A Genetic Screen Identifies Etl4-Deficiency Capable of Stabilizing the Haploidy in Embryonic Stem Cells

Abstract

Mammalian haploid embryonic stem cells (haESCs) hold great promise for functional genetic studies and forward screening. However, all established haploid cells are prone to spontaneous diploidization during long-term culture, rendering application challenging. Here, we report a genome-wide loss-of-function screening that identified gene mutations that could significantly reduce the rate of self-diploidization in haESCs. We further demonstrated that CRISPR/Cas9-mediated Etl4 knockout (KO) stabilizes the haploid state in different haESC lines. More interestingly, Etl4 deficiency increases mitochondrial oxidative phosphorylation (OXPHOS) capacity and decreases glycolysis in haESCs. Mimicking this effect by regulating the energy metabolism with drugs decreased the rate of self-diploidization. Collectively, our study identified Etl4 as a novel haploidy-related factor linked to an energy metabolism transition occurring during self-diploidization of haESCs.

Keywords: Etl4; Pluripotent stem cell; energy metabolism; genetic screen; haploid embryonic stem cells; self-diploidization.

Graphical abstract

Figure 1

A Genetic Screen to Identify Factors Stabilizing Haploidy in Mouse haESCs (A) Schematic representation of the genetic screening procedure. (B) Box plot of average read frequencies of transposon insertions in IML, SML, and CML. Dots represent the top 1% insertions. The upper vertical line represents the 75%–99% percentile group. Boxes indicate the distribution of 25%–75% of insertions. The lower vertical line represents the 1%–25% percentile group. (C) The reads for the top 50 hits in SML and CML were increased compared with IML. (D) Top 10 candidates obtained from the next generation sequencing (NGS) of SML and CML. S and C refer to SML and CML, respectively. Candidate genes obtained from IML, SML, and CML are listed in Table S3. (E) Validation of the NGS results via Sp-PCR. The ratio represents the percentage of each insertion site to the total in Sp-PCR. The f (NGS) means the percentage of each insertion site to the total in NGS. The primers of Sp-PCR are listed in Table S2.

Figure 2

Etl4 Deficiency Facilitates the Maintenance of Mouse haESCs (A) Schematic diagram of the strategy to knockout Etl4 and Adam12 in OdG haESCs by using CRISPR/Cas9 system. Boxes indicate the exon of the genes and the triangles represent the single guide RNA (sgRNA). The sequences of the sgRNAs are listed in Table S1. (B) Validation of Etl4 and Adam12 expression level in mutant and WT OdG haESCs by RT-qPCR. The expression level was normalized by Gapdh (n = 3 independent experiments). (C) Chromosome spreads of Etl4-KO OdG haESCs showing normal haploidy. The right panel depicts the self-diploidized Etl4-KO cells. (D) Morphology of Etl4 and Adam12-deficient OdG haESCs cultured in M2iL medium. Scale bars, 100 μm. (E and F) Flow cytometry analysis of the ratio of haploid cells in (E) Etl4-KO, Adam12-KO and (F) two corresponding WT haESC lines during cell passages. (G) Expressional level of Etl4 in OG3, Etl4-GT, RVT-1, and RVT-2 haESCs tested by RT-qPCR (n = 3 independent experiments). (H) Percentages of haploid cells in OG3, Etl4-GT, RVT-1, and RVT-2 haESCs after 12 days of culture (n = 3 independent experiments). (I) Alkaline phosphatase activity in Etl4-KO and WT OdG haESCs. Scale bars, 50 μm. (J) Immunofluorescence (IF) staining for the indicated pluripotency markers in Etl4-KO and WT OdG haESCs. Scale bars, 50 μm. (K) Validation of indicated pluripotency-related markers' expression level in Etl4-KO and WT OdG haESCs by RT-qPCR. The expression level was normalized by Gapdh (n = 3 independent experiments). (L) Cell proliferation rate of Etl4-KO and WT OdG haESCs (n = 3 independent experiments). Data in (B), (E), (F), (G), and (H) are shown as mean ± SEM, ^∗p < 0.05 and ^∗∗p < 0.01, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001. NS, not significantly different.

Figure 3

Transcriptome of Etl4-deficient OdG haESCs (A) Hierarchical clustering of gene expression profiles from two biologically independent samples based on Pearson correlation coefficient in two Etl4-KO and WT haES clones. Colors from green to red indicate weak to strong correlations. (B) MA plots showing gene expression changes in Etl4-KO haESCs. Red dots indicate upregulated genes and blue dots indicate downregulated genes. (C and D) GO analysis of upregulated genes (C) and downregulated genes (D) for biological processes in Etl4-KO OdG haESCs. (E–G) Heatmap showing the expression of methylation related genes (E), cell-cycle–related genes (F), and Rho A pathway (G) in Etl4-KO and WT OdG haESCs. (H) Significantly enriched KEGG pathways for genes expressed differentially in Etl4-KO OdG haESCs. (I) GSEA for anaerobic glycolysis gene set in WT OdG and Etl4-KO haESCs. For the x axis, genes were ranked based on the ratio of Etl4-KO versus WT haESCs. (J) Heatmap showing changes in RNA expression levels for various enzymes and regulators of central carbon metabolism in Etl4-KO and WT OdG haESCs. The scale represents Z score.

Figure 4

Reduced Self-Diploidization by Promotion of OXPHOS or Inhibition of Glycolysis (A) Mitochondrial stress test to detect mitochondrial energy metabolism and respiratory functions in Etl4-KO and WT OdG haESCs. (B) Quantification of the mitochondrial stress test presented in (A) (n = 3 independent experiments). (C) The glycolytic stress test to measure glycolytic activities in Etl4-KO and WT OdG haESCs. (D) Quantification of glycolytic activities in (C) (n = 3 independent experiments). (E and F) DNA content analysis of haploid cells in OdG (E) and EG1 (F) haESCs treated with 3-BrPA (12.5 μg/mL) and 1,25-(OH)₂D3 (5 μM) for 12 days. Data in (B), (D), (E), and (F) are presented as means ± SEM, ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001.