental potency, they can cross the developmental barrier to become ESC-like cells. Considering these similarities between PGCs and ESCs, Zwaka & Thomson (2005) proposed that ESCs may originate from early germ cells. Again, it is largely unknown how germ-line cells resemble embryo-derived stem cells at the molecular level. However, several independent groups have reported that ESCs can become PGC-like cells or germ-cell-like cells in vitro under the appropriate culture conditions (Hubner et al. 2003; Toyooka et al. 2003; Geijsen et al. 2004). For example, Toyooka et al. (2003) used knock-in ES cells, in which GFP or LacZ gene was expressed from the mouse Vasa homologue (Mvh) locus. As the Mvh is specifically expressed in differentiating germ cells after 11.5 dpc but not in ESC, MVH-positive cells are supposed to be committed to germ-line development. Indeed, such MVH-positive cells could be derived from ESC in vitro, and could participate in spermatogenesis when transplanted into testicular tubules, whereas transplanted ESC formed teratoma inside the tubules. However, the efficiency of ESC–germ cell conversion still remains low, and these "in vitro germ cells" do not seem to be completely functional gametes, implying that some important features intrinsic to germ cells may be lacking in these cells. Recently, Nayernia et al. (2006) reported the derivation of male gametes from ESCs in vitro, which gave rise to viable offspring, although the mice thus produced showed growth abnormalities and eventually died several months after birth.
To determine the relationships between the various categories of embryo-derived stem cells and their derivatives, such as in vitro germ cells and germ-line-derived stem cells, it is important to define the genome-wide expression differences among these cells. It is also important to include PGCs directly isolated from developing embryos in this assay. The expression profiles of PGCs should serve as reference data for these comparisons. Understanding the differences and common features between PGCs and stem cells should provide insight into the genetic programs required for germ-cell development and the maintenance of pluripotency. Particularly, comparison of in vitro-formed PGC (iPGC) with normal PGC will help in understanding developmental status of the iPGCs. Examination of the relationships among ESCs, EGCs and PGCs should also give us a clue to the molecular events underlying genomic reprogramming (Tada et al. 2001).
In this study, we performed the first genome-wide comparisons of the gene expression profiles of fluorescence-activated cell sorter (FACS)-purified PGCs and other stem cell lines. Here we used PGCs isolated from gonads of 11.5–13.5 dpc, because genomic reprogramming is supposed to occur in PGCs of these stages (Hajkova et al. 2002), and the Mvh-positive iPGC is thought to be equivalent to PGCs colonized the fetal gonad in vivo (Toyooka et al. 2003). Furthermore, EGCs analyzed here were derived from 12.5 dpc embryos (Tada et al. 1998). We defined developmental status of these cell types through gene expression analyses, and found that these samples can be divided into two distinct groups: one consisting of all the ESCs and male EGCs, and the other containing all PGC samples, in vitro-formed PGCs, and GS cells. Overall, the accumulated profiling data represent a useful measure for the classification of existing and newly isolated embryo- or germ-line-derived stem cells, and will facilitate the identification of the set of genes that characterizes germ-line development.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
We collected the gene expression profiles of three ESC lines developed by three independent groups, two batches of iPGCs (Toyooka et al. 2003) derived from E14tg2a (XY) ESC, two EGC lines, one GS cell line, and PGCs purified from both male and female mouse embryos of three developmental stages, that is, 11.5–13.5 dpc. We chose these gonadal PGCs for the analysis, because (i) genomic reprogramming is supposed to occur at these stages, (ii) Mvh, a selection marker for the iPGCs, is expressed in these PGCs, and (iii) EGCs analyzed in this study were derived from 12.5 dpc PGCs. Although expression differences can be found among the analyzed PGCs of different stages and different sexes, here we will mainly describe the common features shared by the gonadal PGCs from 11.5 to 13.5 dpc embryos. The gene expression in six different adult tissues was also analyzed. The expression data for these samples were converted to log ratios relative to a common reference RNA (Universal Mouse Reference RNA; Stratagene, La Jolla, CA) so that all the samples could be compared relative to the same reference. Details of the cells and tissues used in this study are described in Table S1 in Supplementary Material. All the log ratio data and P-values are presented in Supplementary Table S2.
To explore the similarities and differences in the expression profiles of all the samples described above, we first performed a pairwise comparison of the micro-array data, and calculated Pearson's correlation coefficients, as summarized in Table 1. The six PGC samples generally showed high mutual correlations, except for 11.5 dpc male PGCs. The correlation coefficients of the 11.5 male PGCs versus other PGCs ranged from 0.70 to 0.76, whereas the rest of the PGC samples had correlation coefficients of 0.855–0.95 when compared with each other. A small fraction of genes showed expression differences between male and female PGCs at these stages. The three independent ESC lines were similar to each other, with correlation coefficients ranging from 0.709 to 0.878. It is noteworthy that all three lines were established by different research groups and that the E14tg2a cells were cultured without feeder cells, whereas the R1 and TMA5 cells were grown on a mouse embryonic fibroblast feeder layer. This suggests that different culture conditions or ESC isolation procedures may not have major effects on their expression profiles. When compared with TMA5 ESCs, the PGCs showed correlation coefficients of 0.424–0.612. We examined 19 070 informative features, and identified up-regulated (> twofold) and down-regulated (< 50%) genes relative to those of TMA5 ESCs (Supplementary Table S3). For example, comparing TMA5 cells with 12.5 dpc female PGCs, we found that 1586 genes (8.3%) were up-regulated in the PGCs, whereas 2546 (13.4%) were down-regulated. In contrast, the expression profile of TMA55G, a male EGC line, showed strong similarity to that of the ESCs (correlation coefficient = 0.867), with only 162 (0.8%) genes up regulated and 272 (1.4%) down-regulated relative to their expression in the ESCs. In contrast, heart samples (correlation coefficient = 0.274) exhibited 1830 (9.6%) up-regulated genes and 2628 (13.8%) down-regulated genes. Therefore, the gonadal PGCs examined are significantly different from ESCs in terms of their global expression profiles.
|
We examined two batches of iPGCs. In vitro PGC–GFP and in vitro PGC–LacZ carry the GFP reporter and LacZ reporter, respectively, knocked into the Mvh (mouse Vasa homologue) gene locus. It is known that Mvh is expressed specifically in PGCs after 11.5 dpc, serving as a convenient marker for PGCs, gonocytes, and oocytes (Tanaka et al. 2000). Embryoid bodies were made with the knock-in ESCs under conditions described previously (Toyooka et al. 2003), and GFP- or LacZ-positive cells were isolated as iPGCs by FACS. The two iPGC samples showed somewhat different but still fairly similar expression profiles. These iPGCs were clearly different from the parental strain, E14tg2a, and resembled the normal PGCs, indicating that the iPGCs were indeed committed to the germ cell lineage. However, differences between the iPGCs and normal PGCs suggested that this commitment might be insufficient.
Differences in gene expression profiles between ESCs and PGCs
Pairwise comparisons of the ESCs and the PGCs indicated that the global expression profiles of these two cell types were significantly different, and we examined this point in detail. We selected 570 differentially expressed genes in the ESCs and PGCs from 12.5 to 13.5 dpc mouse embryos, based on Student's t-test, as described in the Methods section. The analysis also detected 463 genes similarly expressed in the ESC and PGCs. These genes were classified into five different categories (Table 2). The expression patterns of the representative genes in each class are shown in Fig. 1, and lists of the genes belonging to each class are presented as supplementary data (Supplementary Tables S4). In addition to P-value, we also estimated False Discovery Rate (FDR) for testing significance of differential expression. As presented in the Supplementary Table S4, results of the both tests essentially agreed in the most of the cases. Genes in class 1 (234 genes) were up-regulated in the ESC group, whereas the class 5 genes (130 genes) were expressed predominantly in the PGCs. The class 3 genes were expressed at similar levels in both the ESCs and the PGCs. Genes in classes 2 and 4 were expressed in both ESCs and PGCs; the expression levels of class 2 genes were elevated in ESCs, whereas class 4 genes were expressed more abundantly in PGCs. As demonstrated in Fig. 1, Klf4, Dnmt3 l, Upp1 and Tbx3 belong to class 1, and were over-expressed in all the ESCs tested. Asz1, Sin3b, lpo9 and Pes1 are examples of class 5 genes, which were over-expressed in the PGCs. The expression level of Pou5f1 (= Oct3/4) in PGCs was approximately half that in the ESCs (Fig. 1). The expression of Dppa3 (developmental pluripotency associated-3; alternatively called Stella/Pgc7) was elevated in the PGCs relative to its expression in the ESCs, whereas Zfp42 expression was much lower in the PGCs than in the ESCs, suggesting that not all the pluripotency-associated genes are regulated in the same way. This analysis demonstrated that 382 genes (classes 1 and 2) are significantly over-expressed in the ESCs, whereas 188 genes (classes 4 and 5) are up-regulated in the PGCs (Table 2), indicating that these two types of cells have considerably different transcriptional programs.
|
|
Real-time RT-PCR analysis was also carried out for analyzing expression of Klf4, Tbx3, Dazl and Asz1 (Supplementary Figure S2). Expression of the class 1 genes, Klf4 and Tbx3, was significantly higher in the ESCs than in the PGCs, while expression levels of Dazl (class 4) and Asz1 (class 5) were much higher in the PGCs. There results are consistent with the micro-array results shown in Fig. 1.
Multivariate comparison of gene expression profiles of embryo-derived stem cells and germ-line cells
Global gene expression patterns were compared from a different perspective using PCA, which reduces high-dimensional array data into a smaller number of principal components. PCA analysis was used to demonstrate similarities and differences in the transcriptional profiles of the samples analyzed. On a PCA plot, cell types with similar expression profiles can be positioned in proximity to each other (Sharov et al. 2003). We plotted the position of each cell sample against the PC2 and PC3 axes in two-dimensional space (Fig. 2A) or against the PC1, PC2, and PC3 axes in three-dimensional (3D) space (Fig. 2B). These three PCs accounted for about 70% of the variation present in the entire data set (PC1; 0.489, PC2; 0.142, PC3; 0.067; Cumulative proportion of these three PCs was 0.698). It was obvious that stem cells and germ cells can be classified into two groups located at two positions in this 3D space. One group consists of the ESCs, i.e., E14tg2a, TMA5, and R1 cell lines, and male EGC (TMA55G); the other includes all the PGCs, iPGCs, female EGC (TMA58G), and GS cells. The former is hereafter designated the "ES group" and the latter the "PGC group".
|
Another trend was that cells derived from embryos, that is, the ES and PGC groups, were separated from the cells of adult organs along the PC2 axis (Fig. 2A and B). The ordering along the PC2 axis, namely the ES group, the PGC group, testis, and other organs, seems to reflect differences in the developmental potential of each cell type. The PGC group and testis are rather close on the PC2 axis, probably because both express germ-cell-specific genes.
Since the male EGC, TMA55G, and the female EGC, TMA58G, belonged to the ES group and the PGC group, respectively, we next asked if this difference was due to difference in sex chromosome composition between the two cell lines. We newly isolated two female EGCs and two male EGCs from 12.5 dpc embryos as well as two female ESCs, and performed global expression profiling and PCA analysis. As shown in Supplementary Figure S2, all of the new cell lines were classified clearly into the ES group irrespective of their origin, sex chromosome composition or genetic background.
Identification of "signature" genes for the ES and PGC groups
Next, we tried to detect genes that define the characteristics of the ES and PGC groups based on the PCA results. Because the two groups are well separated along the PC3 axis, genes that make a large contribution to PC3 were sought using a loading scatter plot, shown as Fig. 2C. Because genes that have only marginal effects on the PC3 are clustered around the center of the plot, the genes plotted within a distance of less than 1 from the origin (gray dots in Fig. 2C) were eliminated from the analysis. After this data reduction, 799 of 20 281 (3.9%) genes remained. The 799 genes identified above were further categorized according to the positions of the ES group samples (blue dotted circle) and the PGC group samples (red dotted circles), shown in Fig. 2A. The ES group samples were located between lines at angles of 1.92 and 2.63 radians on the PCA plot (Fig. 2A), and genes mapped in the corresponding space (demarcated by blue lines in Fig. 2C) showed enhanced (but not exclusive) expression in the ES group (blue dots in Fig. 2C). Hereafter we call these genes showing enhanced expression in ESC as "signature" genes of the ES group. Similarly, genes plotted between 3.32 and 4.71 radians represent "PGC signature" genes (Fig. 2A and C, red dots). We identified 86 "ES signature" genes and 97 "PGC signature" genes. Genes located in the space between 2.63 and 3.32 radians were expressed similarly in both the ES and PGC groups. These 261 genes are designated "ES/PGC common" genes. Lists of the classified genes are presented in Supplementary Table S6.
We next assigned a gene ontology (GO) term to each gene of each group using GO slim, cut-down versions of GO <http://www.geneontology.org/GO.slims.shtmlgt;. Graphical representation of the "cellular component" GO terms assigned to each gene cluster is shown in Supplementary Fig. S3 (A–C). The ES signature and ES/PGC common graphs show very similar patterns. Both contain a number of genes with the term "nucleus", consisting of about 60% of the total (Fig. S3A,B,D). In contrast, the trend in the PGC signature genes is rather different (Fig. S3C). In the PGC signature genes, "nucleus" genes were reduced to 40%, whereas terms such as "cytoskeleton", "extracellular region", "cytosol", or "Golgi apparatus" are enriched instead. Individual signature genes with their GO terms are listed in Supplementary Tables S7–S9.
Comparison of the signature gene expression patterns of different cell samples by hierarchical clustering
To visualize the expression pattern of each signature gene in each cell sample, and to examine the relationships between EGCs, iPGCs and GS cells with ESCs or PGCs, the signature genes was hierarchically clustered. As shown in Fig. 3A, the ES signature genes showed higher expression in the ES group than in the PGC group, as expected. Furthermore, most of these signature genes had lower expression levels in adult organs, except the testis. The expression of some of the signature genes was hardly detectable in organs such as the heart, brain and liver. The ES signature genes were further classified into two subclusters. Subcluster-2 in Fig. 3A contained genes highly expressed in the ES group cells, such as Upp1, Zfp42 and Fbxo15. These genes generally showed much lower expression in the PGCs. It is noteworthy that differences in gene expression between iPGCs and the PGCs were prominent in this subcluster. About one-thirds of genes belonging to this subcluster in the iPGCs showed strong expression as in the ES group cells. In contrast, the expression patterns of the subcluster-2 genes in the GS cells were similar to those of the PGCs. Genes belonging to the subcluster-1 were also up-regulated in the ES group, and again, their expression was generally much lower in the PGCs. Unlike the subcluster-2 genes, the subcluster-1 genes of the iPGCs showed expression patterns similar to those of the PGCs.
|
The clustering results for the ES/PGC common genes are shown in Fig. 3C. It is apparent that the overall expression patterns of the ES/PGC common genes in cells belonging to either the ES or PGC group were highly similar to each other (Fig. 3C). The expression of these genes in the somatic cells was generally low, indicating that the expression of the ES/PGC common genes is a shared feature of ESCs and PGCs.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
In this study, we conducted the first comprehensive and comparative analysis of the transcriptomes of ESCs, gonadal PGCs, and related stem cell lines. Although we analyzed gonadal PGCs of different stages and different sexes, here we will only discuss the common features shared by the PGCs tested in this study. The present analysis revealed unique features of the PGC transcription profiles. The gonadal PGCs express genes specific to the germ-cell lineage, such as Dazl, Fkbp6, Asz1 and Ddx4, which play important roles in germ cell development (Ruggiu et al. 1997; Tanaka et al. 2000; Yan et al. 2002; Crackower et al. 2003). Although it is known that key regulators of ESC pluripotency and self-renewal, e.g., Oct3/4 (Pou5f1) and Nanog are going to be down-regulated in PGCs after entry into gonads (Yamaguchi et al. 2005), these genes are expressed at considerable levels in the gonadal PGCs studied here. It is uncommon for differentiating cells to express Oct3/4 and Nanog, which usually inhibit normal cellular differentiation (Hochedlinger et al. 2005; Suzuki et al. 2006; Tanaka et al. 2007). Therefore, although committed to a single cell lineage, PGCs are unique cells, in which pluripotency markers and specific lineage markers are co-expressed. It is thought that undifferentiated stem cells may express a variety of genes at low levels and that this leaky expression may be the result of the unique chromatin state of ESCs (Zipori 2004; Meshorer et al. 2006; reviewed by Niwa 2007). PGCs are known to undergo major epigenetic changes during their differentiation (Hajkova et al. 2002; Seki et al. 2007; Sugimoto & Abe 2007), which might entail an even more globally relaxed chromatin structure than that of ESCs. Thus, such a relaxed chromatin state could allow the expression of an even wider variety of genes in the PGCs. However, this does not seem to be the case, because ESCs express generally more genes than the PGCs (see Table 2 and Fig. 3) and the selective repression of the ES signature genes was evident in the gonadal PGCs.
Unique features of the transcriptional program in the PGCs
In this study, significant differences in gene expression profiles between the ESC and the PGCs were detected. We noticed that a number of Oct3/4 target genes were differentially expressed in the PGCs. As it is known that a small change in Oct3/4 level causes alterations in the expression of its target genes (Niwa et al. 2000), the differential expression seen in the PGCs might be due to a reduction of the Oct3/4 level in the PGCs. The downstream target genes of Oct3/4 are classified into two categories: one is activated by Oct3/4, which are associated with pluripotency, whereas some genes are negatively regulated by Oct3/4, which might be involved in lineage-specific differentiation (Lee et al. 2006; Matoba et al. 2006). Zfp42, Upp1, Fbxo15, and Otx2 are examples of the positively regulated targets of Oct3/4, and these are in fact down-regulated in the PGCs. Also, expression of Tbx3 and Klf4, known Oct3/4 target genes required for ESC self-renewal (Ivanova et al. 2006; Takahashi & Yamanaka 2006), was greatly reduced in the gonadal PGCs. Down-regulation of Tbx3 and Klf4 should lead to alterations of developmental status of the PGCs compared to ESC. Furthermore, Oct3/4 and Klf4, together with Sox2, cooperate to control a subset of the Oct3/4 target genes in ESCs (Nakatake et al. 2006). Therefore, because of the reduction of Klf4, the transcriptional profiles of the Oct3/4 target genes must differ between ESCs and the PGCs. These results imply that the transcriptional network composed of Oct3/4, its targets, and other key transcriptional regulators also acts in the gonadal PGC, but in a somewhat different way. Such differences may account for the fact that Oct3/4 deficiency results in different consequences in the ESCs and the PGCs (Niwa et al. 2000; Kehler et al. 2004). On the other hand, most of the targets negatively regulated by Oct3/4 are not activated in the PGCs; many of these genes were unchanged or even down-regulated. This suggests that the repression of those genes is not solely regulated by Oct3/4 and that another layer of regulatory mechanisms operates to prevent somatic differentiation in the PGCs. These differences in positive and negative regulation in the PGCs likely contribute to the establishment of expression profiles distinct from those of ESC.
As described above, it is probable that the PGCs and ESCs have their own transcriptional programs, using different combinations of regulatory factors. However, it is also true that they share large numbers of common components of the transcriptional circuitry, which is not operating in somatic cells, and it is possible that the two systems could be interconverted in response to changes in developmental cues. It is known that the fates of PGCs are dependent on the extracellular environment (Donovan et al. 1986; Matsui et al. 1992; McLaren 2003), and the reprogramming of PGCs into EGCs seems to be mediated by the action of cell signaling factors like fibroblast growth factor 2 (Durcova-Hills et al. 2006). In this context, it is interesting that significantly different GO terms are enriched in the PGC signature genes compared with those of the ESC signatures.
We also found that genes involved in DNA methylation, such as Dnmt3b and Dnmt3 l, were barely detectable specifically in the PGCs, whereas they were abundantly expressed in the ESCs and other stem cells except GS cells. PGCs are known to possess quite different epigenomic status compared to somatic cells (Seki et al. 2007). It is thus possible that differences in DNA methylation status represent the basis for the gene expression program in PGCs, and will serve as a diagnostic marker to distinguish PGCs from other stem cells. Further analysis of the PGC signature genes should provide clues to both intrinsic and extrinsic developmental cues required for PGC development.
Expression patterns of signature genes define the developmental identities of EGCs, iPGCs and GS cells
Our multivariate analysis identified the signature genes of the ES and PGC groups. Most of these genes are expressed at lower levels in somatic tissues, but enhanced expression was observed in each group. These gene sets will thus be valuable markers with which to dissect PGC developmental steps or transformation processes during their transition from germ cells to stem cell lines. Here, we demonstrated the characteristics of the transcriptional programs of EGCs, iPGCs and GS cells using these signature genes. The expression profile of GS cells is much more similar to that of male PGCs than to that of any other stem cells tested, suggesting that GS cells retain, at least partially, to their developmental identity as germ cells. Because GS cells are derived from spermatogonia, spermatogonia and PGCs are likely to share similar transcriptional programs. Considerable numbers of spermatogonia-specific genes (Wang et al. 2001) are indeed expressed in PGCs (this study; Abe K., unpublished). Therefore, it should not be surprising that ES-like cells can be derived from spermatogonia.
Analysis of the iPGC revealed that most of the "ES signature" genes are properly down-regulated in the iPGC except for some of the subcluster-2 genes (Fig. 3A); about one-thirds of these genes remained active in the iPGCs. This difference may be relevant to differences in cellular phenotypes between ESC and the iPGCs, for example, teratoma formation ability of ESC after transplantation. Although many of the "PGC signature" genes are not fully up-regulated in the iPGCs, some of the PGC-specific genes/transcripts are in fact up-regulated, suggesting their commitment for germ-line cells. The iPGCs seem to be committed but not differentiated germ-line cells, representing useful experimental materials with which to study molecular requirements for germ cell formation.
The results also imply that all of the "signature" genes are not regulated in the same way because clusters of genes showed differential expression between the iPGCs and the PGC or ESC. Pursuing differences among these cell types should facilitate greater understanding of PGC development both in vivo and in vitro.
Interestingly, TMA58G, one of the EG cell lines examined, shows an expression profile distinct from other ESCs and EGCs, but rather similar to that of the iPGCs. The repression of the PGC signature genes appears to be nearly complete, but the activation of the ES signature genes is not fully achieved in the TMA58G. Considering the similar expression patterns shared by both TMA58G and the iPGCs, it is tempting to speculate that, although the direction is opposite, the conversion processes of PGCs to EGCs and ESCs to iPGCs proceed in a stepwise manner along the same pathway, and that these cells may be suspended at a stage of similar developmental status. Reprogramming PGCs to pluripotent stem cells is not always arrested at an intermediate stage, because the male EG cell line, TMA55G, shows an expression profile almost indistinguishable from that of ESCs. A difference in sex chromosome composition may account for the differences between TMA55G and TMA58G cells. Zvetkova et al. (2005) reported that the genome of female (XX) ESCs is globally hypomethylated, and it is possible that such epigenetic differences could cause changes in gene expression. However, this is not the case, because we isolated several new XY and XX EGCs and ESCs and found that all the cell lines were classified into the ES group based on expression profiling (Supplementary Fig. S1). Moreover, the detailed analysis by Sharova et al. (2007) revealed that ESCs and EGCs are indistinguishable, regardless of their sex or the developmental stages from which they were derived. Therefore, we assume that TMA58G is a unique cell line, retaining part of the genetic and epigenetic properties of PGCs. It should be noted in this context that TMA58G cells are EGCs that express an activity that erases genomic imprinting, which is not expressed by ESCs (Tada et al. 1998, 2001). Further studies of the differences between ESCs, TMA58G, and PGCs will shed light on the molecular nature of genomic reprogramming processes.
In conclusion, we have compared the expression profiles of the gonadal PGCs with those of embryo-derived and germ-cell-derived stem cells and revealed for the first time not only their similarities but also distinct features of their transcriptional programs. These distinct transcriptional profiles might be generated by modulation of Oct3/4-dependent regulatory network. We have identified sets of genes that characterize the developmental status of each cell type, which will serve as useful molecular markers to delineate germ cell lineages in mice. The knowledge obtained in this study should facilitate a wide range of research in germ cell and stem cell biology, and will also help in establishing better protocols for the derivation of germ cells from ESCs or stem cells from germ-line cells in vitro.
|
Experimental procedures |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
R1 (129X1/SvJ and 129S1/SV-p+Tyr+KitlSl-J/+ hybrid origin; Nagy et al. 1993), E14tg2a (129P2 origin; Smith & Hooper 1987), and TMA5 (129/Sv; Tada et al. 1998) are ESC lines with the normal male karyotype. TMA55G and TMA58G are male and female EGC lines derived from the 12.5 dpc PGCs of embryos with Rb(X.2)2Ad x 129/SV hybrid genetic background (Tada et al. 1998). These ESC and EGC lines were cultured on mitomycin-C-treated primary mouse embryonic fibroblasts in Dulbecco's modified Eagle's medium (Sigma-Aldrich, St. Louis, MO) supplemented with 15% fetal bovine serum (FBS; Vitromex, Geilenkirchen, Germany), 1000 U/mL LIF, 2-mercaptoethanol (Gibco, Grand Island, NY), 1 x nonessential amino acids (Gibco), and penicillin/streptomycin (Gibco). Total RNA was extracted from the stem cells after the feeder cells were removed. E14tg2a cells were maintained on gelatinized dishes without feeder cells. We also used ESCs and EGCs newly established by ourselves. VR-2 and #5 were ESCs derived from blastocysts, and B5, C3, A2 and B3 were EGCs of 12.5 dpc embryos origin. Karyotype analyses of these ESCs and EGCs indicated that they retain normal diploid chromosome composition (data not shown). Both VR-2 and #5 ESCs have normal diploid female XX karyotypes. The VR-2 line was derived from mouse with C57BL/6 x DBA/2 mixed background. The #5 ESC, and the B5 and A2 EGCs were derived from F1 hybrid mice between TgN(deGFP)20Imeg (RIKEN BRC No. 00822) and MSM/Ms (RIKEN BRC No. 00209). The C3 and B3 EGCs were derived from F1 embryos between C57BL/6J and MSM/Ms. Details of these newly established ESCs and EGCs will be described in elsewhere (N. Mise, manuscript in preparation).
The Oct-3/4–GFP transgenic mouse line, TgN(deGFP)18Imeg (RIKEN BRC No. 00821), was provided by the RIKEN BioResource Center <http://www.brc.riken.jp/inf/en/gt;, and was used to collect PGCs from developing mouse embryos (Ohbo et al. 2003). BDF1 female mice were mated with Oct3/4–GFP male mice and the 11.5, 12.5 and 13.5 dpc embryos were dissected out in phosphate-buffered saline supplemented with 4% FBS. Male and female genital ridges could be distinguished by their morphologies at 12.5 and 13.5 dpc. The sexes of the 11.5 dpc embryos were determined by PCR using primers for the Y-linked Zfy genomic locus. All animal experiments were approved by the Institutional Animal Experiment Committee of the RIKEN BioResource Center.
Green fluorescent protein (GFP)-positive PGCs were purified using a FACS (Vantage SE, Becton Dickinson, Franklin Lakes, NJ). The purity of the sorted cells was verified by inspecting the green fluorescence of a small aliquot of cells with a fluorescence microscope and was found to be > 99%. In vitro-formed PGC-like cells (iPGCs) were obtained from Mvh–LacZ or Mvh–GFP knock-in ESCs, as described previously (Toyooka et al. 2003). GS cells were derived from F1 hybrid mouse between ICR and B6C3F1-Tg(CAG-EGFP)OsbCX-FM038 (Green mouse; RIKEN BRC No. RBRC00980) according to the published protocol of Ogawa et al. (2004). The EGFP-positive GS cells grown on a feeder layer were purified by FACS. Total RNAs of adult organs were purchased from Ambion (Austin, TX). All the cells and organs used in this study are summarized in Supplementary Table S1.
RNA labeling and micro-array hybridization
The mouse development 22K oligo array, G4120A (Agilent Technologies, Santa Clara, CA), was used throughout this study. Total RNA was labeled with either Cy3 or Cy5 dye with a low-input linear amplification kit (Agilent Technologies). Universal Mouse Reference RNA (Stratagene) was used as the reference RNA for all hybridization experiments. Hybridized slides were scanned using a micro-array scanner (Agilent Technologies) and the signals were processed with the Feature Extraction software ver. 7.5 (Agilent Technologies).
Analysis of micro-array data
Normalization of the logarithmic ratio of the expression intensity between target and reference RNAs was carried out based on MA plots (Dudoit et al. 2002) using TREBAX, a Java-based micro-array analyzing program (details described in Kobayashi et al. 2007; freely available at <http://kanaya.naist.jp/~skanaya/Web/software/trebax/trebax2.html>). FDR was calculated according to Kobayashi et al. (2007). Most of the experiments were performed with two biological and two technical duplications with dye swapping, and the experiments with iPGC samples were done with technical duplications only. Pearson's correlation coefficient of the global expression data was calculated with the standard formula. Student's t-test was used to compare gene expression between ESCs and PGCs. Averaged data of R1, E14tg2a, and TMA5 data were used as ESC data, and data of male and female PGCs from 12.5 and 13.5 dpc embryos were used as PGC data. Genes with fold-change values of < 1.5 in both the ESC and PGC data were eliminated from this comparison. The relative mean expression values for the remaining 1033 genes of the ESC samples and PGC samples were compared using Student's t-test. These genes were classified into five different categories according to the criteria described in Table 2, which are designated classes 1–5.
Principal component analysis (PCA) was performed using TREBAX. The assignment of GO slim <http://www.geneontology.org/GO.slims.shtml> terms was made by using Biocompass software (NEC, Japan). Expression data of the "signature genes" were normalized by mean subtraction by rows with bioNMF software (Pascual-Montano et al. 2006). The normalized data were then subjected to average linkage hierarchical clustering using Pearson's correlation conducted with the clustering software CLUSTER 3.0 (available at the web site <http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster/>). The clustering results were visualized with Java Tree View (available at the web site <http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster/>).
Quantitative real-time PCR
Some of the micro-array results were validated by quantitative real-time PCR. Total RNA samples of FACS-purified gonadal PGCs from 12.5 dpc embryos and R1 and E14tg2a ESCs were used. Twenty five nano grams of total RNA from the ESCs, and total RNA extracted from about 5000 PGCs were reverse transcribed and amplified using WT-Ovation RNA amplification system (NuGEN, San Carlos, CA) to generate cDNA templates for PCR. Ribo-SPIA, an isothermal linear ampilfication technology used in the Ovation system, allows amplification of whole transcriptomes while retaining the distribution of the transcript numbers in the original starting material (Dafforn et al. 2004). The real-time PCR was conducted using Takara's SYBR premix ExTaq kit (Takara, Ohtsu, Japan). The expression level of tested gene was normalized using the expression level of Gapdh gene.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
* Correspondence: abe{at}rtc.riken.jp
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Brinster, R.L. (2002) Germline stem cell transplantation and transgenesis. Science 296, 2174–2176.
Brook, F.A. & Gardner, R.L. (1997) The origin and efficient derivation of embryonic stem cells in the mouse. Proc. Natl Acad. Sci. USA 94, 5709–5712.
Crackower, M.A., Kolas, N.K., Noguchi, J., et al. (2003) Essential role of Fkbp6 in male fertility and homologous chromosome pairing in meiosis. Science 300, 1291–1295.
Dafforn, A. Chen, P., Deng, G., Herrler, M., Iglehart, D., Koritala, S., Lato, S., Pillarisetty, S., Purohit, R., Wang, M., Wang, S. & Kurn, N. (2004) Linear mRNA amplification from as little as 5 ng total RNA for global gene expression analysis. BioTechniques 37, 3–7.
Donovan, P.J., Stott, D., Cairns, L.A., Heasman, J. & Wylie, C.C. (1986) Migratory and postmigratory mouse primordial germ cells behave differently in culture. Cell 44, 831–838.[CrossRef][Medline]
Dudoit, S., Fridlyand, J. & Speed, T. (2002) Comparison of discrimination methods for the classification of tumors using gene expression data. J. Am. Stat. Ass. 97, 77–87.[CrossRef]
Durcova-Hills, G., Adams, I.A., Barton, S.C., Azim Surani, M. & McLaren, A. (2006) The role of exogenous fibroblast growth factor-2 on the reprogramming of primordial germ cells into pluripotent stem cells. Stem Cells 24, 1441–1449.[CrossRef][Medline]
Geijsen, N., Horoschak, M., Kim, K., Gribnau, J., Eggan, K. & Daley, G.Q. (2004) Derivation of embryonic germ cells and male gametes from embryonic stem cells. Nature 427, 148–154.[CrossRef][Medline]
Ginsburg, M., Snow, M.H.L. & McLaren, A. (1990) Primordial germ cells in the mouse embryo during gastrulation. Development 110, 521–528.
Guan, K., Nayernia, K., Maier, L.S., Wagner, S., Dressel, R., Lee, J.H., Nolte, J., Wolf, F., Li, M., Engel, W. & Hasenfuss, G. (2006) Pluripotency of spermatogonial stem cells from adult mouse testis. Nature 440, 1199–1203.[CrossRef][Medline]
Hajkova, P., Erhardt, S., Lane, N., Haaf, T., El-Maarri, O., Reik, W., Walter, J. & Surani, M.A. (2002) Epigenetic reprogramming in mouse primordial germ cells. Mech. Dev. 117, 15–23.[CrossRef][Medline]
Hochedlinger, K., Yamada, Y., Beard, C. & Jaenisch, R. (2005) Ectopic expression of Oct-4 blocks progenitor-cell differentiation and causes dysplasia in epithelial tissues. Cell 121, 465–477.[CrossRef][Medline]
Hubner, K., Fuhrmann, G., Christenson, L.K., Kehler, J., Reinbold, R., De La Fuente, R., Wood, J., Strauss III, J.S., Boiani, M. & Scholer, H.R. (2003) Derivation of oocytes from mouse embryonic stem cells. Science 300, 1251–1256.
Ivanova, N., Dobrin, R., Lu, R., Kotenko, I., Levorse, J., DeCoste, C., Schafer, X., Lun, Y. & Lemischka, I.R. (2006) Dissecting self-renewal in stem cells with RNA interference. Nature 442, 533–538.[CrossRef][Medline]
Kanatsu-Shinohara, M., Inoue, K., Lee, J., et al. (2004) Generation of pluripotent stem cells from neonatal mouse testis. Cell 119, 1001–1012.[CrossRef][Medline]
Kehler, J., Tolkunova, E., Koschorz, B., Pesce, M., Gentile, L., Boiani, M., Lomeli, H., Nagy, A., McLaughlin, K.J., Scholer, H.R. & Tomilin, A. (2004) Oct4 is required for primordial germ cell survival. EMBO Rep. 5, 1078–1083.[CrossRef][Medline]
Kobayashi, H., Akitomi, J., Fujii, N., Kobayashi, K., Altaf-Ul-Amin, M., Kurokawa, K., Ogasawara, N. & Kanaya, S. (2007) The entire organization of transcription units on the Bacillus subtilis genome. BMC Genomics 8, 197.[CrossRef][Medline]
Lee, T.I., Jenner, R.G., Boyer, L.A., et al. (2006) Control of developmental regulators by Polycomb in human embryonic stem cells. Cell 125, 301–313.[CrossRef][Medline]
Loh, Y.H., Wu, Q., Chew, J.L., et al. (2006) The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nat. Genet. 38, 431–440.[CrossRef][Medline]
Matoba, R., Niwa, H., Masui, S., Ohtsuka, S., Carter, M.G., Sharov, A.A. & Ko, M.S. (2006) Dissecting Oct3/4-regulated gene networks in embryonic stem cells by expression profiling. PLoS ONE 1, e26.[CrossRef]
Matsui, Y., Zsebo, K. & Hogan, B.L. (1992) Derivation of pluripotential embryonic stem cells from murine primordial germ cells in culture. Cell 70, 841–847.[CrossRef][Medline]
McLaren, A. (2003) Primordial germ cells in the mouse. Dev. Biol. 262, 1–15.[CrossRef][Medline]
Meshorer, E., Yellajoshula, D., George, E., Scambler, P.J., Brown, D.T. & Misteli, T. (2006) Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev. Cell 10, 105–116.[CrossRef][Medline]
Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W. & Roder, J.C. (1993) Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc. Natl Acad. Sci. USA 90, 8424–8428.
Nakatake, Y., Fukui, N., Iwamatsu, Y., Masui, S., Takahashi, K., Yagi, R., Yagi, K., Miyazaki, J., Matoba, R., Ko, M.S. & Niwa, H. (2006) Klf4 cooperates with Oct3/4 and Sox2 to activate the Lefty1 core promoter in embryonic stem cells. Mol. Cell. Biol. 26, 7772–7782.
Nayernia, K., Nolte, J., Michelmann, H.W., et al. (2006) In vitro-differentiated embryonic stem cells give rise to male gametes that can generate offspring mice. Dev. Cell 11, 125–132.[CrossRef][Medline]
Nesterova, T.B., Mermoud, J.E., Hilton, K., Pehrson, J., Surani, M.A., McLaren, A. & Brockdorff, N. (2002) Xist expression and macroH2A1.2 localisation in mouse primordial and pluripotent embryonic germ cells. Differentiation 69, 216–225.[CrossRef][Medline]
Niwa, H. (2007) How is pluripotency determined and maintained? Development 134, 635–646.
Niwa, H., Miyazaki, J. & Smith, A.G. (2000) Quantitative expression of Oct-3/4 defines differentiation, dedifferentiation or self-renewal of ES cells. Nat. Genet. 24, 372–376.[CrossRef][Medline]
Ogawa, T., Ohmura, M., Tamura, Y., Kita, K., Ohbo, K., Suda, T. & Kubota, Y. (2004) Derivation and morphological characterization of mouse spermatogonial stem cell lines. Arch. Histol. Cytol. 67, 297–306.[CrossRef][Medline]
Ohbo, K., Yoshida, S., Ohmura, M., Ohneda, O., Ogawa, T., Tsuchiya, H., Kuwana, T., Kehler, J., Abe, K., Scholer, H.R. & Suda, T. (2003) Identification and characterization of stem cells in prepubertal spermatogenesis in mice. Dev. Biol. 258, 209–225.[CrossRef][Medline]
Pascual-Montano, A., Carmona-Saez, P., Chagoyen, M., Tirado, F., Carazo, J.M. & Pascual-Marqui, R.D. (2006) bioNFM: a versatile tool for non-negative matrix factorization in biology. BMC Bioinformatics 7, 366.[CrossRef][Medline]
Ruggiu, M., Speed, R., Taggart, M., McKay, S.J., Kilanowski, F., Saunders, P., Dorin, J. & Cooke, H.J. (1997) The mouse Dazla gene encodes a cytoplasmic protein essential for gametogenesis. Nature 389, 73–77.[CrossRef][Medline]
Seki, Y., Yamaji, M., Yabuta, Y., Sano, M., Shigeta, M., Matsui, Y., Saga, Y., Tachibana, M., Shinkai, Y., Saitou, M. (2007) Cellular dynamics associated with the genome-wide epigenetic reprogramming in migrating primordial germ cells in mice. Development 134, 2627–2638.
Sharov, A.A., Piao, Y., Matoba, R., et al. (2003) Transcriptome analysis of mouse stem cells and early embryos. PLoS Biol. 1, E74.[Medline]
Sharova, L.V., Sharov, A.A., Piao, Y., Shaik, N., Sullivan, T., Stewart, C.L., Hogan, B.L., Ko, M.S. (2007) Global gene expression profiling reveals similarities and differences among mouse pluripotent stem cells of different origins and strains. Dev. Biol. 307, 446–459.[CrossRef][Medline]
Smith, A.G. & Hooper, M.L. (1987) Buffalo rat liver cells produce a diffusible activity which inhibits the differentiation of murine embryonal carcinoma and embryonic stem cells. Dev. Biol. 121, 1–9.[CrossRef][Medline]
Solter, D. (2006) From teratocarcinomas to embryonic stem cells and beyond: a history of embryonic stem cell research. Nat. Rev. Genet. 7, 319–327.[CrossRef][Medline]
Sugimoto, M. & Abe, K. (2007) X chromosome reactivation initiates in nascent primordial germ cells in mice. PLoS Genet. 3, e116.[CrossRef][Medline]
Suzuki, A., Raya, A., Kawakami, Y., Morita, M., Matsui, T., Nakashima, K., Gage, F.H., Rodriguez-Esteban, C. & Izpisúa Belmonte, J.C. (2006) Nanog binds to Smad1 and blocks bone morphogenetic protein-induced differentiation of embryonic stem cells. Proc. Natl. Acad. Sci. USA 103, 10294–10299.
Tada, M., Takahama, Y., Abe, K., Nakatsuji, N. & Tada, T. (2001) Nuclear reprogramming of somatic cells by in vitro hybridization with ES cells. Curr. Biol. 11, 1553–1558.[CrossRef][Medline]
Tada, T., Tada, M., Hilton, K., Barton, S.C., Sado, T., Takagi, N. & Azim Surani, M. (1998) Epigenotype switching of imprintable loci in embryonic germ cells. Dev. Genes Evol. 207, 551–561.[CrossRef][Medline]
Takahashi, K. & Yamanaka, S. (2006) Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 1–14.[CrossRef]
Tanaka, S.S., Toyooka, Y., Akasu, R., Katoh-Fukui, Y., Nakahara, Y., Suzuki, R., Yokoyama, M. & Noce, T. (2000) The mouse homolog of Drosophila Vasa is required for the development of male germ cells. Genes Dev. 14, 841–853.
Tanaka, Y., Era, T., Nishikawa, S. & Kawamata, S. (2007) Forced expression of Nanog in hematopoietic stem cells results in a 
T-cell disorder. Blood 110, 107–115.
Tolostonog, G.V., Wang, X., Shoeman, R. & Traub, P. (2000) Intermediate filaments reconstituted from vimentin, desmin, and glial fibrillary acidic protein selectively bind repetitive and mobile DNA sequences from a mixture of mouse genomic DNA fragments. DNA Cell Biol. 10, 647–677.
Toyooka, Y., Tsunekawa, N., Akasu, R. & Noce, T. (2003) Embryonic stem cells can form germ cells in vitro. Proc. Natl Acad. Sci. USA 100, 11457–11462.
Wang, P.J., McCarrey, J.R., Yang, F. & Page, D.C. (2001) An abundance of X-linked genes expressed in spermatogonia. Nat. Genet. 27, 422–426.[CrossRef][Medline]
Yamaguchi, S., Kimura, H., Tada, M., Nakatsuji, N. & Tada, T. (2005) Nanog expression in mouse germ cell development. Gene Expr. Patterns 5, 639–646.[CrossRef][Medline]
Yan, W., Rajkovic, A., Viveiros, M.M., Burns, K.H., Eppig, J.J. & Matzuk, M.M. (2002) Identification of Gasz, an evolutionarily conserved gene expressed exclusively in germ cells and encoding a protein with four ankyrin repeats, a sterile-
motif, and a basic leucine zipper. Mol. Endocrinol. 16, 1168–1184.
Zipori, D. (2004) Nature of stem cells: state rather than entity. Nat. Rev. Genet. 5, 873–878.[Medline]
Zvetkova, I., Apedaile, A., Ramsahoye, B., Mermoud, J.E., Crompton, L.A., John, R., Feil, R., Brockdorff, N. (2005) Global hypomethylation of the genome in XX embryonic stem cells. Nat. Genet. 37, 1274–1279.[CrossRef][Medline]
Zwaka, T.P., Thomson, J.A. (2005) A germ cell origin of embryonic stem cells? Development 132, 227–233.
Accepted: 13 May 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
