Core pluripotency factors promote glycolysis of human embryonic stem cells by activating GLUT1 enhancer
Human embryonic stem cells (hESCs) depend on glycolysis for energy and substrates for biosynthesis. To understand the mechanisms governing the metabolism of hESCs, we investigated the transcriptional regulation of glucose transporter 1 (GLUT1, SLC2A1), a key glycolytic gene to maintain pluripotency. By combining the genome-wide data of binding sites of the core pluripotency factors (SOX2, OCT4, NANOG, denoted SON), chromosomal interaction and histone modification in hESCs, we identified a potential enhancer of the GLUT1 gene in hESCs, denoted GLUT1 enhancer (GE) element. GE interacts with the promoter of GLUT1, and the deletion of GE significantly reduces the expression of GLUT1, glucose uptake and glycolysis of hESCs, confirming that GE is an enhancer of GLUT1 in hESCs. In addition, the mutation of SON binding motifs within GE reduced the expression of GLUT1 as well as the interaction between GE and GLUT1 promoter, indicating that the binding of SON to GE is important for its activity. Therefore, SON promotes glucose uptake and glycolysis in hESCs by inducing GLUT1 expression through directly activating the enhancer of GLUT1.
Keywordshuman embryonic stem cell pluripotency factors metabolism Glut1 enhancer promoter epigenetics chromosome interaction
Human embryonic stem cells (hESCs) can undergo unlimited self-renewal and maintain the pluripotency to differentiate into all lineages of cells of the human body (De Los Angeles et al., 2015). This metabolic signature of pluripotency is similar to the Warburg effect in human cancers and is primarily dependent on glycolysis (Shyh-Chang and Daley, 2015). In this context, glycolysis produces ATP at a faster rate than oxidative phosphorylation, and glycolytic intermediates are biosynthesis substrates required for unlimited self-renewal of hESCs (Shyh-Chang and Daley, 2015). In addition, glycolysis produces acetyl-CoA to promote histone acetylation, which is required to maintain the epigenetics of hESCs (Moussaieff et al., 2015). The transition from oxidative phosphorylation to glycolysis also promotes the reprogramming of induced pluripotent stem cells (iPSCs) (Folmes et al., 2011). While the core transcriptional factors SRY (sex determining region Y)-box 2 (SOX2), octamer-binding transcription factor 4 (OCT4), and NANOG, collectively denoted SON, are required to maintain pluripotency (Chen et al., 2008), their roles in maintaining the metabolic profile of ESCs remain unclear.
The increase of glucose uptake is required to maintain high levels of glycolysis. GLUT1 plays a key role in glucose uptake in many cell types including ESCs and cancer cells (Shyh-Chang and Daley, 2015; Ancey et al., 2018). The expression of GLUT1 is significantly increased during early embryonic development from the two-cell stage to the blastocyst stage (Morita et al., 1994). Consistent with this finding, GLUT1 is also highly expressed in pluripotent stem cells (Shyh-Chang and Daley, 2015). Studies of GLUT1-deficient and GLUT1-haplodeficient mouse ESCs indicated that GLUT1 is required for the survival of pluripotent stem cells by maintaining high levels of glycolysis (Ohtsuki et al., 2006).
Enhancers are clusters of distal DNA sequences that can increase transcription of their target gene(s) in cis in eukaryote (Pennacchio et al., 2013). The activity of enhancers in the human genome is time- and cell type-dependent. Epigenetic markers commonly used to identify active enhancers include histone H3 acetylated at lysine 27 (H3K27ac) and H3 monomethylated at K4 (H3K4me1) (Deng et al., 2012; Calo and Wysocka, 2013). Chromatin Interaction Analysis with Paired-End-Tag sequencing (ChIA-PET) and Hi-C (Genome-wide 3C) demonstrate that enhancer–promoter interaction through chromosomal looping is necessary for transcriptional activation of genes (Dekker et al., 2002; Dostie et al., 2006; Zhao et al., 2006; Fullwood et al., 2009; Lieberman-Aiden et al., 2009). High-resolution interaction data of ChIA-PET can provide the information of chromatin interaction (Barutcu et al., 2016). When combining with the chromatin immunoprecipitation sequencing (ChIP-seq) data of enhancer histone markers, cohesin ChIA-PET data can help to accurately identify the enhancer-promoter loops (Ji et al., 2016).
In this study, we identified a novel enhancer for GLUT1 in hESCs, which appeared to be evolutionarily conserved in other pluripotent stem cells and cancer cells. In addition, we demonstrate that the binding site of SON within this enhancer is important for the enhancer activity and glucose uptake. Therefore, SON plays important roles in maintaining pluripotency by inducing glycolysis in hESCs.
To identify the enhancer of the GLUT1 gene in hESCs
GE is important for the expression of GLUT1 in hESCs
The SON binding site within GE is important for the enhancer function of GE
GE-deleted hESCs are defective in GLUT1 expression during differentiation
GE is conserved in iPSCs and human cancer cells
GLUT1 is important for high levels of glucose uptake to maintain glycolysis of ESCs and human cancer cells (Ancey et al., 2018). Therefore, it is important to reveal the mechanisms how hESCs regulate the expression of GLUT1. We identified an enhancer element of GLUT1 (GE) in hESCs that is required for the optimal levels of GLUT1 expression. To understand how this enhancer is activated in hESCs, we used the genome-wide ChIP-seq data to identify the proteins that can bind to GE and found that the pluripotency factors Sox2, Oct4 and Nanog all bind to GE. By disrupting this binding site of SON in the genome of hESCs, we showed that the binding sites are important for the binding of SON to GE, the long-range interaction between enhancer and promoter, and the activity of GE. Therefore, SON is important for inducing the expression of GLUT1 by activating its enhancer.
Mutations in GLUT1 can cause GLUT1 deficiency syndrome, which led a neurologic disorder and epilepsy in human (Schneider et al., 2009; Striano et al., 2012). In mouse model, homozygous loss of GLUT1 is associated with embryonic lethality and heterozygous mouse performed incoordination, hypoglycorrhachia and microencephaly, such as epilepsy (Wang et al., 2006; Zheng et al., 2010). In our result, decrease of GLUT1 in GE-deleted teratomas cause the decline of synapsis development and glucose metabolic process, which is essential in neuron development. The GE is essential in the early development of nervous system.
Glycolysis is required to maintian the pluripotency of hESCs (Shyh-Chang and Daley, 2015). Previous studies have shown multiple pathways that could play important roles in promoting glycolysis in ESCs. For example, the stemness factor SALL4 can promote glycolysis by inducing the expression of HIF1a and GLUT1 (Kim et al., 2017). In addition to maintaining the genomic stability of ESCs (Lin et al., 2005; Xu, 2005), the p53-PUMA pathway suppresses the oxidative phosphorylation by limiting pyruvate uptake into the mitochondria (Kim et al., 2019). As the core transcription factors to maintain the pluripotency of hESCs, the knockdown of NANOG, OCT4 and SOX2 will lead to rapid differentiation and death of hESCs, making it difficult to study the roles of SON in regulating the expression of GLUT1 in pluripotent state (Avilion et al., 2003; Mitsui et al., 2003; Ivanova et al., 2006). In this context, previous studies have failed to provide conclusive data on the roles of SON on the regulation of GLUT1 expression. Our data indicated that the mutation of SON binding motif decreases the expression of GLUT1 by disrupting the interaction between the enhancer and promoter of GLUT1. Therefore, SON plays important roles in activating the enhancer of GLUT1 by directly binding to it.
Enhancer activity is often cell type-specific (Pennacchio et al., 2013). As expected, the epigenetic signatures of GE in iPSCs are similar to those in hESCs. The published ChIP-Seq data indicate that OCT4 binds to GE in iPSCs (Fig. 4A). Therefore, it can be speculated that SON activate the transcription of GLUT1 by binding to GE. While the epigenetic signatures of some human cancer cell lines such as HepG2 and HCT116 indicate that GE is active in these cell lines, the shape of the peak signal of H3K27ac is different from that of ES and IPSCs, suggesting that the transcription factors other than SON might be in involved in activating GE in human cancer cells. In addition to GE, the analysis of the epigenetic signatures of enhancers in these human cancer cells identifies multiple potential enhancer elements for GLUT1, suggesting that the expression of GLUT1 might be regulated by multiple enhancer elements in human cancer cell cells.
MATERIALS AND METHODS
Genome editing of hESC culture
H1 and H9 hESC lines were maintained on matrigel-coated plates in complete mTeSR™1 medium, and passaged using accutase or ReleSR. All reagents were obtained from STEMCELL Technologies. The CRISPR/Cas9 technology was used to edit the genome of hESCs as previously described (Rong et al., 2014). For the knockout of GE in hESCs, two expression cassettes encoding the sgRNA sequences (sgRNA1-GE: GGAAAAGGCTGGGAGGCCAG, sgRNA2-GE: GGCTGCTGTGATGCTCGAAT) flanking the deletion region were cloned into a plasmid that expresses a codon-optimized version of Cas9. For the mutation of the SON binding site within GE, the expression cassette encoding one sgRNA (sgRNA-SON: GGAACCTTTGTCATTCAAAC) targeting the SON motif was cloned into a plasmid that expresses a codon-optimized version of Cas9. To transfect the plasmid into the hESCs, hESCs were harvested using accutase and nucleofected in 4D-Nucleofector® Solution. The transfected hESCs were selected with puromycin and individual clones expanded and genotyped. The genotyping primers are following: 5′-AGGTCTCCCAAGTCTAGCGT-3′, 5′-TGATTACCGCAAAGCCCCAA-3′, 5′-CCCAAAACAGGGGATCCTGAA-3′.
The analysis of ChIA-PET and ChIP-seq data
The ChIA-PET and ChIP-seq data were obtained from Gene Expression Omnibus (GEO) and Encyclopedia of DNA Elements (ENCODE). The accession were GSM1505699, GSM1505728, GSM1565766, GSM2534369, GSE57913, GSE44288, GSM1000126, GSE29611, GSE69643 and GSE69646 (Consortium, 2012; Whyte et al., 2013; Dowen et al., 2014; Pope et al., 2014; Yue et al., 2014; Tsankov et al., 2015; Ji et al., 2016). The interaction data derived from cohesin ChIA-PET analysis were displayed in BED12 format that showed the anchors and coordinates of the loop (Ji et al., 2016). The insulator loops were colored in red and the others in green. The ChIP-Seq data were displayed using the UCSC Genome Browser (http://genome.ucsc.edu/). All analyses of hESCs were performed using human (build hg19, GRCh37) RefSeq annotations downloaded from the UCSC genome browser.
Analysis of gene expression profile
The gene expression profiles of hESCs after the knockdown of NANOG, OCT4 and SOX2 were downloaded from GEO (accession GSE34904) and analyzed by Qlucore Omics Explorer 3.3 (http://www.qlucore.se/) (Wang et al., 2012). P-value (two-tailed) was calculated with two-group comparisons.
Chromosome conformation capture (3C) analysis
Real-time PCR was performed as previously described (Zhang et al., 2014). Briefly, total RNA was purified from hESCs with RNeasy Mini Kit (QIAGEN), and total RNA (1 µg) was reversely transcribed into cDNA and analyzed by qPCR. The primers for β-actin are 5′-GCCAACACAGTGCTGTCT-3′ (forward primer) and 5′-AGGAGCAATGATCTTGATCTT-3′ (reverse primer). The primers for GLUT1 are 5′-CTTTGTGGCCTTCTTTGAAGT-3′ (forward primer) and 5′-CCACACAGTTGCTCCACAT-3′ (reverse primer). The primers for NANOG are 5′-CATGAGTGTGGATCCAGCTTG-3′ (forward primer) and 5′-CCTGAATAAGCAGATCCATGG-3′ (reverse primer). The primers for OCT4 are 5′-AGTGAGAGGCAACCTGGAGA-3′ (forward primer) and 5′-ACACTCGGACCACATCCTTC-3′ (reverse primer). The primers for SOX2 are 5′-TGGACAGTTACGCGCACAT-3′ (forward primer) and 5′-CGAGTAGGACATGCTGTAGGT-3′ (reverse primer). The levels of GLUT1 mRNA were normalized to those of β-actin. Changes in mRNA expression were calculated according to the 2−ΔΔCT method (CT, cycle threshold).
Extracellular acidification rate (ECAR) was measured with the Seahorse XFe96 Analyzer (Seahorse Bioscience). H1 cells (2 × 104/well) were seeded in Matrigel-coated 96-well XF Cell Culture Microplate and incubated overnight. The next day, cells were pre-incubated in XF assay media (supplemented with 2 mmol/L L-glutamine) for one hour prior to the assay. Glycolysis Stress Test was performed following manufacturer’s protocol. The obtained ECAR was normalized by fluorescence intensity of DAPI stained nuclei and analyzed using the XF Report Generator (Seahorse Bioscience).
ChIP qPCR assay
Chromatin immunoprecipitation (ChIP) assays were performed using a SimpleChIP Enzymatic Chromatin IP kit (No. 9003; Cell Signaling Technologies). Briefly, hESCs were cross-linked with 1% formaldehyde at room temperature for 10 min. Chromatin was treated with micrococcal nuclease, sonicated, and immunoprecipitated with rabbit anti-acetyl-Histone H3 (Lys27) antibody (8173s; CST), rabbit anti-mono-methyl-Histone H3 (Lys4) antibody (5326s; CST), rabbit anti-Nanog (D73G4) antibody (5232s; CST), rabbit anti-Oct-4 (C30A3C1) antibody (5677s; CST), rabbit anti-Sox2 antibody (D6D9) (5024s; CST), and normal rabbit IgG (negative control) (2729; CST). After the reverse cross-linking and DNA purification, immunoprecipitated DNA was quantified by qPCR (5′-GGTTCTTTCTTCCACCGCGT-3′ and 5′-AGCAAGAATCCCAACCCCG-3′).
Western blotting analysis was performed as we previously described (Kim et al., 2015). Monoclonal antibodies used: anti-Glut1 (ab150299; Abcam) and anti-β-Actin (ab8227; Abcam). The intensity of the bands was quantified using Image Lab software.
GE sequence from various species was compared using BLASTN 2.9.0 in NCBI. The algorithm of Fast Minimum Evolution was used to produce the tree from given distances between sequences of species (Desper and Gascuel, 2004).
GraphPad Prism 5 was used for statistical analysis. For comparisons between two groups of equal sample size, an unpaired two-tailed t test was performed. For comparisons of two groups of paired samples, paired two-tailed t test was performed. P < 0.05 was considered to be statistically significant.
For teratoma fomation of hESCs in immunodeficient mice, 1.5 × 106 hESCs and GE-deleted hESCs were harvested, washed twice with PBS, suspended in PBS with 30% Matrigel, and subcutaneously injected into region around the right (WT hESCs) and left (GE-KO hESCs) hind legs of immunodeficient mice. The teratomas were recovered 40 days after transplantation. Total RNA was purified from the teratomas with Trizol and processed for RNA-seq. All institutional and national guidelines for the care and use of laboratory animals were followed.
RNA-seq and analysis
RNA purity was checked using the kaiaoK5500®Spectrophotometer (Kaiao, Beijing, China). RNA integrity and concentration were assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA, USA). A total amount of 2 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (#E7530L, NEB, USA). Paired-end sequencing was completed on an Illumina HiSeq system. Clean data were renerated after removing adapters and low quality reads. The reference GRCh38 genomes and the annotation file were downloaded from ENSEMBL database (http://www.ensembl.org/index.html). Bowtie2 v2.2.3 was used for building the genome index, and Clean Data was then aligned to the reference genome using HISAT2 v2.1.0. Reads Count for each gene in each sample was counted by HTSeq v0.6.0, and FPKM (Fragments Per Kilobase Millon Mapped Reads) was then calculated to estimate the expression level of genes in each sample. The differentially expressed genes were analyzed with Qlucore Omics Explorer 3.3. Gene ontology (GO) biological process enrichment was analyzed by DAVID (https://david.ncifcrf.gov).
This study was supported by National Natural Science Foundation of China (81430032, U1601222), the leading talents of Guangdong Province Program (No. 00201516), the Key Research and Development Program of Guangdong Province (2019B020235003), Guangdong Provincial Key Laboratory of Cancer Immunotherapy, Major basic research developmental project of the Natural Science Foundation of Guangdong Province (2014A030308018), Science and Technology Innovation Committee of Shenzhen Municipality (JCYJ20180504170301309), and Shenzhen “Sanming” Project of Medicine (SZSM201602102).
CONFLICTS OF INTEREST
Lili Yu, Kai-yuan Ji, Jian Zhang, Yanxia Xu, Yue Ying, Taoyi Mai, Shuxiang Xu, Qian-bing Zhang, Kai-tai Yao, and Yang Xu declare no competing financial interests.
K.J., L.Y., K.Y. and Y.X. designed the research. K.J. and L.Y. performed the majority of experiments with the help of J.Z., Y.X., Y.Y., T.M., S.X., Q-B.Z., K.J., L.Y., K.Y. and Y.X. interpreted the data. Y.X. provided the funding support. K.J., L.Y. and Y.X. were responsible for the initial draft of the manuscript, whereas other authors contributed to the final version.
- Zhao Z, Tavoosidana G, Sjolinder M, Gondor A, Mariano P, Wang S, Kanduri C, Lezcano M, Sandhu KS, Singh U et al (2006) Circular chromosome conformation capture (4C) uncovers extensive networks of epigenetically regulated intra- and interchromosomal interactions. Nat Genet 38:1341–1347CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.