| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EMBRYONIC STEM CELLS |
aLaboratory for Stem Cell Biology, RIKEN Center for Developmental Biology, Kobe, Japan;
bScience of In-Home Medicine, Health and Community Medicine, Nagoya University Graduate School of Medicine, Nagoya, Japan;
cResearch & Development, Cardio Inc., Kobe, Japan
Key Words. Embryonic stem cell • Mesendoderm • Short hairpin RNA • Definitive endoderm
Correspondence: Takumi Era, M.D., Ph.D., Laboratory for Stem Cell Biology, RIKEN Center for Developmental Biology, 2-2-3 Minatojima-minamimachi, Chuo-ku, Kobe 650-0047, Japan. Telephone: 81-78-306-1893; Fax: 81-78-306-1895; e-mail: tera{at}cdb.riken.jp
Received October 25, 2006;
accepted for publication April 8, 2007.
First published online in STEM CELLS EXPRESS April 19, 2007.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
(PDGFR)+ vascular endothelial growth factor receptor 2 (VEGFR2)+ and PDGFR+ mesoderm generations, respectively. Our results obtained from this study are fully consistent with previous knockout studies of those genes. The present study, therefore, demonstrates that the major molecular mechanism underlying in vitro ES cell differentiation largely recapitulates that in actual embryogenesis, and the combination of our culture system and RNAi-mediated gene KD is an useful tool to elucidate the molecular hierarchy in in vitro ES cell differentiation. Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Despite this recent progress, the use of ES cell culture in investigating the process of differentiation has drawn criticism for not accurately reflecting the processes involved in the actual embryo. To solve this problem, we found it necessary to monitor the entire intermediate process in vitro.
One way to characterize the intermediate process is to compare the phenotypes resulting from the inhibition of a particular gene known to be involved in the embryo's differentiation pathway. Indeed, looking at the accumulated data on the in vitro behavior of ES cells bearing null mutation of various genes suggests that the results observed in the ES cell culture are largely consistent with those of actual embryos [12, 13].
The present study aimed at presenting an alternative method for verifying that the in vitro process of mesendoderm differentiation in the simple monolayer cultures is capable of reflecting that of the actual embryonic process and is useful for examining the early processes involved in ES cell differentiation. In our attempt to carry this out, we combined our monolayer cultures with the short hairpin RNA (shRNA)-mediated gene knockdown method [14] and examined the effect of knocking down a number of key molecules whose roles in embryogenesis have been relatively well-characterized in the in vivo differentiation of mesendoderm.
This study demonstrates that the major molecular mechanism underlying the mesendoderm differentiation in ES cell culture recapitulates that in actual embryogenesis. Our results clearly show the genetic hierarchy of the molecular determinants in the specification and differentiation of mesoderm and definitive endoderm via mesendoderm. The combination of in vitro ES cell culture and the shRNA-mediated gene knockdown method is useful for elucidating the molecular networks involving in the early process of ES cell differentiation.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
-minimal essential medium supplemented with 10% fetal calf serum. For Sox1 neuroepithelial cells, we added 10–7 M retinoic acid into the culture from day 2 to day 4 (Condition D). Condition D is modified from the condition described by Dani et al. and is an efficient condition for inducing adipocytes [18].
Design of shRNA Sequence for Each Gene and Construction of Virus Vectors
The sequence effective for gene knockdown (KD) was designed using software available at the Web sites siDirect (http://design.rnai.jp/) and siRNA Target Designer, version 1.51 (http://www.promega.com/siRNADesigner/program/) (supplemental online Fig. 1A). We also designed the scramble sequences for shRNA to confirm the specificity of KD effects (supplemental online Fig. 1B).
The retroviral vector used to express shRNA, pQCXIX-U6-BamHI-EcoRI-HindIII-puro, was based on the pQCXIX retroviral vector (Clontech, Palo Alto, CA, http://www.clontech.com). Each oligonucleotide encoding the short hairpins of targeted gene were ligated into the BamHI/HindIII site. Oligonucleotides encoding only terminal signals flanked by BamHI and HindIII were used as a control.
For Mixl1 overexpression, we used the lentiviral vector because of its stable and string expression during in vitro ES cell differentiation as previously described [19]. In this vector, the expression of transgene inserted is driven by elongation factor-1 promoter, which is constitutively active during ES cell differentiation. As the vector carries the DsRed-T4 cDNA fused with internal ribosomal entry site, infected cells are easily distinguished from uninfected cells by use of a flow cytometer, such as fluorescence-activated cell sorting (FACS).
Production and Infection of Viruses Carrying shRNAs or Mixl1cDNA
The retroviruses and lentiviruses were produced as described previously [20]. For infection, semiconfluent ES cells were incubated with 1 ml of ES medium containing 0.1 ml of virus particles overnight. Puromycin (2 µg/ml) was added 24 hours after infection for selecting the gene-transduced cells. For obtaining ES cells stably expressing Mixl1, cells were purified by FACS based on expression of DsRed. After establishing cell lines, expression of Mixl1 was confirmed by Western blot with an antibody to V5-tag fused to the Mixl1 gene.
FACS Analysis and Cell Purification
Cells were stained as described previously using various combinations of monoclonal antibodies (mAbs) [17]. Antibodies used in this study were Alexa 488-, allophycocyanin (APC)-, and biotinylated conjugated anti-E-cadherin mAb (ECCD2) and APC- and phycoerythrin (PE)-anti-human CD25 mAb (Becton, Dickinson and Company), biotinylated anti-platelet-derived growth factor receptor (PDGFR) mAb (APA5), and APC-conjugated anti-vascular endothelial growth factor receptor 2 (VEGFR2) mAb (AVAS12). PE and PE-Cy7-conjugated streptavidin (BD Pharmingen, San Diego, http://www.bdbiosciences.com/pharmingen) was used for detecting biotinylated anti-ECCD2 and APA5. Analysis of staining with mAbs, cell sorting by FACSAria (Becton, Dickinson and Company), and data analysis by FlowJo (Tree Star, Ashland, OR, http://www.treestar.com) were carried out as described previously [17].
Quantitative Reverse Transcription-Polymerase Chain Reaction and Reverse Transcription-Polymerase Chain Reaction Analysis
Quantitative reverse transcription-polymerase chain reaction (q-RT-PCR) and reverse transcription-polymerase chain reaction analyses were performed as described previously [17]. Primers used in this study are listed in supplemental online Figure 1C.
Statistical Analysis
In q-RT-PCR analysis and statistical analysis of cell population, the statistical significance was calculated by the Student t-test. All experiments were performed three times. In q-RT-PCR analysis, expression levels of the molecules were evaluated three times for one experiment. Expression levels were normalized by expression of GAPDH (each molecule/GAPDH ratio) before calculating the expression ratio to the expression level of control culture. After being normalized, each value of the specific genes was divided by the average value of control culture in the experiment for standardization. In the figures, error bars represent SD and # and * represent p < .05 and p < .01, respectively.
Western Blot Analysis
Western blot analysis was performed as described previously [21]. Primary antibodies used in this study were anti-Eomesodermin (rabbit polyclonal) (kindly provided by Dr. H. Niwa), anti-Mixl1 (rat monoclonal) (6G2 [21]; kindly provided by Dr. L. Robb), anti-Brachyury (goat polyclonal) (sc-17743; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), anti-Gata6 (goat polyclonal) (sc-7244; Santa Cruz Biotechnology), and anti-ß-Actin (ab 8226; Abcam, Cambridge, U.K., http://www.abcam.com) antibodies. Primary antibodies were detected with horseradish peroxidase-conjugated secondary antibodies using ECL reagents (Amersham Biosciences, Little Chalfont, U.K., http://www.amersham.com) according to the manufacturer's instructions.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
, and VEGFR2 were detected by surface staining with specific mAbs [17]. Although Eomesodermin (Eomes) [22], Mixl1 [23], Brachyury (T) [24], and GATA6 [25] are known to be expressed during mesendoderm differentiation, several previous studies have implicated Eomes and Mixl1 as playing central roles in the mesendoderm differentiation [23, 26]. The kinetics of the expression of the four molecules in the ES cell differentiation under Condition B was measured by q-PCR (Fig. 2A). Under this particular condition, we found that the expression level of Eomes, Mixl1, and T dramatically increased between day 0 and day 3. The expression of Eomes and Mixl1 continued to increase until day 4.5 and was maintained until day 6, whereas T expression actually decreased after day 3. We observed GATA6 expression later than the other three markers, and in contrast to these, GATA6 was found to be expressed considerably earlier (day 3) during visceral endoderm (VE) differentiation under Condition A (supplemental online Fig. 2A). We performed gene KD by stably transducing shRNA with a retroviral vector into ES cells (Fig. 2B) and prepared both two shRNA and their scramble shRNA vector constructs, each containing two distinct sequences, to confirm the specificity of their knockdown effects (supplemental online Fig. 1A, 1B). The efficiency of the gene KD was confirmed by comparing the mRNA and protein levels of cells transduced by target shRNA or scramble shRNA sequences following induction of differentiation in the culture (Fig. 2C, 2D).
|
We transduced ES cells either with GATA6 shRNAs (GATA6 KD ES cells), with scramble shRNA for GATA6, or with the vector alone and cultured with either Condition A for 6 days or Condition B for 4.5 days before comparing the proportion of Gsc–Sox17+ VE and Gsc+Sox17+ DE cells. As shown in Figure 3 and supplemental online Figure 2B and 2C, GATA6 KDs (KD1 and KD2) suppressed differentiation of Gsc–Sox17+ VE cells, whereas they had no effect on Gsc+Sox17+ DE cells.
|
Hierarchy of Activin-Induced Early Genes in Mesendoderm Differentiation
Eomes, Mixl1, and Sox17 have been implicated as being central players in DE differentiation via mesendoderm in Xenopus and zebrafish [31–33], although the role of Eomes in DE differentiation has not yet been experimentally shown in mouse. We therefore thought that KD of these genes would be an effective method of confirming that in vitro differentiation of Gsc+Sox17+ cells tracks the same process as actual DE. Among these three factors, Mixl1 and Sox17 have been shown as being involved in the commitment of some parts of DE [23, 34]. In contrast to this, the role of Eomes in mammalian DE development, although clearly implicated in early mesendoderm differentiation in zebrafish [26], remains to be clearly established. We therefore thought that using our in vitro culture to determine whether Eomes is involved in DE differentiation via mesendoderm would be an effective approach.
To evaluate the effect of KD on the DE differentiation pathway, we measured the proportion of three populations in day 4.5 culture: (a) Gsc+Sox17+ DE, (b) ECD+PDGFR+ mesendoderm, and (c) ECD–PDGFR+ mesoderm. We had previously been able to show that ECD and PDGFR are coexpressed at the mesendoderm stage but segregate to PDGFR– endoderm and ECD– mesoderm lineages, respectively [11]. T KD was used as a control, as previous studies had suggested that T does not have a significant cell-autonomous role in DE differentiation [35].
We prepared ES cell lines that were stably transfected by the shRNA of each gene and their scramble shRNA and induced their differentiation under Condition B. We have found that the generation of Gsc+Sox17+ DE is markedly suppressed by Mixl1 KD, whereas T KD shows only a relatively insignificant effect (Fig. 3B, 3C). Our results also reveal that Eomes KD results in the marked inhibition of Gsc+Sox17+ DE differentiation, suggesting that its role is implicated in DE differentiation. Interestingly, the differentiation of Gsc+ cells as a whole is strongly suppressed in Eomes KD, whereas their proportion remained basically unchanged by Mixl1 KD, which is due to an increase in the proportion of Gsc+Sox17– cells (Fig. 3B). The ECD+PDGFR+ mesendoderm generation was markedly suppressed by Eomes KD, although it remained unaffected by Mixl1, T, or GATA6 KD (lower panels in Fig. 3B, 3C). The generation of ECD–PDGFR+ mesoderm was inhibited only by Eomes KD, suggesting that this defect of mesendoderm-derived mesoderm generation was due to the suppression of mesendoderm differentiation. Similar results were obtained from another KDs (KD2s) of each gene (supplemental online Fig. 2C). In contrast, the treatment of scramble shRNA did not affect any populations (supplemental online Fig. 2B). Viewing these results as a whole, then, it would appear that Eomes is involved in the process of mesendoderm differentiation, whereas Mixl1 is implicated in the later differentiation from mesendoderm to DE. This is consistent with previous studies in zebrafish [26, 32], which implicate the involvement of Eomes in earlier mesendoderm differentiation. Despite the involvement of Eomes in this early process, we found that the generation of Gsc+ cells could not be suppressed completely, although its appearance in culture was significantly impaired (supplemental online Fig. 3B).
Relationship Between Eomes and Mixl1 in Mesendoderm Differentiation
A previous study had demonstrated that Mixl1 expression in the primitive streak was lost in Eomes knockout (KO) mice, whereas the expression of other mesoderm markers, such as T and FGF8, was maintained [36]. Moreover, Mixl1 was also found to be essential for the differentiation of some subsets of DE [23]. Despite the lack of experimental evidence, it has been suspected that some features of the Eomes KO mouse phenotype may be a result of the loss of the Mixl1 gene. Consistent with this is our observation that KDs of both Eomes and Mixl1 result in defects in the Gsc+Sox17+ DE population. Based on these findings, we therefore thought that in vitro ES cell differentiation to DE might be suitable for addressing this issue.
We first investigated whether our culture system would be able to detect this hierarchical relation between Eomes and Mixl expressions. Eomes- or Mixl1-KD ES cells were cultured under Condition B for 4.5 days, and the expression of each gene was evaluated by q-RT-PCR and Western blotting. As shown in Figure 4 and supplemental online Figure 4A–4D, we found that Eomes KDs suppressed the expression of Mixl1 as well as Eomes, whereas Mixl1 KDs had no effect on Eomes expression, indicating that Mixl1 expression is under the control of Eomes.
|
Overexpression of Mixl1 during control ES cell differentiation under Condition B enhanced the differentiation of Gsc+Sox17+ DE population (Fig. 5 A, 5B), which is consistent with previous studies using Xenopus [39]. More striking is the observation that Mixl1 overexpression could almost completely rescue the defect of Gsc+Sox17+ cell differentiation in Eomes KD. The Gsc+Sox17+ cells rescued by Mixl1 overexpression expressed other endodermal markers, such as Foxa2 [40], Cerberus 1 [41], and Claudin 6 [42] (supplemental online Fig. 4G). In contrast, we found that restoration of differentiation of ECD+PDGFR+ mesendoderm was almost negligible, although the expression of Gsc did accelerate (Fig. 5A). When these data were taken together, we determined that the major site of Mixl1 involvement can be found at the mesendoderm-to-DE differentiation stage, and it may play some role in mesendoderm differentiation by accelerating Gsc gene expression. As the expression level of Eomes in Eomes KD was unaffected by Mixl1 overexpression (Fig. 5C), the results obtained from q-RT-PCR and overexpression experiments indicate that Mixl1 is, genetically, a downstream molecule of Eomes.
|
Role of Eomes and Mixl1 in Differentiation of Other Mesoderm
Based on the observation that teratoma derived from Eomes-null ES cells contains mesodermal lineages [36], the cell-autonomous role of Eomes in the differentiation of paraxial and lateral mesoderm had been thought to be limited. In contrast to this is a line of evidence indicating that Mixl1 is involved in mesoderm differentiation, such as mesoderm contributing to the heart tube [23] and hematopoietic lineage [43, 44]. Given that Mixl1 is a downstream molecule of Eomes, this would indicate that Eomes is an essential factor involved in mesoderm differentiation. We therefore decided to investigate the role of Eomes and Mixl1 in the differentiation of both lateral and paraxial mesoderm. Eomes-, Mixl1-, and T-gene KD ES cells were prepared and cultured for 4 days under Condition C, which permits the differentiation of lateral and paraxial mesoderm [17]. These molecules are expressed in ES cell-derived mesoderm populations, although their expression levels are variable (supplemental online Fig. 6A). After confirming the efficiency and specificity of KDs in the whole culture under Condition C (supplemental online Fig. 6B–6D), we investigated the effects of KDs on the generation of the mesoderm cells. As shown in Figure 6A and 6B and supplemental online Figure 6E, Eomes KDs, as well as Mixl1 KDs, resulted in a partial suppression of the in vitro differentiation of PDGFR–VEGFR2+ lateral mesoderm (VSP), implicating a role for Eomes and Mixl1 in the differentiation of lateral mesoderm. It is possible that this partial suppression corresponds to the defect in cardiac mesoderm in Mixl1 KO mouse [23]. Our results also suggest that Eomes is involved in the differentiation of a subset of lateral mesoderm. Although we found that suppression of differentiation of VSP cells was only partially affected by Eomes KDs and Mixl1 KDs, the generation of PDGFR+VEGFR2+ double-positive (DP) cells was almost completely suppressed by Eomes KDs, although it was permitted to some extent by Mixl1 KDs (Fig. 6A, 6B; supplemental online Fig. 6E). As the DP cells represent an earlier progenitor that can give rise to both paraxial and lateral mesoderm [17], it can be seen that Eomes is involved in an earlier stage than Mixl1 during differentiation of lateral mesoderm. What is of interest here is that although Eomes KDs resulted in the marked suppression of DP stage, which corresponds to the early progenitors of single-positive populations, a significant level of VEGFR2–PDGFR+ paraxial mesoderm (PSP) cells were generated from Eomes KD ES cells (Fig. 6A, 6B; supplemental online Fig. 6E). These findings are consistent with the previous study showing that Eomes-null ES cells give rise to teratoma cells containing a diverse set of mesoderm lineages, although the cell migration of mesoderm progenitors is severely impaired in the tetraploid chimera with Eomes-null ES cells during gastrulation [36].
|
We also examined the effects of KDs on neuroepithelial cell development. We cultured Sox1gfp/+ ES cells under Condition D and measured the proportion of Sox1+ cells on day 4, revealing that none of these KDs were able to affect the generation of Sox1+ neuroepithelial cell differentiation (Fig. 6C; supplemental online Fig. 7A, 7B). This is consistent with the evidence that Sox1+ cells do not express these molecules except for GATA6 (supplemental online Fig. 7C).
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Here we attempted to address this question by comparing the outcome of both in vitro and in vivo processes resulting from the inhibition of genes whose role has been implicated in the differentiation of DE. We decided to use Eomes, Mixl1, T, and GATA6 as probes to evaluate the process through which Gsc+ mesendoderm and subsequent Gsc+Sox17+ DE are generated. A previous gene KO study has implicated Mixl1 as being involved in the differentiation of DE [23]. Although Mixl1 expression disappeared in Eomes-null embryos [36], several key questions remain to be answered, such as the role of Eomes and Mixl1 in the cell-autonomous process of DE differentiation and the hierarchy among these particular molecules. In an attempt to address these questions, we were able to establish that an ES cell differentiation culture would provide the most suitable conditions in which to conduct our investigation, because it is relatively free from any effects stemming from the complex interaction between developing cells.
In attempting to inhibit gene function, we used the shRNA-mediated gene KD method [14] and were subsequently able to demonstrate that differentiation of Gsc+Sox17+ DE cells is suppressed by KD of Eomes and Mixl1, but not by that of T or GATA6. In contrast, the differentiation of Gsc–Sox17+ VE cells was inhibited by GATA6 KD.
Our results obtained from this study suggest that gene KD is a convenient and efficient method for use in elucidating the intermediate processes leading to the generation of a particular cell type in vitro. This finding is in complete agreement with the previous study demonstrating that gene KD combined with ES cell technology provides a rapid method for determining gene function [47]. Despite the problem of the magnitude of gene suppression by KD being less than gene KO, the phenotypes generated by KD are comparable to those of gene KO. Kunath et al. assessed the phenotypes using the tetraploid embryo chimera method, which can be used to assess the entire process of cell specification and morphogenesis from KD ES cells [47]. In this study, we were able to show that KD can also be combined with cultures that induce differentiation of selective cell lineages, if an appropriate method to monitor the differentiation process is available. Taken together, we anticipate that the combination of gene KD and ES cell culture has the potential to be a popular and convenient method for evaluating the function of a particular gene involved in cell specification processes during early embryogenesis.
Regarding the question of whether the process taking place in culture can recapitulate the actual in vivo process, our KD experiments showed consistency with previous gene KO studies, indicating that in vitro cell specification of Gsc+ECD+PDGFR+ mesendoderm and subsequent Gsc+Sox17+ DE essentially follow nearly the same pathway as that of the actual embryos in a number of ways. First, GATA6 KD showed no effect on the generation of Gsc+Sox17+ DE, whereas it did suppress the generation of Gsc–Sox17+ VE cells. This result corroborates previous studies in GATA6 KO mouse [25, 27, 29], which revealed that this is essential for the differentiation of visceral endoderm but not for differentiation of DE, except for pulmonary development. The differentiation of Gsc+Sox17+ cells, in contrast, was markedly suppressed by Eomes and Mixl1 KD but not by T or GATA6 KD. This is again consistent with the previous studies on Mixl1 KO mice showing the specific involvement of these molecules in DE differentiation [23]. Thus, even though the ES cell per se has an artificial existence and culture conditions for its differentiation can hardly be identical to the embryonic microenvironment, we are convinced that the molecular mechanism underlying the differentiation pathway is robust enough to be repeated in an artificial culture condition. This study demonstrated this in the in vitro ES cell differentiation to DE.
The results of our present study also highlighted the utility of combining gene KD and ES cell culture in examining the molecular mechanisms regulating mesendoderm differentiation. Together with activin, the serum-free medium is highly selective to the differentiation of Gsc+ cells composed of Sox17+ DE and Sox17– mesoderm, resulting in the effect of KD being presented as a simple increase and decrease in those populations that can be generated in the culture. Indeed, the effect of KD of both Eomes and Mixl1 could be most visibly seen in the decrease of Gsc+Sox17+ cells. As far as we are aware, this is the first demonstration of the role of Eomes in mouse DE specification, although its role in DE of zebrafish was recently reported [32]. In the tetraploid chimera analysis with Eomes-null ES cells, embryogenesis is arrested at the gastrula stage because of the failure of node formation [36]. Consistent with this, we observed a marked suppression of generation of the ECD+PDGFR+ mesendoderm, which contains cells corresponding to the node. This mesendoderm defect is likely to be the reason for the decrease of differentiation of ECD–PDGFR+ mesoderm by Eomes KD. In contrast to this finding, although DE generation is markedly suppressed by Mixl1 KD, it showed almost no effect on the generation of ECD+PDGFR+ mesendoderm. These findings appear to indicate that the normal influx of mesendoderm with suppression of DE may be the reason for the compensatory increase of Gsc+Sox17– cells. Summarizing these results shows Mixl1 as having a role in the process of mesendoderm to DE differentiation, whereas Eomes is involved in the earlier step of ES cell differentiation to mesendoderm. It should also be emphasized that even our relatively simple culture system can mimic the expansion of the T-expressing area observed in Eomes- and Mixl1-KO embryos [23, 36]. In fact, our KD of both Eomes and Mixl1 resulted in the total level of T expression increasing markedly, suggesting that both molecules, directly or indirectly, have a suppressive effect on T expression.
Another point of interest for us was that T KD resulted in a decrease of PDGFR+ paraxial mesoderm under Condition C. This observation was unexpected, as the results of chimeric embryos consisting of T KO and wild-type ES cells has been interpreted as indicating that the major role of T in embryogenesis is regulation of cell migration and adhesion [35]. This result is, however, consistent with previous studies on Xenopus animal cap assay that implicate a cell-autonomous role being involved in the specification of mesoderm cells [48]. It is therefore likely that the two T-box molecules, Eomes and T, are both involved in cell specification, as well as in the migration of mesoderm cells. Based on these results, then, ES cell cultures, being less affected by cell-cell interaction, offer a valuable tool that complements phenotypic analysis of KO embryos.
Our study has demonstrated that ES cell culture provides an efficient and convenient tool for analyzing the molecular hierarchy governing cell specification. A previous study had revealed that Mixl1 expression is lost in Eomes KO mice [36]. The condition was also reproducible in ES culture, as Eomes KD suppresses the expression of both Eomes and Mixl1, whereas Mixl1 KD has no effect on Eomes expression. Moreover, earlier studies in Xenopus had suggested that Eomes induces Mix1 expression in the animal cap [49], meaning that the involvement of Mixl1 in DE development as a downstream molecule of Eomes has been widely accepted. This question, however, remained to be addressed by experiments examining whether Mixl1 can rescue the Eomes defect. We therefore decided to address this question using ES cell culture. As expected, we found that Mixl1 overexpression can rescue the two phenotypes resulting from Eomes KD, suppression of Gsc+Sox17+ DE differentiation and enhancement of T expression. Thus, Mixl1 acts as a downstream molecule of Eomes and is involved in the differentiation from mesendoderm to DE in our culture system (Fig. 7).
|
|
CONCLUSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Although the reservation that in vitro culture system may deviate from the actual process always has to be kept in mind, the present study suggests that this combination method is useful for searching for molecular network in in vitro ES cell differentiation.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionDisclosure of Potential...AcknowledgmentsReferences |
|---|
, extended packaging signal. (C): KD efficiency in ES cell differentiation under Condition A (GATA6 KDs) or Condition B (the other KDs). The graph shows the ratio of expression of each target molecule observed in the total of cells of day 6 (Condition A) or day 4.5 (Condition B) culture of ES cells transduced with shRNA versus those with scramble shRNA. We used two types of shRNA (KD1 and KD2) encoding different sequences in the same target to confirm the specificity of their KD effects. All shRNA preparation significantly decreased the expression levels of target genes. The average ratios of target transcript reduction induced by each shRNA treatment were as follows: Eomes KD1, 78%; Eomes KD2, 58%; Mixl1 KD1, 90%; Mixl1 KD2, 74%); T KD1, 74%; T KD2, 63%); GATA6 KD1, 87%; GATA6 KD2, 82%. The statistical significance was calculated by the Student t-test. #, p < .05; *, p < .01. (D): The protein expression in ES cell differentiation under Condition A or Condition B. The suppression of the expression levels was confirmed by Western blot analysis with specific antibodies. The total cell lysates were extracted from day 6 (Condition A, GATA6 KDs) or day 4.5 (Condition B, the other KDs) culture of ES cells transduced with shRNA, scramble shRNA, or the control vector. The protein expressions were strongly suppressed by shRNA treatments. The expression levels were similar in both scramble shRNA-treated and control ES cells. Abbreviations: CMV/MSV, cytomegalovirus/mouse sarcoma virus hybrid promoter; Eomes, Eomesodermin; KD, knockdown; LTR, long terminal repeat; Pgk-P, phosphoglycerate kinase promoter; puroR, puromycin resistance gene; T, Brachyury; U6-P, mouse U6 promoter.
