|Home | About | Journals | Submit | Contact Us | Français|
Nuclear receptor subfamily 0, group B, member 1 (Nr0b1, also known as Dax1) is regarded as an important component of the transcription factor network that governs pluripotency in mouse embryonic stem (ES) cells. Here we generated inducible knockout ES cells for Nr0b1 using the Cre-loxP system and analyzed its precise function. We succeeded in establishing the Nr0b1-null ES cells and confirmed their pluripotency by showing their contribution to chimeric embryos. However, they proliferated slowly with over-expression of 2-cell stage specific transcripts including Zscan4c, which is known to be involved in telomere elongation in ES cells. We revealed that over-expression of Zscan4c prevents normal self-renewal by inducing arrest at G2 phase followed by cell death and that Nr0b1 directly represses the Zscan4c promoter. These data indicated that Nr0b1 is not essential to maintain pluripotency but is involved in the proper activation of 2-cell specific transcripts for self-renewal.
Nr0b1 (also known as Dosage-sensitive sex reversal-adrenal hypoplasia congenital on the X-chromosome gene-1: Dax1) is a unique member of the nuclear family because it lacks the DNA binding domain and works to modulate the function of other nuclear receptors1. It has been well analyzed that Nr0b1 is involved in germ cell development2,3. Nr0b1 is expressed at a high level in ES cells but not in epiblast stem cells (EpiSCs), suggesting the role in the naïve state of pluripotency4. The expression of Nr0b1 is regulated by the LIF signal via Jak-Stat3 pathway5 as well as by Oct3/45, Nanog6, Nr5a26 and Esrrb7. Conversely, Nr0b1 binds to Oct3/4 to inhibit its transcriptional activity and the over-expression of Nr0b1 induces differentiation toward trophectoderm as in the case of Oct3/4 knockout8. It was also reported that Nr0b1 interacts with Esrrb to inhibit its transcriptional activity7. Interestingly, while the biochemical analyses suggested that Nr0b1 interacts with Nr5a2 to suppress its function9,10, it was also demonstrated that Nr0b1 cooperates with Nr5a2 and Steroid receptor RNA activator 1 (Sra1) to activate the Oct3/4 promoter11. What happens if Nr0b1 function is eliminated in ES cells? Niakan et al reported that either knock-down of Nr0b1 by siRNA or knock-out of Nr0b1 by the Cre-lox system induces differentiation of ES cells12. This is consistent with the second report by Khalfallah et al showing that siRNA-mediated knock-down of Nr0b1 causes multi-lineage differentiation13. However, in both cases, the primary effect of the withdrawal of Nr0b1 on the pluripotency-associated transcription factor network was not well analyzed. Here we generated an inducible knockout ES cell line of Nr0b1 with the Cre-lox system to examine its precise role to regulate the transcription factor network. Our data indicated that Nr0b1 is dispensable for maintaining pluripotency but is involved in the transcriptional regulation of 2-cell specific genes in ES cells.
The Nr0b1 gene consists of 2 exons (Fig. 1a). It has been reported that the deletion of exon 2 results in functional ablation. We made a knockout vector in which two loxP sites were inserted in intron 1 and the 3′ of exon 2, with a PGKpacΔtkpA cassette flanked by Frt sites inserted adjacent to the 5′ end of the 3′ loxP site (Fig. 1a). The linearized knockout vector was introduced into male ES cells by electroporation followed by the selection with puromycin. As a result we obtained multiple clones with correct homologous recombination event confirmed by combinations of long-range genomic PCR (Fig. 1b), designated as Nr0b1flFrt/Y ES cells. Then the expression vector of the FLP recombinase (FLPe)14 was transiently transfected by lipofection followed by the selection with gancyclovir, resulting in the generation of ES cells in which the PGKpacΔtkpA cassette was excised, designated as Nr0b1fl-/Y ES cells. Then the piggy-bac vectors for constitutive expression of hormone-inducible Cre (MerCreMer)15 and Egfp were introduced into the Nr0b1fl-/Y ES cells to obtain inducible Nr0b1 knockout (Nr0b1KO/Y) ES cells.
The Nr0b1fl-/Y ES cells carrying MerCreMer and Egfp continued self-renewal as wild-type ES cells and contributed to chimeric embryos after blastocyst injection (see below). When these ES cells were treated with 4-hydroxy tamoxifen (Tx) to activate the Cre recombinase for 2 days followed by dissociation into single cells and plating onto a fresh culture, the colonies with typical ES cell morphology were formed after 6 days at a comparable rate (72%) to that of Nr0b1fl-/Y ES cells without the Tx treatment (Fig. 1c). Then these colonies were isolated and the genotype was examined by PCR. Among the 44 clones examined, 30 clones (68%) were genotyped as Nr0b1KO/Y ES cells that possess the deleted allele by the excision of the floxed region (data not shown), indicating that the loss of Nr0b1 has no primary impact on the clonogenicity of ES cells. Finally we succeeded to establish multiple Nr0b1KO/Y ES cell lines in which proper loss of Nr0b1 protein was confirmed by western blot analysis (Fig. 1d and Supplemental Fig. 1). These data clearly indicated that the function of Nr0b1 is dispensable for continuous self-renewal of ES cells in contrast to previous reports12,13.
During the establishment of Nr0b1KO/Y ES cells, we found that these ES cells formed smaller colonies than Nr0b1fl/Y and the wild-type (Nr0b1+/Y) ES cells at the same culture period (Fig. 2a). The measurement of the proliferation ratios of Nr0b1KO/Y ES cells revealed that it was less than 20% of Nr0b1fl/Y and Nr0b1+/Y ES cells (Fig. 2b). Although ES cells show decreased proliferation ratio when they undergo differentiation in vitro, Nr0b1KO/Y ES cells retained the expression of pluripotency-associated transcription factors such as Oct3/416, Sox217, Nanog18,19, Klf4 and Tbx320 (Fig. 2c) and kept the typical ES cell morphology (Fig. 2a). These data suggested that these Nr0b1KO/Y ES cells should retain proper pluripotency.
To reveal the reason for the slow proliferation of Nr0b1KO/Y ES cells, we examined the cell cycle profile by FACS analysis with propidium iodide (PI) staining of the DNA. ES cells have a characteristic cell-cycle profile, in which the cells at the G1 phase are few and those at the S phase are abundant in comparison to somatic cell types, reflecting the unlimited G1-S transition21. It has been reported that the proportion of G1 is increased in Klf5-null ES cells22 and Sall4-null ES cells23 that show slow proliferation. However, the proportion of the cells at the G1 phase was not altered in Nr0b1KO/Y ES cells (Fig. 2d). Instead, we found significant increase in the G2/M phase in Nr0b1KO/Y ES cells (Fig. 2d), suggesting that Nr0b1 might modulate proliferation without affecting the G1-S transition.
Increased incidence of cell death could cause low proliferation ratio of ES cell population. Thus we tested the proportion of dead cells, and further examined whether the death was caused by apoptosis or not, by double-staining with PI and Annexin-V24. Early apoptotic cells are stained only by Annexin-V, whereas dead cells are stained by both. The proportions of the early apoptotic cells and also the overall dead cells were both increased in Nr0b1KO/Y ES cells (Fig. 2e). These data indicated that the function of Nr0b1 might be involved in suppression of cell death caused by apoptosis.
Next we assessed whether Nr0b1KO/Y ES cells retain pluripotency. The dissociated Nr0b1KO/Y ES cells expressing Egfp were injected into blastocysts and their contribution in embryos dissected at E13.5 was evaluated. As shown in Figure 3, both Nr0b1fl-/Y and Nr0b1KO/Y ES cells contributed to chimera embryos as efficiently as Nr0b1+/Y ES cells. The efficiencies to give rise to chimeric emrbyos were also indistingishable between Nr0b1fl-/Y and Nr0b1KO/Y ES cells (Supplemental Table 2). Therefore, pluripotency is properly maintained in the absence of Nr0b1 in ES cells.
The data shown above indicated that Nr0b1 was required for accelerating cell proliferation and keeping normal mortality but dispensable for maintaining pluripotency. For further investigation of the role of Nr0b1 in ES cells, we performed microarray analyses to examine the effect of Nr0b1 elimination on the global gene expression. We compared the gene expression patterns of Nr0b1fl-/Y ES cells cultured with or without Tx for 4 days (Nr0b1fl-/Y+Tx and Nr0b1fl-/Y−Tx, respectively) and Nr0b1KO/Y ES cells, using the samples of the biological triplicates. The comparison between Nr0b1fl-/Y−Tx and Nr0b1fl-/Y+Tx was expected to show the primary effect of Nr0b1 elimination, whereas that between Nr0b1fl-/Y+Tx and Nr0b1KO/Y would show the effect of the adaptive response to compensate the loss of Nr0b1. There was little difference in the gene expression patterns of wild-type and Nr0b1fl-/Y−Tx ES cells, indicating that the manipulation of Nr0b1 as well as the expression of MerCreMer and Egfp had the minimal impact (data not shown). The comparison between the gene expression patterns of Nr0b1fl-/Y−Tx and Nr0b1fl-/Y+Tx ES cells showed characteristic differences. Among 202 genes significantly up-regulated in Nr0b1fl-/Y+Tx ES cells (Fig. 4a: >2-fold, FDR = 0.05), we found multiple genes categorized as 2-cell specific transcripts including Zscan4c, Tcstv1, and Thoc4/Gm434025 (Fig. 4b). Among the pluripotency-associated genes, Klf4 and Tbx3 were slightly up-regulated, but the genes that have been previously reported to be associated with rapid proliferation (Eras26, Mybl227,28, Klf522 and Sall423) did not alter their expression levels (Fig. 4b). In the comparison between Nr0b1fl-/Y+Tx and Nr0b1KO/Y ES cells, 209 genes were up-regulated and 103 genes were down-regulated (Fig. 4a). Interestingly, the differentiation markers for the endoderm (Gata4, Gata6 Sox7, Sox17 and Hnf4a29) were over-expressed in Nr0b1KO/Y ES cells (Fig. 4b). The 2-cell specific transcripts remained over-expressed in Nr0b1KO/Y ES cells.
To confirm the over-expression of the genes identified by the microarray analysis, we performed immuno-staining for Sox7, Sox17, Gata4 and Tcstv1. Sox7, Sox17 and Gata4 were detected as a cluster in the colonies with differentiated cell morphology, suggesting that the increased expression level of those genes reflects the elevated incidence of spontaneous differentiation of Nr0b1KO/Y ES cells (Supplementary Fig. 2). Tcstv1 (2-cell-stage, variable group, member 1) is known as one of the 2-cell specific transcripts expressed in a subpopulation of ES cells. We raised a polyclonal antibody against Tcstv1 and confirmed its specificity by the staining pattern, which localized at the membrane in a 2-cell-stage specific pattern of Tcstv1, as reported previously30 (Supplementary Fig. 3a). When we stained wild-type ES cells, we found that Tcstv1 expression was detected in ~0.1% of ES cells, which all overlapped with the expression of Rex1-Egfp and Oct3/4-Ecfp in OCRG9 ES cells31 (Supplementary Fig. 3b). When Nr0b1KO/Y ES cells were stained for Tcstv1, the frequency of the Tcstv1-positive cells dramatically increased in the stem cell colonies than in Nr0b1fl-/Y ES cells (Fig. 4c), suggesting that the increased expression levels of the 2-cell specific transcripts reflected the increase in the subpopulation rather than the increase in their expression levels ubiquitously in the stem cell population.
To investigate the effect of Nr0b1 on the kinetics of the subpopulation expressing the 2-cell specific transcripts, we applied a Zscan4c-promoter-mCherry reporter. Zalzman et al reported that the 2.5 kb genomic DNA fragment upstream of Zscan4c (Zinc finger and SCAN domain containing 4c) directed the expression of a fluorescent reporter in ~5% of ES cells and the conversion between the Zscan4c-positive subpopulation and the Zscan4c-negative subpopulation occurred in a reversible manner32. Moreover, depletion of Zscan4c by shRNA-mediated knock-down resulted in a crisis of ES cell population within 1 month by shortening of the telomere length, indicating the functional importance of the transient activation of Zscan4c in a subpopulation of ES cells32. We transfected the Zscan4c-mCherry transgene into Nr0b1fl-/Y and Nr0b1KO/Y ES cells and three independent cell lines expressing mCherry were established for each genotypes. Observation of these ES cells by fluorescent microscopy revealed a significant increase in mCherry-positive cells in Nr0b1KO/Y::Zscan4c-mCherry ES cells (Fig. 4d). Next we applied FACS analysis to quantify the proportion of the Zscan4c-positive subpopulation. In Nr0b1fl-/Y::Zscan4c-mCherry ES cells, 4.9% of ES cells were mCherry-positive (Fig. 4e), which was comparable to the proportion of Zscan4c-positive cells reported previously. In contrast, Nr0b1KO/Y::Zscan4c-mCherry ES cells contained 21.6% of mCherry-positive cells, indicating the dramatic increase in Zscan4c-positive subpopulation, in the same manner found in the immuno-staining for Tcstv1 as shown earlier (Fig. 4c).
Then we performed live-imaging of these ES cells carrying Zscan4c-mCherry. As reported previously, Zscan4c-positive cells appeared from Zscan4c-negative cells (Supplementary Movie 1). The emergence of the Zscan4c-positive cells was much more frequent in Nr0b1KO/Y::Zscan4c-mCherry ES cells than in Nr0b1fl-/Y::Zscan4c-mCherry ES cells, indicating that the increase of Zscan4c-positive cells in Nr0b1KO/Y ES cells was partly due to the altered kinetics in which the conversion from Zscan4c-negative to -positive was enhanced. These data suggested that Nr0b1 negatively regulates the kinetics to generate Zscan4c-positive subpopulation.
It has been reported that Zscan4c is transiently up-regulated in a subpopulation of ES cells to elongate the telomere length via homologous recombination and that the constitutive expression of Zscan4c is not tolerated in ES cells32,33. To investigate the link between the increase in the Zscan4c-positive cells and the slow proliferation accompanied by increased G2 phase and cell death in Nr0b1KO/Y ES cells, we examined the cell-cycle profiles in Zscan4c-positive and -negative subpopulations. mCherry-positive and -negative cells were sorted from Nr0b1KO/Y::Zscan4c-mCherry ES cells and their cell-cycle profiles were analyzed by PI staining (Fig. 5a). As a result, we found that the cells at G2 phase were enriched in Zscan4c-positive cells. The similar phenomenon was observed in Nr0b1fl-/Y::Zscan4c-mCherry ES cells, indicating that the Zscan4c-positive cells have increased G2 phase irrespective of the state of Nr0b1.
This phenomenon seemed consistent with the function of Zscan4c that mediates homologous recombination to elongate the telomere. To test the direct relation between the expression of Zscan4c and the enrichment of G2 phase, we established an ES cell line carrying a tetracycline-inducible Zscan4c transgene. These ES cells continue self-renewal in the absence of the inducer. However, once the expression of Zscan4c transgene was induced by doxycycline (Dox), these ES cells gradually ceased proliferation as reported previously34. We examined the cell-cycle profiles of these ES cells cultured with or without Dox for 2 days and revealed that the ES cells cultured with Dox accumulated at the G2 phase (Fig. 5b). Therefore, the up-regulation of Zscan4c could be a cause to limit the transition from G2 to M phase.
To test the link between the activation of Zscan4c and the incidence of cell death in a physiological context, we analyzed the live-imaging data of Zscan4c-mCherry ES cells. We quantified the total fluorescent intensities of mCherry during one cell cycle, i.e. from the cell division until either the next cell division or cell death. As a result, we found that the cells undergoing cell death expressed higher levels of mCherry than the cells undergoing cell division in both Nr0b1fl/Y and Nr0b1KO/Y ES cells (Fig. 4f). These data strongly suggested that up-regulation of Zscan4c is a main cause to inhibit proliferation of Nr0b1KO/Y ES cells via increased incidence of cell death.
How does the loss of Nr0b1 connect to the increase in the subpopulation expressing 2-cell specific transcripts? To determine whether the Zscan4c-promoter is directly regulated by Nr0b1, we performed luciferase reporter assay. The Zscan4c-promoter was placed upstream of firefly luciferase (Fluc) and co-transfected with renilla luciferase (Rluc) under the control of human cytomegallovious immediate-early promoter into Nr0b1fl-/Y and Nr0b1KO/Y ES cells. Dual luciferase assay of these transfectants demonstrated that the relative Fluc activity driven by the Zscan4c-promoter was dramatically increased in Nr0b1KO/Y ES cells (Fig. 6a), consistent with the results obtained by the Zscan4c-mCherry reporter. When these reporters were co-transfected with Nr0b1 expression vector, the activity of the Zscan4c-promoter was significantly repressed in Nr0b1KO/Y ES cells (Fig. 6a). We also confirmed that these 2-cell specific transcripts including Zscan4c were rapidly up-regulated after repression of Nr0b1 by QPCR analysis (Fig. 6b). These data clearly indicated that the Zscan4c promoter activity is directly regulated by Nr0b1 in ES cells.
To confirm the direct regulation of Zscan4c by Nr0b1, we established Nr0b1KO/Y ES cell lines carrying tetracycline-inducible Nr0b1 transgene by introduction of pPB-hCMV*-1-Nr0b1-IRES-mCherry and pPB-CAG-rtTAM2-IN followed by the selection with G418 in the presence of Dox (Fig. 7a). We isolated the colonies with rapid proliferation and analyzed their characters. These Nr0b1KO/Y::hCMV*1-Nr0b1 ES cells proliferated faster with lower levels of Zscan4c and Tcstv1 expression than in Nr0b1KO/Y ES cells (Fig. 7b and 7c). Then we tested the expression of Nr0b1 protein by western blot. As shown in Supplementary Fig. 4, Nr0b1 expression in Nr0b1KO/Y::hCMV*1-Nr0b1 ES cells was significantly lower than that in Nr0b1fl/Y ES cells, which might be a reason why these ES cells still showed differences from Nr0b1fl/Y ES cells in proliferation speed as well as the expression of 2-cell marker genes. We next examined the impact of Nr0b1 expression on the proportion of 2-cell subpopulation. We evaluated the proportion of Tcstv1-positive cells in Nr0b1KO/Y::hCMV*1-Nr0b1 ES cells cultured with or without Dox by immunostaining and found that the 2-cell subpopulation was decreased when Nr0b1 transgene was activated (Fig. 7d and 7e). The proportion of Tcstv1-positive cells as well as the expression of 2-cell marker genes were reverted to high levels when these ES cells were cultured without Dox (Fig. 7c and 7e), indicating that the expression of 2-cell markers are regulated in a Nr0b1-dependent manner. Moreover, by the activation of Nr0b1 with Dox, the expression of 2-cell marker genes were repressed within 24 hours after the addition of Dox. These data confirmed that the 2-cell marker genes including Zscan4c are direct targets of Nr0b1.
Nr0b1 lacks the DNA binding domain and acts as a partner of other nucelar receptor family members. Nr5a2 is a strong candidate partner of Nr0b1 from previous publications9,10. To test the role of Nr5a2 in ES cells, we made inducible knocout ES cells of Nr5a2 using the Cre-loxP system as in the case of Nr0b1 (figure S5a). As a result, we found that induced knockout of Nr5a2 also caused the induction of 2-cell markers (figure S5c) with reduced proliferation (figure S5b). These data suggested that Nr5a2 might be a partner of Nr0b1 and the heterodimer acts as a transcriptional repressor of 2-cell marker genes including Zscan4c.
Mouse ES cells continue self-renewal in vitro in an unlimited manner34. To continue proliferation, the telomere length should be actively maintained because it shortens along the cycle of chromosome replication and its shortening results in chromosome instability and cell lethality35. The elongation of the telomere by the enzymatic activity of telomerase is regarded as a major pathway to keep the telomere length. The catalytic activity of telomerase is encoded by Telomerase reverse transcriptase (Tert). Tert is expressed in many proliferating cells including stem cells. However, it was reported that the ES cells lacking Tert generated by gene-targeting continue proliferation with gradual shortening of the telomere length and finally reaches the crisis after the culture over 1 year, indicating that there are other systems to keep the telomere length in ES cells36,37,38. In 2010, Zalzman et al reported a novel mechanism to maintain the telomere length in mouse ES cells32. They focused on the functional analysis of Zscan4c in ES cells because they had found that Zscan4c is transcriptionally activated in the late 2-cell stage in mouse pre-implantation embryos and shows a salt-and-pepper expression pattern in ES cells. When they knocked down Zscan4c in ES cells, these ES cells stopped proliferation after 8–9 passages within 1 month by shortening the telomere length. They found that Zscan4c co-operates with Rad50 and Mre11 to maintain the telomere length by homologous recombination as found in some cancer cells. The regulation of Zscan4c expression is quite unique since only 5% of ES cells express it at a certain time point but all of the expression is reversible and transient, and all ES cells in the culture experience the expression of Zscan4c within 1 month to maintain the telomere length. It has remained unclear how this interesting regulation is achieved, but here we found that Nr0b1 is involved in the regulation of the Zscan4c expression.
In ES cells, Zfp206/Zscan10 was previously identified as the transcriptional activator of Zscan4c25. Zscan4c and other 2-cell specific transcripts were down-regulated in Zfp206 knock-down ES cells and Zfp206 was transiently induced at early 2-cell stage preceding the induction of 2-cell specific transcripts in pre-implantation development. However, our microarray analysis indicated that the expression of Zfp206 was not affected by elimination of Nr0b1 (Fig. 4b), suggesting that the regulation of 2-cell specific transcripts by Nr0b1 is independent of Zfp206. It was also reported that Zscan4c is repressed by the inhibition of the phosphatidylinositol-3-OH kinase (PI3K)-Akt pathway in ES cells39. However, Nr0b1 was also down-regulated in this condition, which is inconsistent with our present observation. Recently, Macfarlan et al demonstrated that ES cells possess a subpopulation expressing endogenous retrovirus as well as 2-cell specific transcripts and their expressions were antagonized by repressive chromatin-modifying enzymes such as Kdm1a, Kap1 and G9a40. However, none of these genes were up-regulated in Nr0b1KO/Y ES cells (Fig. 4b). More recently, Dan et al reported that Rif1 directly regulates Zscan4c expression by mediating epigenetic repression41. Interestingly, the phenotype of Rif1 knockdown ES cells is quite similar to that of Nr0b1 knockout ES cells reported in this manuscript such as slow proliferation with increased proportion G2/M phase and induction of 2-cell marker genes. However, Nr0b1 is not included in the list of the genes down-regulated in Rif1 knockdown ES cells and Rif1 expression is not repressed in our microarray data of Nr0b1 knockout ES cells (Fig. 4b). Therefore, the negative regulation of Zscan4c by Nr0b1, which was in fact a direct repression, is unique from the previous observations.
We demonstrated that Zscan4c is a functional target of Nr0b1. The Nr0b1-null ES cells proliferate slowly with accumulation at G2/M phase and high incidence of apoptosis. Over-expression of Zscan4c also induced the similar phenotype efficiently, and the increased proportion of Zscan4c-positive cells found in Nr0b1KO/Y ES cells accords to explain the phenotype. Moreover, our live-imaging analysis of Zscan4c-mCherry ES cells revealed that the cells undergoing cell death show higher levels of Zscan4c activity than the cells undergoing cell division (Fig. 4f). Therefore, the aberrant expression of Zscan4c could be a primary cause of the phenotype observed in Nr0b1KO/Y ES cells. We also found using an inducible knockout line of Nr5a2 that Nr5a2 is a strong candidate partner of Nr0b1 to repress Zscan4c expression. The functional link of Nr0b1:Nr5a2 and Zscan4c confers the connection between the pluripotency-associated transcription factor network and the cell-biological character of ES cells, i.e. the unlimited ability to self-renew. However, since Zscan4c was not expressed in all Nr0b1KO/Y ES cells, the repression of Zscan4c by Nr0b1 is only a part of the regulatory mechanism. Further analysis will be required for revealing the unique mechanism that induces Zscan4c expression in a small subpopulation transiently.
Several reports pointed the role of Nr0b1 in maintaining pluripotency. Two preceding papers demonstrated the essential role of Nr0b1 in self-renewal of ES cells by loss-of-function analyses12,13. However, here we succeeded in establishing Nr0b1-null ES cells. siRNA-mediated knock-down strategy sometimes show severer phenotypes than those observed by gene-targeting, which might be due to the faster kinetics of ablation of the gene function without the time for compensation or the off-target effect. Here we applied the inducible knockout using the Cre-loxP system, providing a rapid kinetics of the functional ablation of Nr0b1. However, most of the knockout ES cells continued self-renewal, which allowed efficient establishment of Nr0b1-null ES cells. The transcriptomic analysis suggested that there was no remarkable change in the gene expression pattern to adopt the loss of Nr0b1 for continuous self-renewal (data not shown). Therefore, the discrepancy from the previous reports might be due to the off-target effect of knock-down strategies applied in the previous reports otherwise the reason for the different result obtained by the inducible knockout remains unclear. We observed enhancement of spontaneous differentiation in Nr0b1KO/Y ES cells although the majority of these ES cells continued self-renewal with slow proliferation ratio. It could be true that, these ES cells might have higher incidence to loose proper pluripotency. We observed that both Nr0b1fl-/Y ES cells and Nr0b1KO/Y ES cells established in the culture without feeder cells lost the ability to contribute to chimeric embryos after injection into blastocysts with over-expression of Klf4 and Tbx3 (data not shown). The role of the feeder cells to settle the over-expression of Klf4 and Tbx3 is obscure, but it may suggest the role of Nr0b1 in stabilizing the pluripotency-associated transcription factor network. Although its function looks not essential to maintain pluripotency as suggested by the comprehensive analyses of the network, here we disclosed the definitive function of Nr0b1 on the regulation of 2-cell specific transcripts to limit their expression in a small subpopulation. Further analysis will reveal the role in the pluripotency-associated transcription factor network that controls pluripotency in biological sense.
EB5 ES cells (derived from male E14tg2a ES cells) were cultured with or without mouse embryonic fibroblast feeder cells treated with Mitomycin C in GMEM supplemented with 10% FCS, 1× sodium pyruvate, 1× non-essential amino acids, 10−4 M of 2-mercaptoethanol and 1000 U/ml of LIF.
For generation of Nr0b1 KO vector, genomic DNA fragments for 5′ and 3′ homology arms (Chr:X, 86191214-86195211 and 86196711-86199810 in GRCm38) as well as the floxed region containing exon 2 (Chr:X, 86195212-86196710 in GRCm38) were amplified from EB5 genomic DNA using the primer pairs Nr0b1-5′, Nr0b1-3′ and Nr0b1-flox, respectively, shown in Suplementary Table 1. The 5′ arm and the floxed region were digested with ClaI and NotI, respectively, and inserted into ClaI-NotI of pBR-MC1DTApA using ligation and In-fusion cloning (Clonetech), resulting in pDTA-Nr0b1-5′+flox. Then the 3′ arm was digested by BglII and NotI and inserted into BglII and NotI of pDTA-Nr0b1-5′+flox, resulting in pDTA-Nr0b1-5′+flox+3′. Finally, the Frt-PGKpacΔtkpA-Frt cassette was introduced into the BglII site of pDTA-Nr0b1-5′+flox+3′, resulting in pDTA-Nr0b1 floxKO.
The PiggyBac vectors pPBCAG-MerCreMer-IH, pPBCAG-Egfp-IZ, pPBCAG-mCherry-IP, pPBCAG-rtTAM2-IN and pPB-hCMV*-1-IRES-mCherry-pA42,43,44 were constructed based on pGG13145. For construction of the Zscan4c reporters, the Zscan4c promoter (Chr:7, 11003915-11006554 in GRCm38) was amplified from EB5 genomic DNA using KOD-Plus-Neo (TOYOBO) using the primer pair Zscan4c-promoter shown in Supplementary Table 1 and subcloned into pPBCAG-mCherry-IP by replacing CAG promoter and SacI-XhoI of pGL3-Basic (Promega), resulting in pPB-Zscan4c-mCherry and pZscan4c-Fluc, respectively. For inducible expression of Zscan4c, the entire coding region was amplified with the primer pair Zscan4c-CDS from EB5 cDNA and subcloned into pPB-hCMV*-1-IRES-mCherry-pA, resulting in pPB-hCMV*-1-Zscan4c-IRES-mCherry. The expression vector of Nr0b1 was generated by insertion of the entire coding region of Nr0b1 amplified from EB5 cDNA with the primer pair Nr0b1-CDS into pPyCAG-IP, resulting in pPyCAG-Nr0b1-IP. For inducible expression of Nr0b1, the XhoI-NotI fragment of Nr0b1 CDS was subcloned into pPB-hCMV*-1-IRES-mCherry-pA, resulting in pPB-hCMV*-1-Nr0b1-IRES-mCherry.
For generation of Nr5a2 KO vector, genomic DNA fragments for 5′ and 3′ homology arms (Chr:1, 136951766-136950374 and 136944805-136943464 in GRCm38) as well as the floxed region containing exons 3 and 4 (Chr:1, 136950373-136944806 in GRCm38) were amplified from EB5 genomic DNA using the primer pairs Nr5a2-5′, Nr5a2-3′ and Nr5a2-flox, respectively, shown in Suplementary Table 1. These PCR fragments were assembled in EcoRV site of pBR-blue II, resulting in pBR-Nr5a2 5′+flox+3′. Finally, the Frt-SA-Ires-neo-pA:PGKpacΔtkpA-Frt cassette was introduced into the BamHI site flanking the 5′ end of the 3′ loxP sequence of pBR-Nr5a2 5′+flox+3′, resulting in pBR-Nr5a2 floxKO.
107 EB5 ES cells were electroporated with 100 μg of linealized Nr0b1 KO vector DNA at 800 V and 3 μF in a 0.4-cm cuvette using a Gene Pulser (Bio-Rad), followed by culture with 1.5 μg/ml of puromycin for 8 days. The resulting stem cell colonies were picked up, expanded and genotyped by PCR using the primer pairs KO-PCR1 and KO-PCR2. The correctly targeted clones (Nr0b1flFRT/Y) were seeded in a well of 48-well plate at 104 cells per well, and transfected with 1 μg of circular pCAG-FLPe-IP plasmid using Lipofectoamine 2000 (Invitrogen) followed by the culture for 3 days. Then these transfected cells were replated and cultured with 1 μM of Gancyclovir for 8 days. The resulting stem cell colonies were picked up, expanded and genotyped by PCR using the primer pair Nr0b1-flox. The clones in which the PGKpacΔtkpA cassette flanked by Frt were correctly removed (Nr0b1fl-/Y) were seeded in a well of 48-well plate at 104 cells per well, and transfected with 0.25 μg of circular pPB-CAG-MerCreMer-IH, 0.25 μg of circular pPB-CAG-Egfp-IZ, and 0.5 μg of circular pCAG-PiggyBac transposase (PBase) plasmid using Lipofectoamine 2000 followed by the culture for 3 days. Then these transfected cells were replated and cultured with 200 μg/ml of Hygromycin B and 20 μg/ml of Zeocin for 8 days. The resulting stem cell colonies were picked up, expanded and assessed for the expression of Egfp by fluorescent microscopic analysis as well as the function of MerCreMer by PCR genotyping of the Tx-treated cells using the primer pair P1. Finally, two independent pairs of Nr0b1flFRT/Y and its descendant Nr0b1KO/Y ES cells were analyzed in this study. For monitoring the expression of Zscan4c, Nr0b1flFRT/Y and Nr0b1KO/Y ES cells were seeded in a well of 48-well plate at 104 cells per well, and transfected with 0.5 μg each of pPB-Zscan4c-mCherry and pCAG-PBase, and 0.05 μg of pPB-CAG-Egfp-IZ plasmid using Lipofectoamine 2000 followed by the culture for 3 days. Then these transfected cells were replated and cultured with 20 μg/ml of Zeocin for 8 days. The resulting stem cell colonies were picked up, expanded and assessed for the expression of mCherry by fluorescent microscopic analysis. To establish ES cells carrying the inducible Zscan4c, ES cells were seeded in a well of 48-well plate at 104 cells per well, and transfected with 0.5 μg each of pPB-hCMV*-1-Zscan4c-IRES-mCherry and pCAG-PBase, and 0.05 μg of pPB-CAG-rtTAM2-IN plasmid using Lipofectoamine 2000 followed by the culture for 3 days. Then these transfected cells were replated and cultured with 160 μg/ml of G418 for 8 days. The resulting stem cell colonies were picked up, expanded and assessed for the expression of mCherry in the presence of Doxycycline (1 μg/ml) by fluorescent microscopic analysis. Establishment of Nr0b1-null ES cells carrying the inducible Nr0b1 was performed in the same way.
The inducible Nr5a2 KO ES cells were generated in the same way as the inducible Nr0b1 KO ES cells except for the serial KO of the second allele.
Dissociated ES cells were injected into a C57BL6 blastocyst by microinjection, which was then transferred into the uterus of a pseudopregnant female ICR mouse. Embryos were collected at 13.5 dpc to evaluate chimera contribution ability of ES cells by analyzing with fluorescence microscopy. All animal experiments confirmed to our Guidelines for the Care and Use of Laboratory animals and were approved by the Institutional Committee for Laboratory Animal Experimentation (RIKEN Kobe Institute).
Polyclonal antibody against Tcstv1 was obtained by immunization of rabbit by purified fusion protein of GST and full-length Tcstv1 followed by the purification of antibody by affinity column. Cells were fixed with 4% of paraformaldehyde for 30 minutes at 4°C, washed with PBS containing 2% FCS, and incubated with anti-Tcstv1 antibody (1:1000) for overnight at 4°C. After washing with PBS, the cells were incubated with donkey anti-rabbit IgG Alexa-Fluor-594-conjugated antibody (Molecular Probes) for 30 minutes at room temperature. Fluorescent images were captured using an IX51 microscope with DP70 digital camera (Olympus) or a Leica SP2 confocal microscope (Leica). Western blot was performed with anti-Nr0b1 (Active Motif, #39983) and anti-Cdk2 (Santa Cruz, sc-163) for the total cell lysate of wild-type, Nr0b1flFRT/Y and Nr0b1KO/Y ES cells.
Cell-cycle analysis were done by staining cells with propidium iodide (SIGMA-ALDRICH) in 2 different experimental steps. As the first step, Nr0b1flFRT/Y and Nr0b1KO/Y ES cells were cultured in GMEM/10%FCS/LIF medium and cell-cycle status were measured. The second step is that cherry positive/negative cells were sorted from Zscan4-mCherry expressing Nr0b1flFRT/Y and Nr0b1KO/Y cells and analyzed.
For apoptotic assay, Annexin V-APC (BD Biosciences) staining was performed according to the manufacturer's protocol using Nr0b1flFRT/Y and Nr0b1KO/Y ES cell. Apoptotic cells were assessed by FACS (BD Biosciences) and analyzed by Flowjo software (Digital biology).
First strand DNA was synthesized from 500 ng of total RNA prepared by QuickGene RNA cultured cell HC kit (KURABO) in 20 μl of the reaction mixture containing oligo-dT primers using a ReverTra Ace first strand synthesis kit (TOYOBO). Real-time PCR was performed with THUNDERBIRD SYBR qPCR Mix (TOYOBO) using CFX384 Real-Time System (Bio-Rad). Sequences of primer pairs are listed in Supplementary Table 1.
DNA microarray analyses were performed using a SurePrint G3 Mouse GE Microarray 8 × 60 K (Agilent Technologies). Microarray results were analyzed using NIA Array Analysis Software. Complete array data will be available on the GEO (NCBI) website.
For live cell imaging, 3 clones of each Nr0b1KO/Y::Zscan4c-mCherry and Nr0b1fl-/Y::Zscan4c-mCherry ES cells were seeded 1000cells per well on a ibidi 8well chamber (Nippon Genetics) coated with poly-L-lysin (100 μg/ml; Sigma-Aldrich) and then with Ecadherin-Fc (5 μg/ml; R&D Systems). Cells were monitored in a humid atmosphere with 5% CO2 at 37°C under an inverted microscope (IX81, Olympus) equipped with a confocal spinning disk (CSU-X1, Yokogawa), a CCD camera (iXon, Andor) and MetaMorph imaging software (Molecular Devices). Time lapse images were taken every 1 hour with a 10× objective lens.
For assessing the Zscan4c promoter activity in Nr0b1flFRT/Y and Nr0b1KO/Y ES cells, 6 × 104 cells were seeded in each well of a 24-well plate and transfected with 0.4 μg of circular pZscan4c-Fluc, 1.6 μg of circular pPyCAG-IP or pPyCAG-Nr0b1-IP, and 0.04 μg of circular pCMV-Rluc plasmid (Promega) using Lipofectoamine 2000 followed by the culture for 24 hours. Luciferase assays were performed using a Dual-luciferase assay kit (Promega).
Supplementary Movie 1
We are grateful for the supporting unit of animal experiments for technical support. This work was supported by CREST by the Japan Science and Technology Agency (JST) and RIKEN grant to H.N.
The authors declare no competing financial interests.
Author Contributions H.N. designed the experiments, performed overexpression study and wrote the manuscript. S.F. established knockout ES cells and performed proliferation assay, immunostaing, QPCR analysis and luciferase assay. S.N.-T. performed FACS analyses, Y.F. performed live cell imaging analysis, M.Y. generated and analyzed the inducible Nr0b1 expressing cells, Y.T. prepared anti-Tcstv1 antibody and S.O. generated chimeric embryos.