SCIENCE CHINA Life Sciences • RESEARCH PAPERS •
September 2010 Vol.53 No.9: 1073–1084 doi: 10.1007/s11427-010-4055-8
Metabolic properties of chicken embryonic stem cells LI Jia1,2, ZHANG BaoLu1,2, HAN HongBing1,2, CAO ZhiCheng1,2, LIAN ZhengXing1,2* & LI Ning1 1
State Key Laboratory of Agrobiotechnology, China Agricultural University, Beijing 100194, China; 2 College of Animal Science & Technology, China Agricultural University, Beijing 100194, China Received March 2, 2010; accepted April 19, 2010
Cellular energy metabolism correlates with cell fate, but the metabolic properties of chicken embryonic stem (chES) cells are poorly understood. Using a previously established chES cell model and electron microscopy (EM), we found that undifferentiated chES cells stored glycogen. Additionally, undifferentiated chES cells expressed lower levels of glucose transporter 1 (GLUT1) and phosphofructokinase (PFK) mRNAs but higher levels of hexokinase 1 (HK1) and glycogen synthase (GYS) mRNAs compared with control primary chicken embryonic fibroblast (CEF) cells, suggesting that chES cells direct glucose flux towards the glycogenic pathway. Moreover, we demonstrated that undifferentiated chES cells block gluconeogenic outflow and impede the accumulation of glucose-6-phosphate (G6P) from this pathway, as evidenced by the barely detectable levels of pyruvate carboxylase (PCX) and mitochondrial phosphoenolpyruvate carboxykinase (PCK2) mRNAs. Additionally, cell death occurred in undifferentiated chES cells as shown by Hoechst 33342 and propidium iodide (PI) double staining, but it could be rescued by exogenous G6P. However, we found that differentiated chES cells decreased the glycogen reserve through the use of PAS staining. Moreover, differentiated chES cells expressed higher levels of GLUT1, HK1 and PFK mRNAs, while the level of GYS mRNA remained similar in control CEF cells. These data indicate that undifferentiated chES cells continue to synthesize glycogen from glucose at the expense of G6P, while differentiated chES cells have a decreased glycogen reserve, which suggests that the amount of glycogen is indicative of the chES cell state. chicken embryonic stem cell, energy metabolism, glycogen, cell fate, glucose-6-phosphate
Citation:
Li J, Zhang B L, Han H B, et al. Metabolic properties of chicken embryonic stem cells. Sci China Life Sci, 2010, 53: 1073–1084, doi: 10.1007/s11427- 010-4055-8
Chicken embryonic stem (chES) cells, derived from the area pellucida of the stage X chicken blastoderm, are characterized by their capacity for prolonged undifferentiated proliferation in culture while retaining the potential to differentiate into all three germ layer cells both in vivo and in vitro [1–3]. Unlike the blastocyst stage mouse embryo, the stage X chicken blastoderm contains 40000–60000 pluripotent precursor cells [1,4,5], which arise from the cleavages of a single egg over 18–20 h. Cell division occurs at approximately 45-min intervals [6]. Because of the extremely short cell cycle and closed oviduct development, these precursor *Corresponding author (email:
[email protected]) © Science China Press and Springer-Verlag Berlin Heidelberg 2010
cells are probably “imprinted” with certain properties of the early generation. For example, chicken primordial germ cells (PGCs), which are derived from the stage X chick embryo, have high glycogen content [7,8]. Several studies have shown that the chES cell cytoplasm is heavily laden with yolk granules [9,10]. Glycogen synthesis and degradation, as well as yolk deposition, are metabolic processes that accompany the development of the chick embryo. Glycogenesis occurs as early as late oogenesis, when a large number of compact clusters of glycogen granules are synthesized by the oocyte [11], and this glycogen is intensively consumed during ovulation [12,13]. After ovulation, a second wave of glycolife.scichina.com
www.springerlink.com
1074
Li Jia, et al.
Sci China Life Sci
gen accumulation occurs and encompasses a period of extensive cell cleavage (stages (EGK) I–(EGK) VI) [4]. This intracellular glycogen is consumed in the entire second half of uterine development, as the chick embryo reaches the blastoderm stage (stages (EGK) VI–(EGK) X) [11,14]. During the final stage of growth, chicken oocytes take up the lipoprotein precursor, vitellogenin (VTG) and very low-density lipoproteins (VLDL) by receptor-mediated endocytosis [15–17], and deliver them to yolk spheres, the storage organelles in the cytoplasm of oocytes. Lipovitellin and phosvitin, which are the breakdown products of VTG, are targeted to and incorporate with other components to form the electron-dense yolk granules [18]. VLDL is proteolytically processed and deposited into the electron-lucent phase of yolk spheres [17,19,20]. Following fertilization, the fertilized egg initiates the first cleavage in the uterus [4]. Because of the large yolk mass, cell division is restricted to the protoplasmic portion of the egg where the zygote nucleus is located. At the cleavage stage (EGK) II, yolk granules extensively penetrate from the yolk sphere into the dividing cells and are distributed to the daughter cells during the following cleavage [21]. The majority of yolk granules at this period are reserved as an energy store, while small amounts of the granules are used for glycogen synthesis [11]. Numerous links have been established between energy metabolism and cell fate decisions. For instance, the rates of oxygen consumption and ATP generation increase during mitosis in several transformed and non-transformed cell systems [22–24], and therefore the depletion of ATP or any of its interconvertible forms causes the cell to arrest in G1 in the absence of DNA damage [25]. Moreover, the glycogen level in tumor tissue from a colon cancer patient was reported to be much higher than the normal tissue, and the glycogen level in the tumor was highest during the G1 phase and progressively decreased in the S phase, suggesting that glycogen metabolism is important for cancer cells to progress to the S phase [26]. A more recent study showed that the inability of TRAF6-deficient CD8+ T cells to switch from glycolytic metabolism to fatty acid metabolism impaired their ability to survive [27]. Furthermore, hypoxia inducible factor (HIF) 1a, which facilitates the transduction pathways that promote self-renewal [28] and inhibits the pathways that promote the differentiation or apoptosis [29,30] of neuronal stem cells (NSC) under low oxygen (O2) tension, could also induce transactivation of glycolysis-related genes, such as Glut1 [31], Hks [32] and Pfk [33]. Deprivation of TCRβ-expressing thymocytes or T cells of IL-7 or Notch leads to a loss of Glut1, thus preventing thymocytes or T cell glucose uptake and metabolism [34,35]. Based on the above-mentioned studies, cellular energy metabolism is required for a number of important cell processes, including cell growth, proliferation and differentiation. Importantly, the cell modulates its cellular energy metabolism through proteins that also regulate cell fate deci-
September (2010) Vol.53 No.9
sions. With an understanding of this connection between metabolism and critical cellular processes, cells can be manipulated by modulating their energy metabolism. Despite several studies showing that the chick embryo metabolizes energy in a special way during cleavage, nothing is known about the metabolic properties of chES cells, impeding our broader understanding of this pluripotent stem cell type in oviparous species. Chicken blastodermal cells that are retrieved from the area pellucida of the stage X (EG & K) embryo can be cultured on a two-dimensional STO feeder layer in the conditioned medium for more than 20 passages [9,10]. However, the derivation of a chES cell line using this method is labor-intensive and time-consuming because the pluripotency is hard to establish, even after several passages [2,6,9,36,37]. It is an unstable experimental system that is difficult to use for the following reasons: chES cells in this culture differentiate easily even under optimized conditions; the experimental data reflect the response of not only chES cells, but also feeder cells, to various stimuli. Previously, we have reported the derivation of chES cells using a chicken blastocyst-like structure (BLS) method [38]. In chicken BLS culture, a group of precursor cells develops as a unit in a sphere and is suspended in space, which largely mimics oviduct growth. Taking advantage of spheroid culture, these cells obtain and maintain chES-like properties for several days without the support of feeder cells, making them a great model for studying chES cells in vitro. Here, using chicken BLSs as experimental materials, we have investigated the metabolic properties of chES cells. We found that chES cells metabolize energy in a special way. We also found solutions to improve chES cell viability in vitro.
1 1.1
Materials and methods Chicken BLS culture
Freshly laid eggs from White Leghorn hens were collected and verified in accordance with the method described by Eyal-Giladi and Kochav [4]. The stage X blastoderm was separated from the egg yolk, and the area pellucida was removed by gentle aspiration with a micropipette into PBS (Invitrogen, Carlsbad, CA, USA) at room temperature (RT). The area pellucida was broken into small pellets using a Pasteur pipette by slowly pipetting up and down five times. The pellets were transferred to a 35 mm low-attachment dish (Fisher) and incubated in the chES cell medium or differentiation medium at 37°C with 5% CO2 and 100% relative humidity. After 7 h of incubation, chicken BLSs formed. Next, around 20–30 chicken BLSs were carefully collected and transferred to a new 35 mm low-attachment dish and placed back into the incubator. Half of the medium was replaced every day. The chES cell medium consisted of Dulbecco’s Modified Eagle Medium (DMEM-H, Invitrogen)
Li Jia, et al.
Sci China Life Sci
with 20% fetal bovine serum (FBS, Invitrogen), 0.1 mmol L−1 β-mercaptoethanol (Sigma, St. Louis, MO, USA), 2 mmol L−1 L-glutamine (Invitrogen), 1×non-essential amino acids (Invitrogen), 1×ES cell nucleotides (Millipore), 1 mmol L−1 pyruvate (Invitrogen), and 2×106 U recombinant human LIF (hLIF, ESGRO). The differentiation medium included all of the components in the chES cell medium except for hLIF.
September (2010) Vol.53 No.9
1075
Total RNA was extracted from chicken BLS cells, stage X chick embryos, primary chicken embryo fibroblast (CEF) cells and adult hen liver using a total RNA isolation kit (Ambion) in accordance with the manufacturer’s instructions. The amount of total RNA was measured with a spectrophotometer at 260 nm. RNA (2 μg) was reverse-transcribed using a Superscript First-Strand Synthesis System (Invitrogen). The cDNAs were used as templates for PCR amplification with gene-specific primers (Table 1). Quantitative differences in gene expression were determined by qPCR using the Platinum® SYBR® Green qPCR SuperMix-UDG system (Invitrogen) and a thermal cycler controlled by the MyiQ Real Time Detection System software (BioRad, Munich, Germany). All experiments were performed in triplicate.
cold 4% paraformaldehyde for 15 min and then permeabilized in PBS with 0.1% Triton X-100 for 10 min. The chicken BLSs were washed three times in PBS for 5 min each time. For alkaline phosphatase (ALP) staining, the chicken BLSs were incubated in alkaline phosphatase staining solution (100 mmol L−1 NaCl, 100 mmol L−1 Tris-HCl pH 9.5, 5 mmol L−1 MgCl2, 1 mg mL−1 NBT, 0.1 mg mL−1 BCIP, Bio-Rad) for 30 min at 37°C in the dark. For periodic acid Schiff (PAS) staining, the chicken BLSs were incubated in periodic acid solution (Sigma) for 5 min and then in Schiff’s solution (Sigma) for 30 min at RT in the dark. For SSEA 1 immunofluorescence staining, the chicken BLSs were blocked in 1% BSA (Sigma) in PBS for 30 min at RT and then incubated in a FITC-conjugated anti-chicken SSEA 1 monoclonal IgG (1:100, Sigma) overnight at 4°C in the dark. After several washes in PBS, chicken BLSs were incubated in Hoechst 33342 (10 μg μL−1, Sigma) solution for 10 min at RT. For observation and photography, the chicken BLSs samples were placed onto microscope slides with cover slips on top. The sides of the cover slip were sealed with nail polish. ALP-stained or PAS-stained chicken BLSs samples were observed under a Leica inverted microscope. SSEA1-FITC fluorescence was excited at 488 nm and observed under a Leica CTR MIC microscope. Images were obtained using a COHU High Performance CCD camera.
1.3
1.4
1.2 RNA extraction and quantitative real-time RT-PCR
Cytochemistry
The chicken BLSs were collected in PBS. After being washed twice in fresh PBS, the chicken BLSs were fixed in Table 1
Oligonucleotide primer sequences used for qPCR
Gene name Glut1 Gys Hk1 Pcx Pck2 Pfk Tert Bmp5 T Vil1 Cxcl12 Isl1 Gapdh
Sequence 5′-GCAATGAGGAGAACAAAG-3′ 5′-ACAATAAACCAAGGGATG-3′ 5′-GGTGAAGGAGAAGTTTGG-3′ 5′-CATAGTAGGAGGGGAAGAC-3′ 5′-CTTCACCTTCTCCTTCCC-3′ 5′-ACCTCCTTGACGATGATG-3′ 5′-CGCATCCGTGGTGTAAAG-3′ 5′-AGCAGGTTGGGCAGGTAG-3′ 5′-GGATGAGCCCACGGAG-3′ 5′-CCCAGCCAGGAGGTGTAG-3′ 5′-GGTGCTGATGCTGCCTAC-3′ 5′-ATCCTCTTTTGCCTGACG-3′ 5′-GAGCGAAGTCATCACAAG-3′ 5′-GCACATAGTCAAGCAAGC-3′ 5′-ACCTCACAGACCCAGACC-3′ 5′-TTCCTTGTAGTGCCTTCG-3′ 5′-TCGCAGTTAGGTAGTTGG-3′ 5′-GCAGGCAGAGGAGTTATC-3′ 5′-CTCAACAGTGCCCTCAAG-3′ 5′-CCATCATCCACCATCTTC-3′ 5′-CAGCCTGACTTACCGATG-3′ 5′-CCCCAGCACAGATACTTTAG-3′ 5′-ATATTCTGAGGGTTTCTCCG-3′ 5′-ACTCGATGTGGTACACCT-TG-3′ 5′-AGAACATCATCCCAGCGTCC-3′ 5′-CAGCAGCAGCCTTCACTACC-3′
GenBank No. NM_205209 AB 090806 AB 083369 NM 204346 XM 001236575 AB 061205 NM_001031007 NM 205148 U 67086 J 03781 NM 204510 NM_205414 NM_204305
Transmission electron microscopy
Chicken BLSs were fixed overnight at 4°C in Karnovsky’s fixative, consisting of 2% paraformaldehyde and 2% glutaraldehyde in 0.05 mol L−1 sodium cacodylate buffer (pH 7.2), and post-fixed in 1% osmium tetroxide for 2 h at 4°C. The samples were then rinsed with cacodylate buffer, dehydrated in a graded series of ethanol, and embedded in epoxy resin. After polymerization for 24 h at 70°C, the trimmed blocks were cut into ultra-thin sections using an ultramicrotome (MT-X, RMC, Tucson, AZ). Ultrathin sections of the specimens were stained with uranyl acetate and lead citrate and then observed under a transmission electron microscope (JEM4000FX, JEOL, Osaka, Japan). 1.5
Quantitative cell death detection assay
To quantify chES cell mortality, cell death was detected by Hoechst 33342 and propidium iodide [29] double staining. Chicken BLSs were rinsed twice with PBS, incubated in PI solution (10 μg μL−1, Sigma) for 20 min at 37°C in the dark, and stained with Hoechst 33342 (10 μg μL−1, Sigma) for 10 min at 4°C in the dark. Hoechst 33342 and PI fluorescence were excited at 352 nm and 536 nm, respectively. All images were obtained using a Leica CTR MIC microscope and a COHU High Performance CCD camera. Hoechst 33342 stained nuclei (blue) and PI stained nuclei (red) were then
1076
Li Jia, et al.
Sci China Life Sci
counted by Image-Pro Plus 5.0 (Media Cybernetics, Inc.). The mortality rate was calculated by dividing the number of red nuclei by the total number of nuclei examined. 1.6
Data analysis
Data were analyzed by one way analysis of variance, followed by a Dunnett’s test using SAS V8 software (SAS Institute, Cary, NC, USA).
2
Results
2.1 Confirmation of the undifferentiated state of chicken embryonic stem cells (chES) in chicken blastocyst-like structures (BLSs) Chicken BLS cells have been characterized as chES cells in our previous study [38]. To confirm that chES cells in chicken BLSs were maintained in an undifferentiated state after long-term culture, we performed three experiments on chicken BLSs cultivated in chES cell medium for seven days. We first detected the cell surface expression of SSEA 1 using immunofluorescence staining. Chicken BLS cells stained positively for SSEA 1 (Figure 1A(a)), as compared with a control where no antibody was added (Figure 1A(d)). We then analyzed the alkaline phosphatase (ALP) activity using an ALP staining assay. Chicken BLS cells had a high level of ALP activity (Figure 1B(a)), relative to a reaction control (Figure 1B(b)). We finally measured the telomerase activity by examining the expression of the chicken telomerase reverse transcriptase (Tert) gene through qPCR. The expression of Tert was significantly higher in chicken BLS cells compared with the stage X chick embryo (XE) and
September (2010) Vol.53 No.9
telomerase-free CEFs (P<0.05, Figures 1C and D), which had very low basal expression. The high mRNA levels of TERT indicated high telomerase activity in chicken BLS cells. After long-term culture in vitro, chES cells were maintained in an undifferentiated state in chicken BLSs. 2.2 ChES cells initiate the differentiation program in suspended growth period without the support of LIF ChES cells are capable of differentiation in response to retinoic acid and form embryoid bodies (EBs) [9,36]. Previous studies have shown that chicken BLS cells initiate spontaneous differentiation in the absence of LIF in vitro [38]. In this case, the chicken BLSs were cultured in a tissue culture dish with differentiation media. Under these conditions, the chicken BLSs were suspended in culture for a short time and then attached to the tissue culture dish within 48 h. After more than one week of differentiation, the chicken BLS cells progressively differentiated into neural cells [38], endothelial cells and muscle cells (unpublished data) as shown by immunofluorescence staining. To trace the differentiation processes during the suspended growth period, we first checked the levels of ALP in chicken BLSs maintained without LIF. Chicken BLS cells had ALP activity during the first 12 h after LIF withdrawal (Figure 2A). However, most cells in the chicken BLS rapidly lost their ALP activity by 24 h (Figure 2B), and this loss was most noticeable at 36 h (Figure 2C). By contrast, the control chicken BLS cells, which were cultured in a low-attachment dish and in the presence of LIF, had a high level of ALP activity during the entire time period (Figures 2D–F). Loss of ALP activity indicates that chicken BLS cells initiate a distinct differentiation program.
Figure 1 Confirmation of chES cells in an undifferentiated state after long-term culture. A, Chicken BLSs cultured in the chES cell medium for seven days are stained with an antibody against SSEA 1 (green; a); reaction control (d); nuclei of cells in a and d are counterstained with Hoechst 33342 (blue; b and e); Merged images (c and f). B, Chicken BLSs cultured in the chES cell medium are stained for ALP activity (a); reaction control (b). C and D, The expression of the chicken Tert gene was detected in chicken BLSs maintained in the chES cell medium, XE and CEFs using qPCR. Each bar represents the average expression in three repeats. An asterisk (*) indicates a significant difference (P<0.05). Scale bar=200 μm.
Li Jia, et al.
Sci China Life Sci
Figure 2 Alkaline phosphatase (ALP) activity during chES cell differentiation. A–C, ALP staining of chicken BLSs at 12, 24 and 36 h after LIF withdrawal. A, Expression of ALP in all cells in a chicken BLS at 12 h. B and C, Loss of ALP activity by the vast majority of cells in chicken BLSs at 24 h (B) and 36 h (C). Arrows indicate purple stained, ALP-positive cells in B and C. D–F, Expression of ALP activity in chicken BLSs grown in the presence of LIF at 12 (D), 24 (E) and 36 h (F). Scale bar=200 μm.
Studying the onset of marker gene expression allowed us to determine the time course of differentiation. We detected both the expression pattern of chicken Tert as well as early mesodermal, ectodermal and endodermal marker genes in chicken BLS cells during a 36-h period after LIF removal using qPCR (Figure 3). The stage X chick embryo was used as a control to mark the start of differentiation. We found that the Tert mRNA level was downregulated significantly by 24 h after LIF withdrawal (P<0.05). The early endodermal marker, villin 1 (Vil1), was expressed at 12 h and dramatically upregulated at 24 h (P<0.05). A substantial amount of brachyury (T) mRNA, which encodes a marker for the primary mesoderm, was detected by 24 h. Except for an early mesodermal marker, genes specific to mesoderm-derived cell lineages were also detected during this period. LIM/homeodomain (Isl1) transcripts, which encode a marker for cardiac mesoderm and definitive endoderm derivatives, were already detected during the first 12 h and were upregulated significantly by 24 h (P<0.05). The expression of bone morphogenetic protein (Bmp) 5, which encodes a marker for the dorsal mesoderm-derived cell lineage, was not detected until 12 h and was then dramatically upregulated by 24 h (P<0.05). The ectodermal marker, chemokine (C-X-C motif) ligand (Cxcl) 12 transcript was expressed at the 12-h time point and was significantly upregulated by 24 h (P<0.05). The induction of these markers was not observed in chicken BLS cells maintained in the presence of LIF (data not shown). Taken together, these data indicate that the expression of genes from all three germ layers are induced in chicken BLS cells during the 12–24 h after LIF withdrawal. 2.3 Undifferentiated chES cells have a large stockpile of yolk protein and glycogen stores Previous reports have shown that the cytoplasm of cleavage
September (2010) Vol.53 No.9
1077
stage embryonic cells is rich in yolk granules and glycogen reserves but that the reserves are utilized completely as cell shedding progresses [11,14]. To check if chES cells have the same properties, chicken BLS cells in undifferentiated chES cell culture were analyzed ultrastructurally over time using EM and were compared with chicken BLS cells that had undergone differentiation. In general, the undifferentiated chicken BLS cells contained large nuclei (arrowheads in Figures 4A–C), and their cytoplasm was filled with energy storage organelles. One storage organelle was a yolk granule with a round or oval shape and a size range of 0.2–2.3 μm in diameter (arrows in Figures 4A–D). We found that not only did the number of yolk granules decrease, but the content of yolk granules also changed over time. First, the granules were electron-dense and proteinaceous (arrow in Figure 4A); then we observed mosaic structures with both protein residues and an exposed lipid phase (arrow in Figure 4B); and finally, we found electron-lucent lipid-rich droplets (arrow in Figure 4C). Interestingly, a number of proteinaceous or lipid-rich yolk granules were observed either close to or directly associated with mitochondria (arrows in Figure 4D), indicating an energy function of the yolk proteins. We also found the same phenomenon in chicken BLS cells that were undergoing differentiation (figures not shown), indicating that yolk protein utilization does not depend on the chES cell state. Furthermore, the cytoplasm of chicken BLS cells in undifferentiated chES cell cultures had abundant glycogen reserves (arrowhead in Figure 4E), which are seen as clusters of open spaces associated with various numbers of glycogen granules (arrow in Figure 4E). By contrast, we did not find many glycogen reserves in the cytoplasm of differentiating chicken BLS cells. To further confirm that glycogen abundance depends on the state of chES cells, we performed PAS staining, which is primarily used to identify glycogen in tissues, in chicken BLSs cultured either in the presence or absence of LIF over time. Indeed, all of the cells in the chicken BLSs that were grown in the presence of LIF consistently stained positive (Figures 5A–C), whereas the number of PAS positive cells in the chicken BLSs declined by 24 h after LIF withdrawal (Figures 5E and F). This suggests that chES cells store glycogen in their undifferentiated state; however, they decrease their glycogen reserves upon differentiation. 2.4 Metabolic flux occurs largely in the glycogenic direction in undifferentiated chES cells The storage of glycogen indicates a specific energy metabolism of undifferentiated chES cells. Therefore, we studied the gene expression levels of enzymes in various metabolic pathways to investigate the metabolic properties of chES cells. To detect glycolytic and glycogenic metabolism, we measured the mRNA levels of glucose transporter protein type 1 (GLUT1), which facilitates the transport of
1078
Li Jia, et al.
Sci China Life Sci
September (2010) Vol.53 No.9
Figure 3 Expression of differentiation markers during chES cell differentiation. To determine the time course of differentiation, the expression pattern of chicken Tert, as well as five germ layer markers, including villin1 (Vil1) (early endoderm), brachyury (T) (primary mesoderm), LIM/homeodomain 1 (Isl1) (cardiac mesoderm and definite endoderm lineage), bone morphogenetic protein 5 (Bmp5) (dorsal mesoderm lineage) and chemokine (C-X-C motif) ligand 12 (Cxcl12) (ectoderm), were detected in chicken BLSs at 7, 12, 24 and 36 h after LIF withdrawal using qPCR. The XE represents the start of differentiation. All of the gene expression data were normalized to the expression of Gapdh. Each bar represents the average expression of three repeats. An asterisk (*) indicates a significant difference (P<0.05).
glucose across the plasma membrane of cells: hexokinase 1 (HK1), which is a rate-limiting enzyme that phosphorylates glucose into glucose-6-phosphate (G6P) in the first step of glycolysis, glycogen synthase (GYS), which catalyzes the transfer of the glucosyl moiety of UDP-glucose to the glycogen polymer and controls the rate of glycogen synthesis, and phosphofructokinase (PFK), which converts fructose 6-phosphate into fructose 1,6-bisphosphate and controls the committed step of glycolysis. Quantitative real-time RT-PCR was used to measure the expression levels of these genes in
the XE, ALP positive chicken BLS cells (undifferentiated chES cells), ALP negative chicken BLS cells (differentiated chES cells) and control primary CEFs (Figure 6). Both undifferentiated chES cells and the XE had significantly less abundant GLUT1 mRNA compared with the control (P<0.05). By contrast, the GLUT1 transcript was much more abundant in differentiated chES cells as compared with the control. This result indicates that glucose use is low in undifferentiated chES cells and in the XE, but it is high in differentiated chES cells. Once glucose enters the cell, it is
Li Jia, et al.
Sci China Life Sci
Figure 4 Ultrastructural analysis of undifferentiated chES cells. Chicken BLS cells cultured in the chES cell medium were analyzed ultrastructurally over a time course. A–C, Yolk granules (arrows) in the cytoplasm of a chES cell at different time points. A, The yolk granule is proteinaceous (arrow) at 7 h. B, The yolk granule has a mosaic structure with an exposed lipid phase due to the partial degradation of yolk proteins at 24 h (arrow). C, The yolk granule becomes an electron-lucent lipid droplet due to the complete degradation of yolk proteins at 36 h (arrow). Arrowheads in A–C indicate the large nuclei in chES cells. D, Yolk granules (arrows) are either close to or directly associated with the mitochondria (arrowheads) in a chES cell. E, A glycogen reserve (arrowhead) appears as an open space that is composed of a cluster of glycogen granules (arrow) in the cytoplasm of a chES cell. Scale bar, 2 μm.
Figure 5 PAS staining of chES cells. A–C, PAS staining of chicken BLSs after LIF withdrawal over a time course. A, All cells in a chicken BLS stained positively (red) at 12 h. B and C, Progressive loss of PAS positive stain by cells in chicken BLSs at 24 (B) and 36 h (C). Asterisks (*) indicate the PAS positive staining area in B and C. D–F, PAS staining of chicken BLSs grown in the presence of LIF at 12 (A), 24 (B) and 36 h (C). n equals the number of chicken BLSs studied. Scale bar, 200 μm.
phosphorylated and then flows into different metabolic shunts. Both undifferentiated chES cells and the XE express significantly lower levels of PFK transcripts as compared with the control (P<0.05), whereas differentiated chES cells had moderately more abundant PFK transcripts as compared with the control. This result indicates a lower glycolysis rate in both undifferentiated chES cells and the XE as well as a slightly higher glycolysis rate in differentiated chES cells as compared with primary CEFs. Unexpectedly, the HK1 message was significantly more abundant in the undifferentiated chES cells than in the control and differentiated chES
September (2010) Vol.53 No.9
1079
Figure 6 Expression of glycolytic and glycogenic genes in chES cells. To detect both glycolytic and glycogenic metabolism, the expression of several glycolytic genes, Glut 1, Pfk and Hk, and the glycogenic gene, Gys, was detected in a stage X chick embryo (XE), ALP positive chicken BLSs (U), ALP negative chicken BLSs (D) and control CEFs using qPCR. Quantitative gene expression data were normalized to the expression level of Gapdh. Each bar represents the average expression of three repeats. An asterisk (*) indicates a significant difference (P<0.05).
cells (P<0.05), suggesting that undifferentiated chES cells convert more glucose into G6P. However, increased G6P did not boost the transcriptional level of inducible glycolytic genes, indicating that G6P did not accumulate in the undifferentiated chES cells. Consistently, undifferentiated chES cells significantly upregulated GYS gene expression as compared with the control (P<0.05), implying that G6P is converted into glycogen immediately after being generated. By contrast, we detected only a moderate increase in the HK1 mRNA level and no change in the GYS mRNA level in differentiated chES cells as compared with the control. Additionally, a moderate increase in the level of HK1 mRNA and a significant increase in the level of GYS mRNA was detected in the XE as compared with the control (P<0.05), indicating that metabolic trends are similar between the XE and undifferentiated chES cells. Taken together, we conclude that the glucose flux is directed mainly toward the glycogenic pathway in undifferentiated chES cells; however, glucose flux is redirected to the glycolysis pathway in differentiated chES cells. To detect gluconeogenic metabolism, we measured the mRNA levels of two enzymes that have catalytic and rate-controlling roles in the gluconeogenesis pathway: pyruvate carboxylase (PCX), which catalyzes the carboxylation of pyruvate to form oxaloacetate, and mitochondrial phosphoenolpyruvate carboxykinase (PCK2), which converts oxaloacetate into phosphoenolpyruvate. The mRNA levels were compared among the XE, ALP positive chicken
1080
Li Jia, et al.
Sci China Life Sci
September (2010) Vol.53 No.9
Figure 7 Expression of gluconeogenic genes in chES cells. To detect gluconeogenic metabolism, the expression of two gluconeogenic genes, Pck2 and Pcx, was detected in a stage X chick embryo (XE), alkaline phosphatase (ALP) positive chicken BLSs (U), ALP negative chicken BLSs (D) and an adult hen liver control (L) using qPCR. Quantitative gene expression data were normalized to the expression level of Gapdh. Each bar represents the average expression of three repeats. An asterisk (*) indicates a significant difference (P<0.05).
BLS cells, ALP negative chicken BLS cells and the adult hen liver control (starved overnight, Figure 7). Unexpectedly, there were no detectable mRNA levels of PCK2 and PCX in undifferentiated chES cells and the XE, whereas lower mRNA levels of these two enzymes were detected in the differentiated chES cells as compared with the adult hen liver control (P<0.05). This result indicates that undifferentiated chES cells block gluconeogenesis. 2.5 Exogenous G6P rescues severe cell death in undifferentiated chES cells According to various chES cell culture conditions, glucose is provided as the primary energy source, and pyruvate serves as a backup for energy crisis [1,2,9,36]. However, the experiment performed thus far indicated that glucose was not a priority energy substrate for chES cells in their undifferentiated state. We quantitatively detected cell mortality over time in chicken BLSs maintained in either chES cell medium (glucose based medium, UC in Figure 8A) or differentiation medium (glucose based medium, DC in Figure 8B) using Hoechst 33342 and PI double staining. We calculated the mortality rate by dividing the number of dead cells (PI-stained) by the total number of cells (Hoechst 33342-stained) examined. The UC group had an average mortality rate of 3%±1% at 7 h, which increased further to 57%±5% at 36 h. By contrast, the DC group had an average mortality rate of 2%±1% at 7 h, which only moderately increased to 15%±1% over the same period. This result shows that undifferentiated chES cells suffered under the current glucose culture conditions. To verify whether exogenous energy, or energy substrates, could rescue undifferentiated chES cell death,
Figure 8 Detection of chicken BLS cell mortality in the glucose based culture medium. Chicken BLSs cultured in either the chES cell medium (UC) or the differentiation medium (DC) were stained with Hoechst 33342 and PI over a time course. A, Merged pictures show chicken BLS cell death in UC at 7, 12, 24 and 36 h. An arrowhead indicates a live cell stained with Hoechst 33342 (blue). An arrow indicates a dead cell stained with Hoechst 33342 and PI (pink). B, Merged pictures show chicken BLS cell death in DC at 7, 12, 24 and 36 h. C, Quantitative analysis of chicken BLS cell mortality in UC and DC. Hoechst 33342-stained nuclei and PI-stained nuclei were counted, and the mortality rate was calculated by dividing the number of PI-stained nuclei by the total number of nuclei examined. n represents the total number of chicken BLSs counted in UC or DC. The figure shows the mean±SEM of three different experiments. An asterisk (*) indicates a significant difference (P<0.01). Scale bar, 200 μm.
chicken BLSs in undifferentiated chES cell cultures were incubated with or without ATP, G6P, glucose or pyruvate. The average mortality rate of chicken BLSs for each treatment was calculated at 7, 12, 24 and 36 h. Before testing for ATP, G6P, glucose or pyruvate in the experimental groups, standard curves were generated to determine the ideal concentration of each energy substrate (data not shown). As shown in Figure 9A, the average mortality rate of chicken BLSs without treatment (U in Figure 9A) increased from 2.5%±1% to 55%±3% over 36 h. However, the average mortality rate of these cells decreased by 77% (P<0.05) at 36 h after the administration of 5 mmol L−1 G6P. ALP staining indicated that all cells in chicken BLSs had high levels of ALP activity at 36 h of G6P treatment (data not shown). Treatment with glucose (as high as 35 mmol L−1), pyruvate (5 mmol L−1) or ATP (2 mmol L−1) had no effect over 36 h. We performed the same experiment in chicken BLSs that had undergone differentiation (D in Figure 9B). The average mortality rate in differentiating cells without treatment increased from 3%±1% to 14.5%±2% over 36 h.
Li Jia, et al.
Sci China Life Sci
Figure 9 Quantitative analysis of cell mortality in chicken BLSs after treatment with energy or different energy substrates. Chicken BLSs cultured in either the chES cell medium (U) or the differentiation medium (D) were treated with energy (ATP) or various energy substrates, including G6P, glucose and pyruvate for 36 h. Cell death was detected by Hoechst 33342 and PI double staining, and the number of Hoechst 33342 and PI-stained nuclei were counted. The mortality rate was calculated by dividing the number of PI-stained nuclei by the number of Hoechst 33342-stained nuclei. A, The effect of various supplements on the mortality rate of undifferentiated chES cells. B, The effect of various supplements on the mortality rate of differentiating chES cells. Each bar represents the mean±SEM of three different experiments.
However, none of the exogenous energy or energy substrates statistically decreased the average mortality rate in these cells during this time period (Figure 9B). Taken together, the data indicate that the endogenous G6P level in undifferentiated chES cells could not satisfy their growth in vitro.
3
Discussion
In our present study on the metabolic properties of chicken embryonic stem (chES) cells, we demonstrated that undifferentiated chES cells maintained a high glycogen level by synthesizing glycogen extensively from the glucose that enters the cells, which withdraws a large amount of glucose-6-phosphate (G6P) from the glucose flux. Different from the uterine chick embryo, the yolk proteins serve as the source of carbon for pyruvate, rather than the sugar phosphate, and glycogen in undifferentiated chES cells. In this scenario, the G6P level cannot be maintained, which causes severe chES cell death. By contrast, chES cells ter-
September (2010) Vol.53 No.9
1081
minate glycogen synthesis and decrease the glycogen reserves once they commence spontaneous differentiation, which causes a metabolic switch from glycogenic metabolism to glycolytic metabolism. Taken together, our results suggest that a chES cell is a glycogenic cell, and the cellular glycogen level is an indicator of chES cell state. Our findings also indicate that, except for keeping them in an undifferentiated state, chES cells should be maintained in an extracellular milieu that contains an appropriate amount of G6P. Development of the uterine chick embryo accompanies a wave of glycogen synthesis that is based on yolk degradation, and glycogen breakdown provides energy for complex cell activities, such as cleavage, cell compaction and migration [4,11,14]. In this study, we found that once glucose enters undifferentiated chES cells, it is directly converted to glycogen rather than used as an energy source, and this glycogen is not degraded unless chES cells have differentiated. This allows large stores of glycogen to be continuously present in the undifferentiated chES cells during rapid proliferation. It is most likely that glycogen itself participates in cell division. In fact, a number of studies have reported glycogen in association with the nuclear envelope, the endoplasmic reticulum, and the annulate lamellae of embryonic and transformed cells [14,39–41]. Glycogen has also been reported inside the nucleus [42–46]. Moreover, glycogen consistently appears in the Xenopus embryo during the cleavage stage [47]. In addition, glycogen has been shown to promote nuclear formation, where its role is both structural and catalytic [48]. Thus, it is attractive to propose that undifferentiated chES cells benefit from synthesizing glycogen, which might facilitate cell cleavage during their extensive proliferation, although this has not been tested yet. Consistent with this hypothesis, we found that glycogen reservoirs began to diminish as soon as differentiation markers began to be expressed in the chES cells (between 12 and 24 h), and we found that glucose flux was redirected to glycolysis in differentiated chES cells. Cell proliferation and differentiation are traditionally perceived as reciprocal processes, with cell cycle withdrawal required for terminal differentiation [49–52]. It is most likely that differentiating cells decrease the amount of glycogen present in the cells in response to the declining proliferation rate. Furthermore, this response may be under the control of signals that not only modulate cellular proliferation and differentiation but also regulate cellular energy metabolism. In fact, one of the serine/threonine protein kinases, glycogen synthase kinase 3 (GSK3), and its related phosphoinositide 3-kinase (PI3K)/ Akt/GSK3 signaling pathway may be involved in this regulation. As a downstream target of the PI3K/Akt pathway, GSK3, which is inhibited upon phosphorylation by Akt, actively regulates cell proliferation and differentiation [53–55]. Conversely, GSK3 modulates glycogen metabolism through inactivation of its rate-controlling enzyme, glycogen synthase (GYS) [56]. While our study could not
1082
Li Jia, et al.
Sci China Life Sci
fully explain the role of glycogen in undifferentiated chES cells, further experiments are required to pinpoint where and when these glycogens are translocated during cell proliferation and differentiation to draw a rough picture of their cellular function. However, we can conclude from our current findings that the glycogen level reflects the chES cell state. A high glycogen level indicates an undifferentiated state of chES cells, and vice versa. It has been reported that the number of cytosolic yolk granules is decreased in chES cells during culture [9,10]. In our present study, we found that except for the mitotic loss of yolk granules, chES cells utilized their protein component. This conclusion is based on the observation that a majority of the yolk granules transformed from proteinaceous (electron-dense) particles into lipid-rich (electron-lucent) droplets over time. Moreover, amino acids from the yolk protein breakdown can be partially oxidized to malate, which is then converted to pyruvate in the mitochondrion [57]. Because of the low glycolytic rate, we predict that these amino acids contributed to the major pyruvate pool in undifferentiated chES cells. Generally, the pyruvate from this pool would be converted to phosphoenolpyruvate and metabolized outside the mitochondrion to generate sugar phosphates, such as G6P, through gluconeogenesis, if the glycolytic rate is low in cells [57]. However, undifferentiated chES cells completely block the gluconeogenic pathway, and thus pyruvate in these cells cannot be converted to sugar phosphates under the low glycolytic rate. This block occurs because some hormones, such as insulin, glucagon and glucocorticoid, and their receptors, which transcriptionally regulate gluconeogenic gene (Pcx and Pck2) expression [58–61] are not expressed until the gastrulation stage in the chicken [62,63]. We observed that yolk granules were either close to or directly associated with mitochondria. We can conclude from these findings that undifferentiated chES cells utilize amino acids from yolk protein breakdown in the mitochondria to generate pyruvate and ATPs, which become the main energy source for undifferentiated chES cells in vitro. In this study, we found that cell death was induced in the maintenance of stemness of chicken BLS cells in vitro. However, chES cell death in chicken BLSs was not caused by its spherical structure, because when the sphere was maintained in the differentiation medium, we observed a dramatic decline in cell death. Moreover, this cell death could be rescued by exogenous G6P, but not by exogenous energy (ATP) or other energy substrates, such as pyruvate or glucose. It has been reported that G6P can transmit an anti-death signal in the Xenopus egg and oocyte and that caspase-2 induced cell death would occur, if the levels of G6P committed to the pentose-phosphate-pathway cannot be maintained [64]. According to this, the G6P level is insufficient to inhibit undifferentiated chES cell death. Furthermore, this may be attributed, at least partially, to the intensive glycogen accumulation by undifferentiated chES
September (2010) Vol.53 No.9
cells, which “pull” a large amount of G6P into the glycogenic pathway. Consistently, we observed a dramatic decline in cell death once the chES cells stopped synthesizing glycogen during differentiation. In addition, yolk protein breakdown, which has been shown to be the main G6P source for the Xenopus egg and oocyte [57], does not contribute to cytosolic G6P production in undifferentiated chES cells. Taken together, we conclude that G6P, once generated, is shunted directly into the glycogenic pathway, and only a small amount is available to enter the glycolytic pool and the pentose phosphate pathway. Within the developing animal, the G6P level can be maintained as long as yolk protein stores are continually feeding G6P production. However, this route is blocked by undifferentiated chES cells in vitro. Therefore, severe cell death is induced when chES cells are maintained in an extracellular milieu without this critical substrate. All of the conclusions in this research were made by studying chES cells in a previously established model in which chES cells can only be maintained for a limited time, and hence this precludes their application on chES cells for several passages. However, our present study identifies some aspects of the metabolic properties of chES cells in the undifferentiated state (Figure 10) versus the differentiated state, indicating that chES cells are a great model to study the cell signaling that links the energy metabolic profile with cell proliferation and differentiation. The studies presented here also provide solutions for improving chES cell viability in vitro. With increased cell viability, early passage chES cells can be used as a tool to introduce genetic modifications and increase the efficiency of transgenic chicken production.
Figure 10 A model of the metabolic pathways in undifferentiated chES cells. Glucose, once transferred into the undifferentiated chES cell, is phosphorylated by HK1 into G6P. G6P is immediately incorporated into glycogen by GYS catabolism. Only small amounts of G6P flow into the glycolytic pathway to generate pyruvate and into the pentose phosphate pathway to generate NAPDH. Amino acids from the yolk protein breakdown are partially converted into pyruvate in the mitochondrion by the malic enzyme. This pyruvate will go into the Krebs cycle to generate energy (ATP) for undifferentiated chES cells.
Li Jia, et al.
Sci China Life Sci
This work was supported by the National Basic Research Program of China (Grant No. 2006CB102100) and the National Natural Science Foundation of China (Grant No. 30471234). 1
2
3 4
5
6 7
8
9 10
11 12
13
14
15
16
17 18
19
20
21
22
23
Petitte J N, Clark M E, Liu G, et al. Production of somatic and germline chimeras in the chicken by transfer of early blastodermal cells. Development, 1990, 108: 185–189 Pain B, Clark M E, Nakazawa H, et al. Long-term in vitro culture and characterisation of avian embryonic stem cells with multiple morphogenetic potentialities. Development, 1996, 122: 2339–2348 Van de lavoir M, Mather-Love C. Avian embryonic stem cells. Methods Enzymol, 2006, 418: 38–64 Eyal-Giladi H, Kochav S. From cleavage to primitive streak formation: a complementary normal table and a new look at the first stages of the development of the chick. I. General morphology. Dev Biol, 1976, 49: 321–337 Carsience R S, Clark M E, Verrinder Gibbins A M, et al. Germline chimeric chickens from dispersed donor blastodermal cells and compromised recipient embryos. Development, 1993, 117: 669–675 Etches R J, Clark M E, Zajchowski L. Manipulation of blastodermal Cells. Poult Sci, 1997, 76: 1075–1083 Ginsburg M, Eyal-Giladi H. Primordial germ cell development in cultures of dispersed central disks of stage X chick blastoderms. Gamete Res, 1989, 23: 421–427 Jung J G, Kim D K, Park T S, et al. Development of novel markers for the characterization of chicken primordial germ cells. Stem Cells, 2005, 23: 689–698 Petitte J N, Liu G, Yang Z. Avian pluripotent stem cells. Mech Dev, 2004, 121: 1159–1168 van de Lavoir M C, Mather-Love C, Leighton P, et al. High-grade transgenic somatic chimeras from chicken embryonic stem cells. Mech Dev, 2006, 123: 31–41 Eyal-Giladi H, Raveh D, Feinstein N, et al. Glycogen metabolism in the prelaid chick embryo. J Morphol, 1979, 161: 23–38 Greenfield M L. The oocyte of the domestic chicken shortly after hatching, studied by electron microscopy. J Embryol Exp Morphol, 1966, 15: 297–316 Paulson J, Rosenberg M D. The function and transposition of lining bodies in developing avian oocytes. J Ultrastmct Res, 1972, 40: 25–43 Eyal-Giladi H, Feinstein N, Friedlander M, et al. Glycogen metabolism and the nuclear envelope-annulate lamella system in the chick embryo. J Cell Sci, 1985, 73: 399–407 Perry M M, Gilbert A B. Yolk transport in the ovarian follicle of the hen (Gallus domesticus): lipoprotein-like particles at the periphery of the oocyte in the rapid growth phase. J Cell Sci, 1979, 39: 257–272 Barber D L, Sanders E J, Aebersold R, et al. The receptor for yolk lipoprotein deposition in the chicken oocyte. J Biol Chem, 1991, 266: 18761–18770 Shen X, Steyrer E, Retzek H, et al. Chicken oocyte growth: receptor-mediated yolk deposition. Cell Tissue Res, 1993, 272: 459–471 Deeley R G, Mullinix K P, Wetekam K, et al. Vitellogenin synthesis in the avian liver. Vitellogenin is the precursor of the egg yolk phosphoproteins. J Biol Chem, 1975, 250: 9060–9066 Griffin H D, Perry M M, Gilbert A B. Yolk formation. In: Freeman B M, ed. Physiology and Biochemistry of the Domestic Fowl. London: Academic Press, 1984. 345–380 Stevens L. ed. Lipids and their metabolism. In: Avian Biochemistry and Molecular Biology. Cambridge: Cambridge University Press, 1996. 46–64 Kochav S, Ginsburg M, Eyal-Giladi H. From cleavage to primitive streak formation: a complementary normal table and a new look at the first stages of the development of the chick. II. Microscopic anatomy and cell population dynamics. Dev Biol, 1980, 79: 296–308 Ishiguro S, Yamaguchi H, Oka Y, et al. Change in energy metabolism in the cell cycle of mouse L cells. Cell Struct Func, 1978, 3: 331–340 Carvalhala A V, Coroadinhaa A S, Alvesa P M, et al. Metabolic
September (2010) Vol.53 No.9
24
25
26
27
28
29
30
31
32
33
34 35 36 37
38 39 40
41
42 43 44
45
46
1083
changes during cell growth inhibition by the IRF-1 system. Enzyme Microb Technol, 2002, 30: 95–109 Chignola R, Milotti E A. Phenomenological approach to the simulation of metabolism and proliferation dynamics of large tumour cell populations. Phys Biol, 2005, 2: 8–22 Linke S P, Clarkin K C, Di Leonardo A, et al. A reversible p53-dependent G0/G1 cell cycle arrest induced by ribonucleotide depletion in the absence of detectable DNA damage. Genes Dev, 1996, 10: 934–947 Takahashi S, Satomi A, Yano K, et al. Estimation of glycogen levels in human colorectal cancer tissue: relationship with cell cycle and tumor outgrowth. J Gastroenterol, 1999, 34: 474–480 Pearce E L, Walsh M C, Cejas P J, et al. Enhancing CD8 T-cell memory by modulating fatty acid metabolism. Nature, 2009, 460: 103–107 Gustafsson M V, Zheng X, Pereira T, et al. Hypoxia requires notch signaling to maintain the undifferentiated cell state. Dev Cell, 2005, 9: 617–628 Panchision D M, Pickel J M, Studer L, et al. Sequential actions of BMP receptors control neural precursor cell production and fate. Genes Dev, 2001, 15: 2094–2110 Chen H L, Panchision D M. Concise review: bone morphogenetic protein pleiotropism in neural stem cells and their derivatives-alternative pathways, convergent signals. Stem Cells, 2007, 25: 63–68 Bergeron M, Yu A Y, Solway K E, et al., Sharp F R. Induction of hypoxia-inducible factor-1 (HIF-1) and its target genes following focal ischaemia in rat brain. Eur J Neurosci, 1999, 11: 4159–4170 Iyer N V, Kotch L E, Agani F, et al. Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor1α. Gene Dev, 1998, 12: 149–162 Minchenko A, Leshchinsky I, Opentanova I, et al. Hypoxia-inducible factor-1-mediated expression of the 6-phosphofructo-2-kinase/fructose2, 6-bisphosphatase-3 (PFKFB3) gene. Its possible role in the Warburg effect. J Biol Chem, 2002, 277: 6183–6187 Jacobs R S, Rathmell J C. Lymphocyte selection by starvation: glucose metabolism and cell death. Trends Immunol, 2006, 27: 4–7 Xiong J, Yankee T N. Notch and IL-7 regulate cell fate decisions in TCRβ+ DN and ISP thymocytes. J Immunol, 2009, 182: 85.10 Pain B, Chenevier P, Samarut J. Chicken embryonic stem cells and transgenic strategies. Cells Tissues Organs, 1999, 165: 212–219 Swanberg S E, Payne W S, Hunt H D, et al. Telomerase activity and differential expression of telomerase genes and c-myc in chicken cells in vitro. Dev Dyn, 2004, 231: 14–21 Li J, Pan Q, Li J, et al. Research of blastocyte-like structure in chicken. Sci China Ser C-Life Sci, 2005, 48: 481–486 Kessel R G. The annulate lamellae-from obscurity to spotlight. Electron Microsc Rev, 1989, 2: 257–348 Kessel R, Beams H W. Freeze fracture and scanning electron microscope studies on the envelope and perinuclear cytomembranes (parabasal apparatus) in the protozoan, Lophomonas blattarum. J Submicrosc Cytol Pathol, 1990, 22: 367–378 Hubbard M J, Dent P, Smythe C, et al. Targetting of protein phosphatase 1 to the sarcoplasmic reticulum of rabbit skeletal muscle by a protein that is similar or identical to the G subunit that directs the enzyme to glycogen. Eur J Biochem, 1990, 189: 243–249 Mori M, Dempo K, Abe M, et al. Electron microscopic study of intranuclear glycogen. J Electron Microsc (Tokyo), 1970, 19: 163–169 Karaski S. Cytoplasmic and nuclear glycogen synthesis in Novikoff ascites hepatoma cells. J Ultrastruct Res, 1971, 35: 181–196 Ferrans V J, Maron B J, Buja L M, et al. Intranuclear glycogen deposits in human cardiac muscle cells: ultrastructure and cytochemistry. J Mol Cell Cardiol, 1975, 7: 373–386 Granzow C, Kopun M, Zimmermann H P. Role of nuclear glycogen synthase and cytoplasmic UDP glucose pyrophosphorylase in the biosynthesis of nuclear glycogen in HD33 Ehrlich-Lettre ascites tumor cells. J Cell Biol, 1981, 89: 475–484 Kopun M, Granzow C, Krisman C R. Comparative study of nuclear and cytoplasmic glycogen isolated from mutant HD33 ascites cells. J Cell Biochem, 1989, 39: 185–195
1084 47
48
49
50
51
52 53
54
55
Li Jia, et al.
Sci China Life Sci
Dworkin M B, Dworkin-Rastl E. Metabolic regulation during early frog development: glycogenic flux in Xenopus oocytes, eggs, and embryos. Dev Biol, 1989, 132: 512–523 Hartl P, Olson E, Dang T, et al. Nuclear assembly with lambda DNA in fractionated Xenopus egg extracts: an unexpected role for glycogen in formation of a higher order chromatin intermediate. J Cell Biol, 1994, 124: 235–248 Andres V, Walsh K. Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J Cell Biol, 1996, 132: 657–666 Martinez L A, Chen Y, Fischer S M, et al. Coordinated changes in cell cycle machinery occur during keratinocyte terminal differentiation. Oncogene, 1999, 18: 397–406 De Falco G, Comes F, Simone C. pRb: master of differentiation. Coupling irreversible cell cycle withdrawal with induction of muscle-specific transcription. Oncogene, 2006, 25: 5244–5249 Miller J P, Yeh N, Vidal A, et al. Interweaving the cell cycle machinery with cell differentiation. Cell Cycle, 2007, 6: 2932–2938 Marcus E A, Kintner C, Harris W. The role of GSK3beta in regulating neuronal differentiation in Xenopus laevis. Mol Cell Neurosci, 1998, 12: 269–280 Hoeflich K P, Luo J, Rubie E A, et al. Requirement for glycogen synthase kinase-3beta in cell survival and NF-kappaB activation. Nature, 2000, 406: 86–90 Wang Q, Zhou Y, Wang X, et al. p27 Kip1 nuclear localization and cyclin-dependent kinase inhibitory activity are regulated by glycogen synthase kinase-3 in human colon cancer cells. Cell Death Differ, 2008, 15: 908–919
September (2010) Vol.53 No.9
56 57
58
59
60
61
62
63
64
Rayasam G V, Tulasi V K, Sodhi R, et al. Glycogen synthase kinase 3: more than a namesake. Br J Pharmacol, 2009, 156: 885–898 Dworkin M B, Dworkin-Rastl E. Regulation of carbon flux from amino acids into sugar phosphates in Xenopus embryos. Dev Biol, 1990, 138: 177–187 Sutherland C, O’Brien R M, Granner D K. New connections in the regulation of PEPCK gene expression by insulin. Philos Trans R Soc Lond B Biol Sci, 1996, 351: 191–199 Scott D K, Strömstedt P E, Wang J C, et al. Further characterization of the glucocorticoid response unit in the phosphoenolpyruvate carboxykinase gene. The Role of the glucocorticoid receptor-binding sites. Mol Endocrinol, 1998, 12: 482–491 Barthel A, Schmoll D. Novel concepts in insulin regulation of hepatic gluconeogenesis. Am J Physiol Endocrinol Metab, 2003, 285: 685–692 Boonsaen T, Rojvirat P, Surinya K H, et al. Transcriptional regulation of the distal promoter of the rat pyruvate carboxylase gene by hepatocyte nuclear factor 3β/Foxa2 and upstream stimulatory factors in insulinoma cells. Biochem J, 2007, 405: 359–367 Girbau M, González-Guerrero P R, Bassas L, et al. Insulin receptors and insulin-like growth factor I receptors in embryos from gastrula until organogenesis. Mol Cell Endocrinol, 1992, 90: 69–75 Lu J W, McMurtry J P, Coon C N. Developmental changes of plasma insulin, glucagon, insulin-like growth factors, thyroid hormones, and glucose concentrations in chick embryos and hatched chicks. Poult Sci, 2007, 86: 673–683 Nutt L K, Margolis S S, Jensen M, et al. Metabolic regulation of oocyte cell death through the CaMKII-mediated phosphorylation of caspase-2. Cell, 2005, 123: 89–103