EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS
1
Key Laboratory of Gene Engineering of the Ministry of Education, Institute of Healthy Aging Research and SYSU‐BCM Joint Research Center, School of Life Sciences, Sun Yat‐sen University, Guangzhou, 510275, China; 2Key La‐ boratory of Reproductive Medicine of Guangdong Province, School of Life Sciences and the First Affiliated Hospi‐ tal, Sun Yat‐sen University, Guangzhou, 510275, China; 3Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, One Baylor Plaza, Hou‐ 4 ston, Texas 77030, USA; Guangdong Gastroenterology Institute, Depart‐ ment of Gastrointestinal Surgery, The Sixth Affiliated Hospital of Sun Yat‐sen University, China; 5Department of Urology, The First People's Hospital Affiliated to Guangzhou Medical Uni‐ 6 versity; International Peace Maternity and Child Health Hospital, School of Medicine, Shanghai Jiao Tong Universi‐ ty, Shanghai 200030, China; 7 Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, Texas 77030,USA. *Correspondence should be addressed to : Junjiu Huang, Key Laboratory of Gene Engineering of the Ministry of Education, Institute of Healthy Aging Research and SYSU‐BCM Joint Re‐ search Center, School of Life Sciences, Sun Yat‐sen University, Guangzhou, 510275, China, E‐mail: J. H. (e‐ mail:
[email protected]) and Z. S. (e‐mail:
[email protected]). Received November 17, 2014; accept‐ ed for publication February 18, 2015; available online without subscription through the open access option. ©AlphaMed Press 1066‐5099/2015/$30.00/0 This article has been accepted for pub‐ lication and undergone full peer review but has not been through the copyedit‐ ing, typesetting, pagination and proof‐ reading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1002/stem.2000
Actl6a Protects Embryonic Stem Cells from Dif‐ ferentiating into Primitive Endoderm WEISI LU1,2,3, LEKUN FANG4, BIN OUYANG5, XIYA ZHANG1,2, SHAOQUAN ZHAN1,2, XUYANG FENG1,2, YAOFU BAI1,2, XIN HAN1,2, HYEUNG KIM3, QUANYUAN HE3, MA WAN3, FENG‐TAO SHI6, XIN‐HUA FENG7, DAN LIU3, JUNJIU HUANG1,2,*, ZHOU SONGYANG1,2,* Key words. Actl6a embryonic stem cells self‐renewal primitive endo‐ derm Tip60‐p400 complex ABSTRACT Actl6a (actin‐like protein 6A, also known as Baf53a or Arp4) is a subunit shared by multiple complexes including esBAF, INO80, and Tip60‐p400, whose main components (Brg1, Ino80, and p400 respectively) are crucial for the maintenance of embryonic stem cells (ESCs). However, whether and how Actl6a functions in ESCs has not been investigated. ESCs originate from the epiblast (EPI) that is derived from the inner cell mass (ICM) in blastocysts, which also give rise to primitive endoderm (PrE). The molecu‐ lar mechanisms for EPI/PrE specification remain unclear. In this report, we provide the first evidence that Actl6a can protect mouse ESCs (mESCs) from differentiating into PrE. While RNAi knockdown of Actl6a, which appeared highly expressed in mESCs and downregulated during differentiation, in‐ duced mESCs to differentiate towards the PrE lineage, ectopic expression of Actl6a was able to repress PrE differentiation. Our work also revealed that Actl6a could interact with Nanog and Sox2, and promote Nanog bind‐ ing to pluripotency genes such as Oct4 and Sox2. Interestingly, cells deplet‐ ed of p400, but not of Brg1 or Ino80, displayed similar PrE differentiation patterns. And mutant Actl6a with impaired ability to bind Tip60 and p400 failed to block PrE differentiation induced by Actl6a dysfunction. Finally, we showed that Actl6a could target to the promoters of key PrE regulators (e.g., Sall4 and Fgf4), repressing their expression and inhibiting PrE differ‐ entiation. Our findings uncover a novel function of Actl6a in mESCs, where it acts as a gatekeeper to prevent mESCs from entering into the PrE lineage through a Yin/Yang regulating pattern. STEM CELLS 2014; 00:000–000
STEM CELLS 2014;00:00‐00 www.StemCells.com
©AlphaMed Press 2014
2
INTRODUCTION Cells from the inner cell mass (ICM) can differentiate into either epiblast (EPI) or primitive endoderm (PrE) in blastocyst stage, with EPI giving rise to all three germ layers and PrE developing into the visceral (VE) and pa‐ rietal (PE) endoderm [1, 2]. PrE facilitates EPI patterning and gives rise to the yolk sac that provides nutrient ex‐ change for the developing embryo [3]. Derived from EPI, embryonic stem cells (ESCs) can propagate indefi‐ nitely (self‐renewal) and differentiate into all embryonic lineages (pluripotency), processes that are governed by the regulatory network anchored by core transcription factors including Nanog, Oct4, and Sox2 [4‐6]. These factors co‐occupy and repress a significant portion of developmental genes, forming feed‐forward loops to maintain ESCs [7]. For example, Nanog could maintain mESC self‐renewal in the absence of LIF and Nanog−/− embryos failed to generate viable epiblasts, suggesting that Nanog is essential for both mESC maintenance and ICM development [8, 9]. Nanog and Gata6, the markers for EPI and PrE, re‐ spectively, are coexpressed in all ICM cells at the early blastocyst stage, and become mutually exclusive during embryo development [10]. Therefore, the ICM contains Nanog‐expressing EPI progenitors and Gata6‐expressing PrE progenitors [11]. After the segregation of EPI and PrE is complete, PrE progenitors activate more PrE‐ related genes such as Gata4, Sox17, Sox7, Pdgfra, which contribute to PrE stabilization [12]. Currently, Nanog, Gata6, Sall4, and FGF signaling are most well character‐ ized factors for EPI/PrE segregation. Nanog‐/‐ mESCs differentiated into PrE‐like cells, whereas all ICM cells in Gata6‐/‐ embryos adopted an EPI fate [13, 14]. Sall4 was shown to regulate mESCs by co‐occupying multiple tar‐ get genes with Nanog/Oct4/Sox2, and maintain the PrE‐ derived extraembryonic endoderm (XEN) cells by con‐ trolling PrE‐associated genes such as Gata6, Gata4, Sox17, and Sox7 [15]. Deletion of FGF signaling genes such as Fgf4 and Fgfr2 in mouse embryos abrogated PrE formation [16, 17]. Conversely, addition of exogenous Fgf4 could convert ICM cells to Gata6‐expressing PrE progenitors [18]. Notably, a recent study showed that exogenous FGF4 could not rescue the PrE progenitors in Gata6 null mutant embryos, suggesting that Gata6 is likely to act at the top of the networks in PrE specifica‐ tion[19]. Therefore, how the coordinated action of dif‐ ferent factors helps to determine EPI/PrE entry remains to be established. In our study of potential regulators for mESC maintenance and PrE differentiation, we identified Actl6a, an actin‐related protein also known as Baf53a or Arp4, as a novel player in maintaining the undifferenti‐ ated state of mESCs. Actl6a is a subunit in the esBAF (Brg1), INO80 (Ino80), and Tip60‐p400 (p400) complex‐ es, which have been implicated in mESC maintenance. For instance, Brg1 can promote STAT3 binding to main‐ tain pluripotency, Ino80 facilitates pluripotent gene www.StemCells.com
activation, and p400 depleted cells exhibited gene ex‐ pression profiles that significantly overlapped with that of Nanog [20‐22]. Recent studies suggest that Actl6a is necessary for the maintenance of several adult stem cells: the neuronal progenitor BAF complex, which con‐ tains Actl6a, is essential for the self‐renewal capacity of neural stem cells; conditional knockout of Actl6a re‐ vealed its importance for maintaining adult hematopoi‐ etic stem cells; Actl6a could repress Klf4 induction to enforce the epidermal progenitor state [23‐25]. In addi‐ tion, homozygous deletion of Actl6a proved embryonic lethal at day E6.5 [24]. However, the precise function of Actl6a in mESCs and in cell fate decisions remains to be elucidated. Our findings indicate that Actl6a could on one hand regulate mESCs self‐renewal through working with Nanog and Sox2, and on the other prevent ESCs from differentiating into PrE by repressing Sall4 and Fgf4, a process likely achieved through the action of the Tip60‐p400 complex.
RESULTS Actl6a knockdown led to differentiation of mESCs to the PrE lineage Actl6a appears to exhibit the highest mRNA level in mESC lines based on BioGPS data (Supporting infor‐ mation Fig. S1A) (http://biogps.org/#goto=genereport&id=56456). To further examine Actl6a expression, we monitored Actl6a protein levels in mESCs following three different conditions for differentiation induction (Supporting in‐ formation Fig. S1B). As expected, Nanog protein expres‐ sion was downregulated in all three cases. And a similar reduction was also apparent for Actl6a, suggesting pos‐ sible ESC‐specific function for Actl6a. To gain insight into the function of Actl6a, we gen‐ erated mESCs transiently knocked down for Actl6a. All three siRNA oligos were able to effectively reduce Actl6a levels (>80%) (Fig.1A). Actl6a knockdown induced cell differentiation including flattened cell morphology and weaker alkaline phosphatase (AP) activities, with‐ out affecting cell viability (Fig. 1B and Supporting infor‐ mation Fig. S1C‐D). Further analyses by western blot, qRT‐PCR, and immunostaining revealed a corresponding decrease in the expression of key pluripotency factors including Nanog, Oct4, Sox2, Klf4, Tbx3, and Esrrb (Fig. 1A and Supporting information Fig. S1D‐E). These ob‐ servations suggested that Actl6a depletion in mESCs could trigger cell differentiation. Indeed, when we ex‐ amined the expression of various lineage markers in Actl6a knockdown cells (Fig. 1C), we observed the strongest induction in endoderm markers (Foxa2, Sox17), followed by ectoderm (Mash1) and mesoderm (Brachyury) markers, while ectoderm (Nestin), meso‐ derm (Goosecoid), and trophectoderm (Hand1 and Eomes) markers showed little or no induction (Fig.1C).
©AlphaMed Press 2014
3 Foxa2 and Sox17 are crucial regulators for endo‐ derm development [26, 27], and appear highly ex‐ pressed in primitive endoderm (PrE) as well [28, 29]. Therefore, we tested whether Actl6a depletion could lead to upregulation of other PrE‐associated genes. In‐ deed, qRT‐PCR results confirmed that the expression of most of the PrE markers examined (Gata6, Gata4, Sox7, Hnf4a, and Pdgfra) was dramatically upregulated (~30‐ 140 fold) in Actl6a‐depleted cells compared to control siRNA‐treated cells (Fig. 1D). Next, we carried out a 12‐ day embryoid body (EB) formation assay to recapitulate early mouse embryo development using cells depleted of Actl6a. Notably, cysts started to appear in siActl6a EBs at day 8 and continued to grow (Fig.1E). Cystic EBs have a typical PrE‐like morphology and contain a mix‐ ture of PrE‐derived visceral endoderm (VE) and parietal endoderm (PE) cells [30, 31]. When we analyzed these siActl6a EBs by qRT‐PCR, we found higher expression of key PrE markers (Gata6, Gata4, Sox7, and Pdgfra), while the expression patterns of other lineage markers re‐ mained similar to control EBs (Fig. 1F). Furthermore, we observed higher Actl6a protein levels in mESCs com‐ pared to PrE‐derived extraembryonic endoderm cells (XEN) or mouse embryonic fibroblasts (MEF) (Support‐ ing information Fig.S1F). Taken together, these data support the notion that Actl6a inhibition can lead mESCs to differentiate into the PrE lineage.
Forced Actl6a expression can prevent mESCs from differentiating into PrE Given that Actl6a inhibition induced PrE specification, we speculated that restoring Actl6a expression should block this process. To test this idea, we generated mESC lines that could be induced to express HA‐Flag tagged Actl6a (Actl6a‐HF) at levels comparable to endogenous Actl6a (Fig. 2A‐B). We confirmed that the morphology, AP activity, and expression profiles of lineage markers of the inducible Actl6a‐HF cells were similar to parental mESCs (Supporting information Fig. S2A‐B). These cells were then cultured in suspension without LIF for six days to allow EB formation. A portion of the EBs were then harvested for gene expression analysis; the re‐ maining EBs were transferred to gelatin‐coated dishes for an additional day of culture to determine the per‐ centages of differentiated vs. undifferentiated EBs (based on the migration of PrE cells from EBs) [32]. Without doxycycline (‐Dox), PrE cells became scattered following attachment and spread into a ring of epithelial cells, with strong staining of the PrE marker Gata4 in cells that had migrated away from EBs (Fig.2D‐E). With doxycycline (+Dox), cells were less dispersed, and the few cells that had migrated away from the periphery exhibited weak Gata4 staining (Fig. 2D‐E). In addition, Actl6a overexpression in the presence of doxycyline also repressed PrE markers expression, whereas the expression of other lineage markers was either similar to or higher than that in untreated EBs (‐Dox) (Fig. 2F). To further validate our results, two additional Actl6a‐HF clones (No.2 and No.3) were similarly ana‐ www.StemCells.com
lyzed (Supporting information, Figure S2C‐E). Again, EBs derived from these cells were defective in PrE differen‐ tiation (Supporting information, Figure S2F‐H), indicat‐ ing that maintaining Actl6a expression is sufficient to block mESC entry into the PrE lineage.
Actl6a works with Nanog for mESC self‐ renewal maintenance Our data indicate downregulated Nanog expression with prolonged Actl6a inhibition (Fig.1A, Supporting information Fig.S1D). Given that Nanog depletion is sufficient to induce ESCs to differentiate into PrE [33], we speculated that Actl6a might work in concert with Nanog in mESCs. We tested this idea by first examining whether these two proteins could interact with each other through co‐immunoprecipitation experiments using our inducible Actl6a‐HF cells. As shown in Fig. 3A, Actl6a‐HF was able to bring down endogenous Nanog. In addition, Actl6a could also interact with Sox2, but not Oct4 (Fig. 3A). When we knocked down Actl6a in mESCs stably expressing HA‐Flag tagged Nanog, ectopic ex‐ pression of Nanog could rescue the differentiation phe‐ notypes caused by Actl6a depletion (Fig. 3B‐D). To fur‐ ther probe the interaction between Actl6a and Nanog in ESCs, we examined the effect of Actl6a depletion on the recruitment of endogenous Nanog to its binding targets [34]. Nanog expression remained fairly steady with transient Actl6a knockdown (two days) (Fig. 3E, left), allowing us to carry out ChIP‐qPCR assays using anti‐ Nanog antibodies to assess Nanog occupancy. We found that Actl6a depletion significantly reduced locali‐ zation of Nanog to its target genes, including Nanog itself and other pluripotency factors Sox2, Oct4, and Jarid2 (Fig. 3E, right). Collectively, these results indicate that Actl6a works with Nanog in maintaining mESC self‐ renewal and promoting Nanog binding to its targeting genes.
Actl6a functions as a subunit of the Tip60‐ p400 complex in blocking ESCs differentiation Although the Tip60‐p400, esBAF, and INO80 complexes to which Actl6a belongs are important regulators for ESCs [20, 21, 35], whether and how they function in lineage specification is still poorly understood. We therefore set out to examine the effects of inhibiting the main subunit of each complex, p400 (Tip60‐p400), Brg1 (esBAF), and Ino80 (INO80). Knocking down each of the three genes was sufficient to induce altered cell morphology (Supporting information, Fig.S3A‐B) and decreased expression of pluripotency markers Nanog and Oct4 (data not shown) compared with control cells. However, only inhibition of p400, but not Brg1 or Ino80, significantly increased the expression of a number of key PrE markers (e.g., Gata6, Gata4, Sox7, Hnf4a, and Pdgfra) and gave rise to cystic EBs (Supporting infor‐ mation, Fig. S3C‐E). These observations implicate the Tip60‐p400 complex in the prevention process of PrE differentiation. ©AlphaMed Press 2014
4 Previous studies have shown that actin‐related pro‐ teins (ARPs) can bind directly to the HSA domain of sev‐ eral ATPases [36], and that the M3 mutant of human ACTL6A (E388A/R389A/R390A) was impaired in its abil‐ ity to bind TIP60 [37]. We therefore compared the abili‐ ties of wild‐type and mutant Actl6a to interact with Tip60 and the HSA domain of p400 (Fig. 4A). We found that wild‐type Actl6a could co‐precipitate with Tip60 and p400 HSA, while the M3 mutant was deficient in its interaction with both proteins (Fig. 4B‐C). We then gen‐ erated cells stably expressing RNAi‐resistant wild‐type or M3 mutant Actl6a and introduced Actl6a siRNA oli‐ gos into these cells (Supporting information, Fig.S3F). As expected, ectopic expression of wild‐type Actl6a was able to restore the morphology, AP activities, as well as pluripotency factors expression in siActl6a treated cells (Fig.4D, supporting information, Fig. S3G). In contrast, siAclt6a‐treated cells that expressed the M3 mutant continued to display weak AP staining and marker ex‐ pression patterns similar to control cells (Fig.4D, sup‐ porting information, Fig. S3G). We further compared the activity of wild‐type and mutant Actl6a in EB for‐ mation assays as described in Fig.2D. Following attach‐ ment on gelatin‐coated plates, ~80% of EBs derived from the GFP‐expressing control cells and nearly 65% of EBs derived from Actl6a M3 mutant expressing cells were differentiated; in comparison, only 35% of wild‐ type Actl6a expressing EBs had differentiated (Fig. 4E). These results combined underline the importance of Actl6a interaction with Tip60 and p400 in Actl6a‐ mediated maintenance of mESC pluripotency.
Actl6a prevents PrE differentiation through repressing Sall4 and Fgf4 How does Actl6a work to prevent PrE differentiation? Given that Sall4 and Fgf4 are essential for PrE develop‐ ment [18, 38], we speculated that Actl6a might regulate PrE differentiation through targeting these two genes and therefore performed ChIP‐qPCR using antibodies against endogenous Actl6a in mESCs. As shown in Fig. 5A, enrichment of Actl6a on both Sall4 and Fgf4 pro‐ moter regions was apparent. Notably, Actl6a bound to both the proximal and distal regions of the Sall4 pro‐ moter (Fig. 5A, left panel). Knocking down Actl6a in mESCs led to induction of both Sall4 and Fgf4, whereas knockdown of Sall4 or Fgf4 had little effect on Actl6a expression (Fig. 5B, C). To better understand the path‐ ways mediated by Actl6a, we assessed the impact of Sall4 and Fgf4 reduction on mESCs already depleted of Actl6a. As expected (Fig. 1D), Aclt6a knockdown alone resulted in an increase in PrE markers expression. How‐ ever, this induction was abolished when Sall4 or Fgf4 was also depleted (Fig. 5D, E), supporting the notion that Actl6a may repress PrE differentiation through in‐ hibition of Sall4 and Fgf4 activities. Finally, we per‐ formed EB formation assays using the double knock‐ down cells. Here, the double knockdown cells were sim‐ ilar to negative control cells, producing fewer and smaller cystic EBs than cells depleted for Actl6a alone www.StemCells.com
(Fig. 5F, red arrow). The above findings place Actl6a directly upstream of Sall4 and Fgf4, where it negatively regulates the differentiation of ESCs into the PrE lineage by repressing Sall4 and Fgf4. DISCUSSION Actin‐related proteins are often integrated into multiple complexes and participate in diverse cellular processes [39]. Notably, of all the actin‐related proteins [25], only Actl6a appears enriched in mESCs (http://biogps.org/#goto=genereport&id=56456). In line with Actl6a being essential for mouse embryogene‐ sis [24], our data indicate that Actl6a is an important player in maintaining ESCs and support a Yin‐Yang mod‐ el of Actl6a‐mediated ESC regulation (Fig. 6). Based on this model, Actl6a collaborates with pluripotency fac‐ tors such as Nanog and Sox2 to ensure ESC self‐renewal on one hand, and inhibits key PrE factors including Sall4 and Fgf4 on the other, functioning as a crucial compo‐ nent of the Tip60‐p400 complex to prevent differentia‐ tion into the PrE lineage (Fig.6). Notably, knocking out Actl6a in hematopoietic stem/progenitor cells impaired both cell proliferation and viability, while Actl6a stable knockdown affected proliferation but not survival in neural stem/progenitor cells [24, 40]. In our system, we did not observe significant changes in mESC viability with Actl6a knockdown, perhaps Actl6a exerts different functions in different types of stem cells. Actl6 deletion could induce ESCs quickly differentiat‐ ing into PrE‐like cells, and Nanog overexpression could rescue Actl6a knockdown induced PrE differentiation, which is in agreement with previous reports of Nanog being a key repressor of the PrE lineage [8, 41]. We showed that Actl6a can interact directly with Nanog and Sox2, but not with Oct4, the latter is contrary to a pre‐ vious report that identified Aclt6a as an Oct4‐associated protein in mESCs by mass spectrometry [42]. It is possi‐ ble that Oct4‐Actl6a interaction may be too weak for co‐IP analysis. Actl6a depletion impaired Nanog binding to its target loci such as Nanog, Oct4, Sox2, and Jarid2 (Fig.3F); however, no enrichment of Actl6a was detect‐ ed near Nanog promoter regions (data not shown). It is possible that Actl6a may enhance Sox2 binding to the Nanog promoter through its interaction with Sox2 [43, 44], which in turn upregulates Nanog expression and promotes Nanog binding to target genes. Though Actl6a is a subunit in the esBAF (Brg1), INO80 (Ino80), and Tip60‐p400 (p400) complexes, only knockdown of p400 (not Brg1 or Ino80) resulted in PrE differentiation phenotypes similar to Actl6a inhibition. In fact, Brg1 silencing was previously shown to drive ESCs to differentiate towards ectoderm and mesoderm lineages [45, 46], and Ino80 appeared to mainly act as a transcriptional activator for pluripotency genes [21]. These findings together imply that different complexes, despite sharing certain subunits, have distinct functions in controlling mESC self‐renewal and cell fate determi‐ nation. When we disrupted Actl6a interaction with ©AlphaMed Press 2014
5 Tip60 or p400, we abrogated its ability to rescue Actl6a knockdown cells, suggesting that Actl6a may function through facilitating the proper assembly of the Tip60‐ p400 complex. In addition, the ATPase activity of the RSC chromatin remodeling complex was shown to be impaired if its catalytic subunit Sth1 lacked the ARP‐ recruiting HSA domain [36]. It is therefore also possible that the Actl6a‐p400 HSA interaction may regulate the ATPase activity of p400, which in turn helps maintain the undifferentiated state of ESCs. We present evidence here that Actl6a could occupy the promoter regions of Sall4 and Fgf4 and inhibit their expression to prevent PrE differentiation. Because both Sall4 and Fgf4 were also targeted by p400 (based on a ChIP‐on‐chip study) [20], it is possible that the Actl6a‐ containing Tip60‐p400 complex acts as an important upstream regulator in determining cell fate. Interesting‐ ly, Sall4 was induced to a higher level than Fgf4 in Actl6a depleted cells, and Sall4 inhibition led to more dramatic repression of PrE genes. It was shown that Sall4 homozygous mutant mouse blastocysts failed to develop to EPI/PrE lineages and had impaired ability to generate ES or XEN cell lines [47]. In comparison, ICM from Fgf4 null mutant mouse blastocysts lacked the PrE layer but still capable of generating XEN cell lines in the presence of exogenous FGF [16]. Therefore, Sall4 may acts as a nodal point that governs initial cell fate deci‐ sions regarding PrE differentiation. The relationship between various PrE players warrants further investiga‐ tion. Recently, XEN cells have emerged as a useful model for studying PrE lineage development [48, 49], however, the process of generating XEN cells from embryos and ESCs remains time‐consuming [50]. Given our data that Actl6a knockdown significantly upregulated PrE genes in mESCs and promoted cystic EB formation, silencing Actl6a represents an attractive strategy for optimizing the generation of XEN cells. In addition, siRNA‐based approaches are also valuable for clinical applications, with relative ease of delivery and absence of permanent alterations of the genome [51]. For instance, dual‐ specific kinase MKK4 silencing could enhance liver re‐ generation capacity in mouse models [52]. As PrE also contributes to cardiac formation, our findings point to possible alternative methods such as silencing Actl6a to improve the efficiency of PrE differentiation into specif‐ ic cell types, which should benefit future regenerative medicine [53‐55]. CONCLUSIONS In summary, our data identify Actl6a, an actin‐like pro‐ tein, as an essential player for maintaining the undiffer‐ entiated state of mESCs. Actl6a is highly expressed in mESCs and downregulated during differentiation. RNAi knockdown of Actl6a in mESCs induced differentiation, especially towards the PrE lineage, and ectopic expres‐ sion of Actl6a or Nanog rescued such phenotypes. We showed that Actl6a could interact directly with Nanog www.StemCells.com
and promote its binding to target pluripotency genes. Notably, similar PrE differentiation results were ob‐ served with RNAi‐mediated inhibition of p400, but not Brg1 or Ino80, indicating that the Tip60‐p400 complex was involved in this process. Disrupting the interaction of Actl6a with either Tip60 or p400 impaired Actl6a function in mESCs. Finally, we showed that Actl6a could directly target and negatively regulate the expression of PrE regulators Sall4 and Fgf4. Our study indicates that Actl6a could protect mESCs by preventing their entry into the PrE lineage through a Yin‐Yang regulatory pro‐ cess.
MATERIALS AND METHODS Cell lines, vectors, and RNAi The mouse ESC lines used were AB2.2 (passage#18, Darwin Core facility, Baylor College of Medicine) and A172‐loxP (passage#17, gift from Dr. Thomas P. Zwaka). mESCs were maintained under feeder free conditions, on tissue culture dishes coated with 0.1% gelatin (Sig‐ ma‐Aldrich) in high glucose DMEM medium (Hyclone) that was supplemented with 15% (v/v) fetal bovine se‐ rum, β–mercaptoethanol (55 M), GlutaMax‐I supple‐ ment (2mM), MEM non‐essential amino acids (0.1 mM), and LIF (1,000 U/ml, Millipore). HEK293T cells were cul‐ tured as previously described [56]. XEN cells (pas‐ sage#3, gift from Dr. Malgorzata Borowiak) were cul‐ tured as previously described [57]. To induce differenti‐ ation, AB2.2 cells were cultured without LIF, with 1% dimethyl sulfoxide (DMSO) and in the absence of LIF, or with 5M retinoic acid (RA) and in the presence of LIF[58]. Sequences encoding murine Actl6a, Nanog, Tip60, and the HSA domain of p400 were PCR‐amplified from AB2.2 cells, and cloned into either a MSCV‐based retro‐ viral expression vector for N‐terminal tagging with HA‐ FLAG (under the control of EF1 promoter) or the pDEST‐27 vector (Invitrogen) for N‐terminal tagging with GST. Actl6a M3 mutant and RNAi‐resistant con‐ struct were generated by Quick Change site‐directed mutagenesis (Stratagene). The construct for inducible C‐ terminal HA‐Flag tagged Actl6a was generated by ligat‐ ing mouse Actl6a cDNA to HA‐Flag sequence and then cloned into the p2Lox targeting vector [59].The vector was then electroporated (500uF, 250V) into A172‐loxP cells (1X107) together with pSalk‐Cre (total DNA amount of 20g)[59]. Stable lines were established by selecting the cells withG418 (300g/ml). Individual clones were selected and expanded. mESCs were transfected with siRNA oligos targeting various genes using Lipofectamine 2000 (Invitrogen), and harvested 48 hours after the second round of trans‐ fection, unless indicated otherwise. The siRNA oligos were purchased from Sigma. Please see Supplementary information, Table 1 for siRNA sequences used in this study. ©AlphaMed Press 2014
6
Embryoid body (EB) formation assay and al‐ kaline phosphatase (AP) staining AP staining was performed using the alkaline phospha‐ tase detection kit (SCR004, Millipore). EB formation assays were carried out as previously described [60], with minor modifications. For cystic embryoid body formation, cells were seeded as hanging drops (20l) at a density of 4X104cells/ml and cultured without LIF for two days before being transferred to 100mm Petri dish‐ es and maintained in suspension for an additional 10 days. EBs were visualized or harvested at different time points. For EB formation assay using the inducible cell lines, cells were cultured with doxycycline (500ng/ml) for 48 hours before being transferred to a 10cm Petri dish at a density of 4X106 cells with medium containing doxycycline and without LIF for six days. For microscopy analysis, the EBs were transferred to gelatin‐coated tissue culture dishes and cultured for 24 hours prior to imaging.
Immunoprecipitation (IP), western blotting, immunostaining, antibodies HEK293T cells were transfected with various expression constructs and harvested 48 hours after transfection for IP analysis as previously described [61]. Briefly, cells were lysed in 1xNETN buffer (containing 1mM DTT and protease inhibitors (Gendepot, P3100‐010)). The super‐ natant was incubated with anti‐FLAG M2 agarose beads (Sigma, A2220) for >2 hours at 4C. The immunoprecipi‐ tates were then eluted in 2x Laemmli buffer (BioRad, 161‐0737). Whole cell lysates were also obtained by direct lysis of cells in 2xLaemmli buffer at 90C. All sam‐ ples were resolved by SDS‐PAGE and transferred onto polyvinylidene fluoride membranes (BioRad, 1620177) for blotting with appropriate antibodies. For im‐ munostaining, cells grown on glass cover slips were fixed in 4% paraformaldehyde, permeablized with 0.2% Triton X‐100, and analyzed as described previously [61]. The following antibodies were used for IP and west‐ ern blotting: anti‐Actl6a (A301‐391A), anti‐GST‐HRP (A190‐122P), and anti‐Nanog (A300‐397A) (Bethyl La‐ boratories); anti‐Ruvbl1 (12300S, Cell Signaling); anti‐ Oct4 (sc‐5279), anti‐Gata4 (sc‐9053), and anti‐GAPDH (sc‐25778) (Santa Cruz); and anti‐FLAG (A8592) and an‐ ti‐FLAG M2 affinity gel (A2220) (Sigma). For im‐ munostaining, the antibodies included anti‐Actl6a (A301‐391A)(Bethyl Laboratories), anti‐Oct4 (sc‐5279) and anti‐Gata4 (sc‐9053) (Santa Cruz).
Quantitative Real‐Time PCR (qRT‐PCR) qRT‐PCR was performed as previously described [62]. Briefly, total RNA was isolated using the RNeasy Mini Kit (Qiagen). Following cDNA synthesis using the iScript Select cDNA Synthesis Kit (BioRad), quantitative PCR was performed using SYBR Green PCR Master Mix (Ap‐ www.StemCells.com
plied Biosystems) with an ABI StepOnePlus Real‐Time PCR System. Results were normalized to GAPDH tran‐ scripts and analyzed using the delta‐delta Ct method to calculate the relative fold change in gene expression. The specific primer sequences are listed in Supplemen‐ tary information, Table 2. Statistical analyses were per‐ formed using GraphPad Prism5.
Chromatin Immunoprecipitation (ChIP) assay ChIP assays were performed using the fast ChIP proto‐ col [63]with minor modifications. Briefly, ESCs were fixed for 15min at room temperature with 1.42% for‐ maldehyde. The sonicated chromatin was incubated with antibodies at 4C overnight with gentle rotation. The immunoprecipitates were mixed with protein A‐ agarose beads (sc2001, Santa Cruz) and rotated for 2hrs at 4C. Chelex 100 resin (BioRad) was then added to the immunoprecipitates and input DNA samples, followed by incubation at 100C for 10 minutes to reverse cross‐ linking. After proteinase K treatment for 30 minutes, DNA was recovered and subjected to real‐time PCR analysis using appropriate primers listed in the Supple‐ mentary information, Table 3. The following antibodies were used for ChIP assay: anti‐Nanog (BL1663)(Bethyl Laboratories), and anti‐Actl6a(ab3882) and Rabbit IgG (ab37415)(Abcam). Statistical analyses were performed using GraphPad Prism5. ACKNOWLEDGMENTS This work was supported by National Basic Research Program (973 Program 2010CB945400 and 2012CB911201), National Natural Science Foundation of China (NSFC 91019020, 91213302 and 31371508), Zhujiang Program of Science and Technology Nova in Guangzhou (2011J2200082); NIGMS GM081627. We would also like to acknowledge the support of the C‐ BASS/GRSA shared resource of the Dan L. Duncan Can‐ cer Center (P30CA125123). We thank Dr. Thomas P. Zwaka for generously providing the A172‐LoxP ES cell line and p2Lox and pSalk‐Cre vectors, and Dr. Mal‐ gorzata Borowiak for the XEN cell line. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST The authors indicate no potential conflicts of interest.
AUTHOR CONTRIBUTIONS W.L., M.W, and Z.S.: designed experiments; W.L., L.F., B.O., X.Z., S.Z., X.F., Y.B., X.H., H.K., Q.H., J.H., and F.S.: performed experiments and analyzed data; W.L., D.L., J.H. and Z.S.: wrote and edited the manuscript.
©AlphaMed Press 2014
7
REFERENCES 1 Kim PT, Ong CJ. Differentiation of definitive endoderm from mouse embryonic stem cells. Results and problems in cell differentiation. 2012;55:303‐319. 2 Fernandez‐Diaz LC, Laurent A, Girasoli S et al. The absence of Prep1 causes p53‐ dependent apoptosis of mouse pluripotent epiblast cells. Development. 2010;137:3393‐ 3403. 3 Artus J, Chazaud C. A close look at the mammalian blastocyst: epiblast and primitive endoderm formation. Cellular and molecular life sciences : CMLS. 2014. 4 Young RA. Control of the embryonic stem cell state. Cell. 2011;144:940‐954. 5 Marson A, Levine SS, Cole MF et al. Connecting microRNA genes to the core transcriptional regulatory circuitry of embryonic stem cells. Cell. 2008;134:521‐533. 6 Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature. 1981;292:154‐156. 7 Boyer LA, Lee TI, Cole MF et al. Core transcriptional regulatory circuitry in human embryonic stem cells. Cell. 2005;122:947‐956. 8 Mitsui K, Tokuzawa Y, Itoh H et al. The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell. 2003;113:631‐642. 9 Chambers I, Colby D, Robertson M et al. Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells. Cell. 2003;113:643‐655. 10 Plusa B, Piliszek A, Frankenberg S et al. Distinct sequential cell behaviours direct primitive endoderm formation in the mouse blastocyst. Development. 2008;135:3081‐ 3091. 11 Rossant J, Chazaud C, Yamanaka Y. Lineage allocation and asymmetries in the early mouse embryo. Philosophical transactions of the Royal Society of London Series B, Biological sciences. 2003;358:1341‐ 1348; discussion 1349. 12 Stephenson RO, Rossant J, Tam PP. Intercellular interactions, position, and polarity in establishing blastocyst cell lineages and embryonic axes. Cold Spring Harbor perspectives in biology. 2012;4. 13 Chambers I, Silva J, Colby D et al. Nanog safeguards pluripotency and mediates germline development. Nature. 2007;450:1230‐1234. 14 Bessonnard S, De Mot L, Gonze D et al. Gata6, Nanog and Erk signaling control cell fate in the inner cell mass through a tristable regulatory network. Development. 2014;141:3637‐3648. 15 Lim CY, Tam WL, Zhang J et al. Sall4 regulates distinct transcription circuitries in different blastocyst‐derived stem cell lineages. Cell stem cell. 2008;3:543‐554. 16 Kang M, Piliszek A, Artus J et al. FGF4 is required for lineage restriction and salt‐and‐ pepper distribution of primitive endoderm factors but not their initial expression in the mouse. Development. 2013;140:267‐279. 17 Arman E, Haffner‐Krausz R, Chen Y et al. Targeted disruption of fibroblast growth factor (FGF) receptor 2 suggests a role for FGF
www.StemCells.com
signaling in pregastrulation mammalian development. Proceedings of the National Academy of Sciences of the United States of America. 1998;95:5082‐5087. 18 Yamanaka Y, Lanner F, Rossant J. FGF signal‐dependent segregation of primitive endoderm and epiblast in the mouse blastocyst. Development. 2010;137:715‐724. 19 Schrode N, Saiz N, Di Talia S et al. GATA6 levels modulate primitive endoderm cell fate choice and timing in the mouse blastocyst. Developmental cell. 2014;29:454‐467. 20 Fazzio TG, Huff JT, Panning B. An RNAi screen of chromatin proteins identifies Tip60‐ p400 as a regulator of embryonic stem cell identity. Cell. 2008;134:162‐174. 21 Wang L, Du Y, Ward JM et al. INO80 Facilitates Pluripotency Gene Activation in Embryonic Stem Cell Self‐Renewal, Reprogramming, and Blastocyst Development. Cell stem cell. 2014;14:575‐ 591. 22 Ho L, Miller EL, Ronan JL et al. esBAF facilitates pluripotency by conditioning the genome for LIF/STAT3 signalling and by regulating polycomb function. Nature cell biology. 2011;13:903‐913. 23 Yoo AS, Sun AX, Li L et al. MicroRNA‐ mediated conversion of human fibroblasts to neurons. Nature. 2011;476:228‐231. 24 Krasteva V, Buscarlet M, Diaz‐Tellez A et al. The BAF53a subunit of SWI/SNF‐like BAF complexes is essential for hemopoietic stem cell function. Blood. 2012;120:4720‐4732. 25 Bao X, Tang J, Lopez‐Pajares V et al. ACTL6a enforces the epidermal progenitor state by suppressing SWI/SNF‐dependent induction of KLF4. Cell stem cell. 2013;12:193‐203. 26 Ang SL, Wierda A, Wong D et al. The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development. 1993;119:1301‐1315. 27 Kanai‐Azuma M, Kanai Y, Gad JM et al. Depletion of definitive gut endoderm in Sox17‐null mutant mice. Development. 2002;129:2367‐2379. 28 Hwang JT, Kelly GM. GATA6 and FOXA2 regulate Wnt6 expression during extraembryonic endoderm formation. Stem cells and development. 2012;21:3220‐3232. 29 Artus J, Piliszek A, Hadjantonakis AK. The primitive endoderm lineage of the mouse blastocyst: sequential transcription factor activation and regulation of differentiation by Sox17. Developmental biology. 2011;350:393‐404. 30 Magyar JP, Nemir M, Ehler E et al. Mass production of embryoid bodies in microbeads. Annals of the New York Academy of Sciences. 2001;944:135‐143. 31 Martin GR, Wiley LM, Damjanov I. The development of cystic embryoid bodies in vitro from clonal teratocarcinoma stem cells. Developmental biology. 1977;61:230‐244. 32 Shimoda M, Kanai‐Azuma M, Hara K et al. Sox17 plays a substantial role in late‐stage differentiation of the extraembryonic endoderm in vitro. Journal of cell science. 2007;120:3859‐3869. 33 Hyslop L, Stojkovic M, Armstrong L et al. Downregulation of NANOG induces
differentiation of human embryonic stem cells to extraembryonic lineages. Stem Cells. 2005;23:1035‐1043. 34 Wu Q, Chen X, Zhang J et al. Sall4 interacts with Nanog and co‐occupies Nanog genomic sites in embryonic stem cells. The Journal of biological chemistry. 2006;281:24090‐24094. 35 Kidder BL, Palmer S, Knott JG. SWI/SNF‐ Brg1 regulates self‐renewal and occupies core pluripotency‐related genes in embryonic stem cells. Stem Cells. 2009;27:317‐328. 36 Szerlong H, Hinata K, Viswanathan R et al. The HSA domain binds nuclear actin‐ related proteins to regulate chromatin‐ remodeling ATPases. Nature structural & molecular biology. 2008;15:469‐476. 37 Nishimoto N, Watanabe M, Watanabe S et al. Heterocomplex formation by Arp4 and beta‐actin is involved in the integrity of the Brg1 chromatin remodeling complex. Journal of cell science. 2012;125:3870‐3882. 38 Zhang J, Tam WL, Tong GQ et al. Sall4 modulates embryonic stem cell pluripotency and early embryonic development by the transcriptional regulation of Pou5f1. Nature cell biology. 2006;8:1114‐1123. 39 Dion V, Shimada K, Gasser SM. Actin‐ related proteins in the nucleus: life beyond chromatin remodelers. Current opinion in cell biology. 2010;22:383‐391. 40 Lessard J, Wu JI, Ranish JA et al. An essential switch in subunit composition of a chromatin remodeling complex during neural development. Neuron. 2007;55:201‐215. 41 Frankenberg S, Gerbe F, Bessonnard S et al. Primitive endoderm differentiates via a three‐step mechanism involving Nanog and RTK signaling. Developmental cell. 2011;21:1005‐1013. 42 Ding J, Xu H, Faiola F et al. Oct4 links multiple epigenetic pathways to the pluripotency network. Cell research. 2012;22:155‐167. 43 Gagliardi A, Mullin NP, Ying Tan Z et al. A direct physical interaction between Nanog and Sox2 regulates embryonic stem cell self‐ renewal. The EMBO journal. 2013;32:2231‐ 2247. 44 Rodda DJ, Chew JL, Lim LH et al. Transcriptional regulation of nanog by OCT4 and SOX2. The Journal of biological chemistry. 2005;280:24731‐24737. 45 Singhal N, Esch D, Stehling M et al. BRG1 Is Required to Maintain Pluripotency of Murine Embryonic Stem Cells. BioResearch open access. 2014;3:1‐8. 46 Ho L, Ronan JL, Wu J et al. An embryonic stem cell chromatin remodeling complex, esBAF, is essential for embryonic stem cell self‐renewal and pluripotency. Proceedings of the National Academy of Sciences of the United States of America. 2009;106:5181‐ 5186. 47 Elling U, Klasen C, Eisenberger T et al. Murine inner cell mass‐derived lineages depend on Sall4 function. Proceedings of the National Academy of Sciences of the United States of America. 2006;103:16319‐16324. 48 Artus J, Douvaras P, Piliszek A et al. BMP4 signaling directs primitive endoderm‐ derived XEN cells to an extraembryonic
©AlphaMed Press 2014
8 visceral endoderm identity. Developmental biology. 2012;361:245‐262. 49 Kruithof‐de Julio M, Alvarez MJ, Galli A et al. Regulation of extra‐embryonic endoderm stem cell differentiation by Nodal and Cripto signaling. Development. 2011;138:3885‐3895. 50 Niakan KK, Schrode N, Cho LT et al. Derivation of extraembryonic endoderm stem (XEN) cells from mouse embryos and embryonic stem cells. Nature protocols. 2013;8:1028‐1041. 51 Khanjyan MV, Yang J, Kayali R et al. A high‐content, high‐throughput siRNA screen identifies cyclin D2 as a potent regulator of muscle progenitor cell fusion and a target to enhance muscle regeneration. Human molecular genetics. 2013;22:3283‐3295. 52 Wuestefeld T, Pesic M, Rudalska R et al. A Direct in vivo RNAi screen identifies MKK4 as a key regulator of liver regeneration. Cell. 2013;153:389‐401. 53 Holtzinger A, Rosenfeld GE, Evans T. Gata4 directs development of cardiac‐
inducing endoderm from ES cells. Developmental biology. 2010;337:63‐73. 54 Brown K, Doss MX, Legros S et al. eXtraembryonic ENdoderm (XEN) stem cells produce factors that activate heart formation. PloS one. 2010;5:e13446. 55 Liu W, Brown K, Legros S et al. Nodal mutant eXtraembryonic ENdoderm (XEN) stem cells upregulate markers for the anterior visceral endoderm and impact the timing of cardiac differentiation in mouse embryoid bodies. Biology open. 2012;1:208‐219. 56 Shi FT, Kim H, Lu W et al. Ten‐eleven translocation 1 (Tet1) is regulated by O‐linked N‐acetylglucosamine transferase (Ogt) for target gene repression in mouse embryonic stem cells. The Journal of biological chemistry. 2013;288:20776‐20784. 57 Niakan KK, Ji H, Maehr R et al. Sox17 promotes differentiation in mouse embryonic stem cells by directly regulating extraembryonic gene expression and indirectly antagonizing self‐renewal. Genes & development. 2010;24:312‐326.
58 Khalfallah O, Rouleau M, Barbry P et al. Dax‐1 knockdown in mouse embryonic stem cells induces loss of pluripotency and multilineage differentiation. Stem Cells. 2009;27:1529‐1537. 59 Fujita J, Crane AM, Souza MK et al. Caspase activity mediates the differentiation of embryonic stem cells. Cell stem cell. 2008;2:595‐601. 60 Price FD, Yin H, Jones A et al. Canonical Wnt signaling induces a primitive endoderm metastable state in mouse embryonic stem cells. Stem Cells. 2013;31:752‐764. 61 Chen LY, Liu D, Songyang Z. Telomere maintenance through spatial control of telomeric proteins. Molecular and cellular biology. 2007;27:5898‐5909. 62 Liu Y, Kim H, Liang J et al. The death‐ inducer obliterator 1 (Dido1) gene regulates embryonic stem cell self‐renewal. The Journal of biological chemistry. 2014;289:4778‐4786. 63 Nelson JD, Denisenko O, Bomsztyk K. Protocol for the fast chromatin immunoprecipitation (ChIP) method. Nature protocols. 2006;1:179‐185.
See www.StemCells.com for supporting information available online. STEM CELLS ; 00:000–000
www.StemCells.com
©AlphaMed Press 2014
9
Figure 1. Knockdown Actl6a in mouse embryonic stem cells (mESCs) leads to primitive endoderm (PrE) differentia‐ tion. (A‐D) mESCs were cultured in the presence of LIF, transfected with control oligos (siControl) or three different oligos targeting Actl6a, and analyzed in the following assays. (A) Cells were western blotted for Actl6a to assess knockdown efficiency, and Nanog and Oct4 expression. GAPDH was used as a loading control. (B) Cells were visual‐ ized by phase‐contrast microscopy (upper panel) and stained for alkaline phosphatase (AP) activities (lower panel). (C) Cells were analyzed by qRT‐PCR for the expression of the indicated lineages markers. Error bars indicate SD (n=3). *, p