Dppa3 facilitates self-renewal of embryonic stem cells by stabilization of pluripotent factors

Developmental pluripotency-associated 3 (Dppa3, also called Stella or PGC7) is a principal maternal protein specially expressed in pre-implantation embryos, embryonic stem cells (ES cells) and primordial germ cells (PGCs). It plays critical role in the regulating of DNA methylation in zygotes and oocytes. However, the effect of Dppa3 in ES cells on the stability of proteins is still unclear. In this study, we first identified the potential interacting proteins with Dppa3 using immunoprecipitation-mass spectrometry (IP-MS). After GO analysis, we further constructed Dppa3-silenced ES cells and ES cell lines overexpressing with different lengths of Dppa3 to explore the mechanisms of Dppa3 on protein stability. IP-MS results showed that Dppa3 interacted with quite a few subunits of 26S proteasome. Full length of Dppa3 stabilized Uhrf1 and Nanog by inhibiting its degradation. Silencing Dppa3 promoted degradation of Nanog protein. Our results indicated that Dppa3 safeguard the stability of Uhrf1 and Nanog by inhibiting proteasome-associated degradation in ES cells. These findings shed light on new function of Dppa3 in maintaining stability of proteins and provides a valuable resource for understanding the roles of Dppa3 in embryonic stem cells.


Introduction
Embryonic stem cells (ES cells) are derived from the inner cell mass (ICM) of the blastocyst and can be maintained pluripotency indefinitely [8]. With the ability of self-renewal and potential differentiation to multiple cell lineages, human ES cells hold great promise for the treatment of patients with degenerative disorders [33]. Developmental pluripotency-associated protein 3 (Dppa3, also known as Stella or PGC7) is initially identified as a maternal gene preferentially expressed in germ cells and early embryos [29]. Dppa3 plays crucial roles in early embryonic development by modulating transcriptional program and regulating epigenetic modification [20,23]. Dppa3 −/− male mice showed normal fertility when mated with wild-type (Dppa3 +/+ ) or heterogenetic (Dppa3 ± ) females. However, the pre-implantation development of embryos derived from Dppa3 −/− oocytes is abnormal, suggesting that maternally supplied Dppa3 is important in the cleavage stages of pre-implantation development [26]. However, the epigenetic behavior of Dppa3 is quite different on paternal genome and maternal genome in the zygotes. After fertilization, paternal genome is demethylated by Tet3 (ten-eleven translocation 3). Dppa3 specifically binds to the maternal genome on histone H3K9 dimethylation (H3K9me2) and protects the maternal genome against Tet3-mediated conversion of 5mC to 5hmC during the wave of DNA demethylation in Open Access *Correspondence: liuna@nankai.edu.cn † Shuang Zhao, Chuanyu Zhang and Jia Xu have contributed equally to this work 1 School of Medicine, Nankai University, 94# Weijin Road, Tianjin 300071, China Full list of author information is available at the end of the article early embryogenesis [22,23,26,29] and the mechanism of Dppa3 on DNA methylation in zygotes is distinguish different with that in oocytes, suggesting the complexity functions of Dppa3 in different development stages [14]. Shin et al. found that the cleavage of full length of Dppa3 is also necessary for early embryonic development [32].
In addition to the function of Dppa3 in embryogenesis, it is also heterogeneously expressed in embryonic stem cells. Dppa3 is required for generation of fully reprogrammed induced pluripotent stem cells (iPS cells), and increases the efficiency of nuclear transfer reprogramming [5,12,14,22,36,41]. With the development of single-cell analysis technology, ES cells display heterogeneous and metastable state that fluctuate between naïve state and primed state [21,24]. A defined medium known as "2i medium" (MEK inhibitor: PD0325901, and GSK3 inhibitor: CHIR99021) contributes to the establishment of a ground pluripotent state in ES cells, which renders them similar to the cells of the inner cell mass [25,43]. In contrast, primed ES cells are similar to epiblast stem cells (EpiSCs), which can be induced by FGF2 and Activin A (FGF + ActA) [6]. Compared with primed ES cells, ground or naïve ES cells have a higher level of Dppa3 expression [12,28]. Epigenome in Dppa3-positive ES cells are similar to that in ICM, characterized with hypomethylated DNA [12,28,46]. The function pattern of Dppa3 in ES cells is different from embryo. In ES cells and somatic cells, Dppa3 causes the demethylation of whole genome [7,9]. Dppa3 is also critical for the maintenance of Dlk-Dio3 imprinting [13,41].
The stability of proteins in all activities remains controllable balance. There are two main degradation systems in cellular activities. One is autophagy-lysosome system. The other is ubiquitin-proteasome system (UPS) [27]. UPS plays a significant role in zygotic gene activation, DNA repair, and developmental programs [4,10,30,31] Dppa3 is involved in degradation of proteins in early embryogenesis, but the role of which in regulating protein stability in ES cells remains poorly understood. Here, we demonstrated that Dppa3 regulates proteins stability in ES cells. Specifically, we found that the accumulation of Uhrf1 in ES cells with full length of Dppa3 overexpression. Furthermore, Dppa3 overexpression also increased Nanog in ES cells. This provides new insights into the function of Dppa3 in ES cells.

Plasmid construction and establishment of Dppa3 overexpression cell lines
We used pCMV6-entry (OriGene Technologies) to amplify Flag sequence first and then cloned it into plasmid plch3.7. Primers were linked with restriction enzymes Nhe I (Takara) and Psp5 II (Takara) for further enzymatic digestion. Then, introduce plch3.7-Flag as scaffold for plasmids construction. Mouse Dppa3 CDS contains 450 base pairs. 1-447, 1-180, and 181-447 base pairs (except for the last termination codon) were cloned into plch3.7-Flag with restriction enzymes Nhe I (Takara) and Xho I (Takara). Primers of three different lengths of Dppa3 are shown in Additional file 1: Table S1. All three different lengths of Dppa3 were followed by Flag sequences. We named 1-447, 1-180, and 181-447 of Dppa3 plasmids as plDf, plD1, and plD61. Plasmid plch3.7 is a load control compared to other three plasmids. D3 wild-type embryonic stem cell lines and 293 T cell lines were transfected with plasmids mentioned above using Lipo 2000 reagent (Invitrogen) with serum-free medium Opti-MEM I ® according to the manufacturer's instructions. The cells were cultured by nonpenicillin or streptomycin culture medium for 2 days. Transfected ES cells and 293 T cells were then selected by puromycin (0.5 μg/ml) for 10-14 days. For transfected ES cells, individual clones were picked for prolong culture and analysis. For transfected 293 T cells lines, the protein level of Flag was evaluated by western blot assay.

Generation of Dppa3 KD ES cells
shRNAs (Additional file 1: Table S2) were synthesized and cloned into pSIREN-RetroQ vector. The reconstructed vectors were transfected into ESCs using Lipofectamine 2000 (Invitrogen). ESCs were selected by puromycin (1.5 μg/ml) for 7 days, and positive clones were next used for analysis.

Real-time polymerase chain reaction
Total RNA was isolated by TRIzol ™ Reagent (Thermo Fisher Scientific ™ ) and reversely transcribed into cDNA from 1 μg of total RNA using Transcriptor First Strand cDNA Synthesis Kit (Roche). Real-time polymerase chain reaction (RT-PCR) was performed with the Opticon ® System (Bio-Rad, Hercules, CA) using Hieff ™ qPCR SYBR ® Green Master Mix (No Rox) (Yeasen) to amplify expressions of relative genes. Relative gene expression fold change was identified using the 2 −ΔΔCt method. The sequences of primers used in this study are shown in Additional file 1: Table S3.

Alkaline phosphatase staining
ES cells were seeded at a density of 5 × 10 4 cells per well into 24-well plates and cultured for one passage. Alkaline phosphatase (AP) staining was performed with the Alkaline Phosphatase Staining Kit (Yeasen). The cells were fixed with 4% paraformaldehyde for 10 min after washing with PBS and incubated with AP solution at room temperature for 10-15 min. PBS was then added to terminate the reaction. The stained cells were observed and recorded for the number of colonies under the microscope. Data were analyzed by counting at least 300 colonies for each sample.

CCK-8 proliferation assay
To investigate whether overexpressing different lengths of Dppa3 could affect the proliferation of ES cells, the ES cells were seeded into 96-well plates at a density of 10 4 cells per well for each overexpression cell lines. After 12 h, the reagent of Cell Counting Kit-8 (Solarbio) was added to the medium and incubated for an additional 1 h. The absorbance value of each well was recorded at 450 nm using a microplate reader (Thermo Labsystems, Vantaa, Finland).

Immunofluorescence staining
Cells were fixed in 4% paraformaldehyde for 20 min and incubated at 4 °C in blocking buffer (PBS containing 5% BSA). Cells were incubated in the presence of primary antibodies including Dppa3 (1:1000; Abcam) and Nanog (1:1000; Bethyl) at 4 °C overnight and then washed three times in PBS. Cells were then incubated with Alexa Fluor 488 secondary antibody (1:1000; Invitrogen) for 2 h at room temperature. Nuclei were stained with DAPI (1:10,000; Invitrogen).

Hematoxylin-Eosin staining
Dppa3 overexpression ES cells and Dppa3 knockdown ES cells were injected into nude mice for teratoma analysis. After 4 weeks, the nude mice were killed and the teratoma was collected for the HE staining according to standardized procedures. Teratoma was washed in PBS, fixed in 4% paraformaldehyde (pH 7.4), and embedded in paraffin. Teratoma was stained using hematoxylin for 5-10 min, rinsed with distilled water for 1 min, and separated color with 0.5% alcohol hydrochloric acid at 37 °C. Teratoma was then stained with eosin for 2-5 min, dehydrated using graded ethanol, vitrified by dimethylbenzene, and mounted with neutral balsam at 37 °C.

Statistical analysis
Data were analyzed by Student's t test using SPSS software. Statically significant differences were defined as *p < 0.05, **p < 0.01, ***p < 0.001. The results were shown as mean ± SEM.

Identify Dppa3-interacted proteins in embryonic stem cells
Dppa3 is heterogenetic expressed in ES cells and plays important role in somatic cell reprogramming. To further understand the mechanisms of Dppa3 in ES cells, we firstly identified the proteins which interact with Dppa3 using immunoprecipitation-mass spectrum (IP-MS) after the cell lysis was enriched by Dppa3 antibody (Fig. 1A). GO results show that the Dppa3-interacted proteins were mainly involved in post-translational modification, ribosomal structure and biogenesis, protein transportation, coenzyme transport and metabolism, and energy production (Additional file 1: Fig.  S1A). Using STRING website, we found that several Dppa3-interacted proteins play critical important roles in the pluripotency of ES cells, such as Dppa2/4, Uhrf1, Pou5f1, and Nanog [15,19] (Additional file 1: Fig. S1C). We analyzed the function enrichment of Dppa3-interated proteins using DAVID website. The results showed that Dppa3 is involved in the regulation of protein stabilization (Fig. 1B). KEGG analysis turned out that Dppa3 interacts with many of proteasome-associated proteins (Fig. 1D) as well as proteins function in posttranslational modification (Additional file 1: Fig. S1B). For instance, Psmc5, Nsflc, Psmb6, Psmd7, and Ube2k [16,44] are components of ubiquitin proteasome system, which participate in UPS-mediated degradation of protein (Fig. 1C). Our results showed that Dppa3 also interacts with quite a few subunits of 26S proteasome (Fig. 1E), suggesting Dppa3's potential role in regulating protein stability in ES cells.

Full length of Dppa3 stabilizes Uhrf1 in ES cells
To investigate the role of Dppa3 on protein stability, we established Dppa3 overexpression ES cells (Dppa3 OE) and found that Uhrf1 was significantly increased in Dppa3 OE cells (Fig. 2A). However, the mRNA level of Uhrf1 was reduced in Dppa3 OE cells compared to control ES cells (Fig. 2B), suggesting that Dppa3 might regulate Uhrf1 at protein level. Immunofluorescence results also confirmed that Dppa3 overexpression increased Uhrf1 positive cells (Fig. 2C). The expression of Dppa3 was upregulated in ground-state ES cells which were induced by 2i/LIF cultural medium and Uhrf1 has similar expression pattern with Dppa3, further indicating the potential regulation function of Dppa3 to Uhrf1 (Additional file 1: Fig. S2B and C).
Next, we investigated the effects of Dppa3 on protein synthesis and degradation process of Uhrf1 using MG132 (a proteasome inhibitor) and CHX (cycloheximide, a eukaryote protein synthesis inhibitor), respectively. When the proteasome was inhibited using MG132, Uhrf1 generally accumulated from 0 to 1 h in Dppa3 overexpression group (Fig. 2D). After CHX treatment, Uhrf1 degradation was significantly inhibited in Dppa3 OE ES cells (Fig. 2E).
IP-MS results showed that Dppa3 might involve in UPS-mediated protein degradation. The whole length of Dppa3 (150 amino acids) is cleaved into two truncated forms (N-terminus of Dppa3 (1-60 AA) and C-terminus of Dppa3 (61-150 AA)) in embryogenesis after fertilization. N-terminus of Dppa3 participates in vesicle trafficking along with the early embryo development, and the C-terminus of Dppa3 is easily degraded by UPS after cleavage [32], as seen in our results. Next, we constructed ES cell lines overexpression with full length of Dppa3 (Dppa3 1−150 ) and two truncated versions of Dppa3 (Dppa3 1−60 and Dppa3 61−150 ), respectively (Fig. 3A). It is shown in western blot that only full length of Dppa3 increased the protein level of Uhrf1 (Fig. 3B). Next, we used MG132 and CHX to study the regulation mode of three different lengths of Dppa3 on Uhrf1. Uhrf1 protein in MG132 (20 mM, 1 h) treated group overexpression of full length Dppa3 was higher than the other two truncated groups (Fig. 3C). When treat -Dppa3 1−150 group with CHX in concentration of 2 mM for 1 h, Uhrf1 protein still showed higher level than control group (Fig. 3D), suggesting only the full length of Dppa3 could stabilize Uhrf1 protein. This conclusion is also validated in 293 T cell lines (Additional file 1: Fig. S3A and B).

Dppa3 stabilizes Nanog from degradation in embryonic stem cells
Stability of proteins regulated by Dppa3 not only observed in Uhrf1, but also discovered in pluripotent transcriptional factor Nanog. Dppa3 overexpression increased Nanog expression in ES cells (Additional file 1: Fig. S4A). As shown in Fig. 4A, after CHX treatment, the protein level of Nanog in control group was extremely degraded within 1 h. However, overexpression of Dppa3 in ES cells significantly retarded the degradation of Nanog protein. This result indicated that Dppa3 stabilized the Nanog in ES cells. Immunofluorescence results indicated that Dppa3 was mainly expression in nucleus of ES cells (Fig. 4B), and colocalized with Nanog in nuclear (Fig. 4C). We also have identified the interaction between Dppa3 and Nanog by means of co-IP when used antibody against Nanog and Dppa3, respectively (Fig. 4D, E), suggesting Dppa3 might stabilize Nanog by direct or indirect interaction. Nanog is a well-known pluripotent factor. We next investigate the pluripotency of Dppa3 OE ES cells. Our data showed that full length of Dppa3 could increase the expression of ground-state marker genes (Additional file 1: Fig. S4C), while all three different lengths of Dppa3 have no effect on proliferation (Additional file 1: Fig. S4B). Furthermore, full length of Dppa3 could increase the percentage of AP (alkaline phosphatase, a marker of pluripotency) positive clones (Additional file 1: Fig. S4D); these results indicate Dppa3, especially the full length of Dppa3, facilitates the pluripotency maintenance of ES cells.

Silencing of Dppa3 promotes degradation of Nanog protein
To further verify the effect of Dppa3 on the degradation of Nanog in ES cells, we established Dppa3 knockdown (Dppa3 KD) ES cell line and verified by RT-PCR (Additional file 1: Fig. S5A). We next examined the expression level of Nanog in Dppa3 KD ES cells. Nanog protein levels were reduced in Dppa3 KD ES cells using western blot and IF analysis (Fig. 5B, C). By contract, Oct4 keep unchanged in Dppa3 knockdown cells (Fig. 5B). Nanog is heterogenetically expressed in ES cells, with a reduced percentage in Dppa3 KD ES cells compared with the level in control ES cells (61.9% vs 66.7%) (Fig. 5D). Though the ability of Nanog synthesis in Dppa3 KD cells keep unchanged (Fig. 5E), Nanog protein was extremely decreased within 1 h in Dppa3 KD ES cells (Fig. 5F). Knockdown of Dppa3 has no effect on the proliferation of ES cells (Additional file 1: Fig. S5B). The clone displays flatter morphology in Dppa3 KD cell line than control group (Fig. 5A). AP staining showed decreased AP-positive clones in Dppa3 knockdown cells (Additional file 1: Fig. S5C). Real-time PCR experiment showed silencing Dppa3 results in a massive downregulation of groundstate markers (Additional file 1: Fig. S5D). In teratoma forming experiment, Dppa3 knockdown cell line can only form smaller teratoma with undeveloped ectoderm, mesoderm, or endoderm (Additional file 1: Fig. S5E and F). These results indicated that silencing of Dppa3 promotes degradation of Nanog protein and impairs pluripotency of ES cells.

Discussion
Dppa3 is a maternal gene required for oogenesis and early embryogenesis, but which has different mechanisms on the regulation of DNA methylation in zygotes and oocytes [3,5,18,22]. Besides of the embryos and oocytes, Dppa3 is also reported to regulate the methylation level in other kind of cells as well [37]. Enforced Dppa3 in NIH3T3 could trigger global DNA demethylation by interacting with Uhrf1 [9]. In some cancer cells Dppa3 could also mediate demethylation by preventing Tet3 function [37]. A recent research showed that Dppa3 promotes tumor oncogenic dedifferentiation through remodeling DNA methylation state [42]. In addition to the effects in embryos and oocytes, Dppa3 is also specifically expressed in embryonic stem cells and plays critical role in the regulation of DNA methylation and pluripotency in ES cells. Dppa3 facilitates germline and endodermal differentiation of human ES cells [40]. Our previous study also showed that Lin28a induced primed ES cells by downregulating Dppa3 [28]. In early embryos, Dppa3 is cleaved into two fragments. The cleaved maternal Dppa3 protein is essential for early embryogenesis by vesicular trafficking [32]. However, it is unclear whether Dppa3 impacts protein stability in ES cells. Here, our data showed Dppa3 interacted with regulatory particle non-ATPase (Rnp8, Rnp11, Rnp1) and regulatory particle triple-ATPase (Rpt2, Rpt6, Rpt5, Rpt3) and almost all α subunits (α1, α2, α4, α5, α6, α7) and β1, β5, and β6 subunits (Additional file 1: Fig. S1D). Ubiquitin is covalently attached to target proteins to signal their degradation by 26S proteasome [38]. UPS provides dominated regulation mechanisms within nucleus. Degradation mediated by lysosome exists in cytoplasm. It is vital for removing misfolded and damaged proteins, and proteins whose functions are no longer needed. UPS is also involved in cell cycle progression, metastasis, DNA damage, and regulatory network in diseases [34,39]. Moreover, UPS is required for activating biological function in early embryonic development, because female pronuclear accumulates thousands of maternal proteins during the oogenesis which are essential for cleavage stage. During early embryonic development stage, numerous proteins undergo different ways to ensure the right path through development period.
Our results also indicate that Dppa3 protects Uhrf1 and Nanog from degradation in ES cells, which may provide further understanding of the role Dppa3 paly in the ES cells. Forced expression of Dppa3 in ES cells stabilized Nanog and Uhrf1, in which Dppa3 might serve  as an ubiquitinated substrate to the inner catalytic chamber within the 20S core particle and is cleaved, which may prevent other substrate to be bind by proteasome and stop the following degradation. Several evidences have shown the crystal structure of Uhrf1 binds with hemi-DNA to recruit Dnmt1 toward the replicable foci [1,2,11]. The Uhrf1 C-terminal RING finger functions as an ubiquitin E3 ligase to cause histone H3 ubiquitination at site of Lys18 and/or Lys23, which is subsequently recognized by Dnmt1 to promote its localization onto replication foci [17]. Degradation of Uhrf1 is mediated by ubiquitin, subsequently Uhrf1 is submitted to 26S proteasome for degradation. The existence of Dppa3 takes over the domain of Uhrf1 that is the same site for its own degradation. Uhrf1 cannot degraded by UPS and results in accumulation within cells. Specifically, we also evaluate the splicer of Dppa3 and found that the accumulation of Uhrf1 is accompanied with full length of Dppa3 protein level in ES cells. Uhrf1 is better characterized in molecular mechanism in DNA methylation maintenance [35,45].

Conclusion
Together, we provide new molecular insight into the function of Dppa3 in ES cells. We identified new functions of Dppa3 in protein stability regulation, Nanog and Uhrf1 regulation in ES cells. Our findings advance our knowledge of the functions and mechanisms of Dppa3 in ES cells, which may contribute to the future application of stem cells in regenerative medicine.