Fig 2: SOX17, TFAP2C, and BLIMP1 do not create human primordial germ-cell-like cells.(A) The transgene expression levels in the designated hiPSC clones stimulated with (+) or without (-) Dox (1.0 µg/ml) for 24 h. Two clones were examined for each transgene combination. To quantify the transgene expression levels by qRT-PCR, primer pairs for the rabbit ß-globin polyA sequence (rBGpA) or for the transgene coding sequences were used, and the ?Ct was calculated from the average Ct value of two housekeeping genes, RPLP0 and PPIA (set as 0). The expression levels of SOX17, TFAP2C and BLIMP1 in d2 human primordial germ-cell–like cells (BT+AG+ cells) measured by primer pairs for detecting their 3' untranslated regions are shown as references. n.d., not determined. (B) Western blot analysis of the expression of SOX17, BLIMP1, and TFAP2C in the SOX17/TFAP2C/BLIMP1 hiPSC clone without (-) or with (+) 1.0 µg/ml of Dox for 24 h. a-TUBULIN was used as loading control. (C, D) Expression dynamics of rBGpA (transgenes) and the indicated endogenous genes in the TFAP2C/BLIMP1 (C) and SOX17/BLIMP1 (D) clones induced by bone morphogenetic protein 4 (black) or Dox (red). d1: whole aggregates; d2/d4: BT+AG+ cells for induction by bone morphogenetic protein 4, BT+ cells for induction by Dox. For each gene, the ?Ct was calculated from the average Ct value of two housekeeping genes, RPLP0 and PPIA (set as 0). The bars indicate the mean value of each time point. (C, D) Two independent experiments for the TFAP2C/BLIMP1 (C) and SOX17/BLIMP1 (D) clones were performed.Source data are available for this figure.

Fig 3: SOX17, TFAP2C, and BLIMP1 do not create human primordial germ-cell-like cells (hPGCLCs).(A) A schematic of the vectors used for Dox-inducible expression. The coding sequence of each gene was cloned in the designated position. D4Z4: D4Z4 macrosatellite repeat insulator; EF1a: promoter sequence of human EEF1A1; rtTA: reverse tetracycline trans-activator; IRES: internal ribosome entry site; Neo: Neomycin resistance gene; tetO: Tet operator sequence; rBGpA: rabbit ß-globin polyadenylation signal; Puro/Hygro: resistance gene for puromycin/hygromycin. (B) Heat map representation of the expression levels of the indicated genes in the designated hiPSC clones stimulated with (+) or without (-) Dox (1.0 µg/ml) for 24 h. Two clones were examined for each transgene combination. To quantify the expression levels of the transgenes or endogenous genes by qRT-PCR, primer pairs for the rBGpA or 3' untranslated regions were used, respectively, and the ?Ct was calculated from the average Ct value of two housekeeping genes, RPLP0 and PPIA (set as 0). (C) The protocol for hPGCLC induction. iMeLC aggregates were induced for hPGCLC fate by bone morphogenetic protein 4 (BMP4) or Dox (1.0 µg/ml) in the presence of stem cell factor, EGF, and leukemia inhibitory factor. ActA: activin A; CHIR: CHIR99021. (D) Phase-contrast images of iMeLCs in the designated clones. No apparent morphological differences were seen among the clones. Representative images of at least two independent experiments are shown (shown in Fig 1F). Bar: 50 µm. (E) Bright-field and fluorescence (TFAP2C-EGFP [AG] and BLIMP1-tdTomato [BT]) images, and FACS analyses for BTAG expression in floating aggregates of the indicated transgene-expressing clones at day 4 of the indicated stimulation. (-): induction only with stem cell factor, EGF and leukemia inhibitory factor. Representative images of at least two independent experiments are shown (see Fig 1F). Bars, 200 µm. (F) Percentage of BT+AG+ cells of the indicated transgene-expressing clones with the indicated stimulations at day 2 (left) and day 4 (right). Dots represent values for each experiment and the bars represent their averages. The numbers of inductions performed are shown in parenthesis. (G, H) Expression dynamics of rBGpA (transgenes) and the indicated endogenous genes in the SOX17/TFAP2C (G) and SOX17/TFAP2C/BLIMP1 (H) clones induced by BMP4 (black) or Dox (red). d1: whole aggregates; d2/d4: BT+AG+ cells for induction by BMP4, BT+ cells for induction by Dox. For each gene, the ?Ct was calculated from the average Ct value of two housekeeping genes, RPLP0 and PPIA (set as 0). (G, H) Three independent experiments with two SOX17/TFAP2C clones (G) and three SOX17/TFAP2C/BLIMP1 clones (H) were performed.

Fig 4: GATA3/2, SOX17, and TFAP2C generate human primordial germ-cell-like cells.(A) Expression of GATA3 and GATA2 at d2 in bone morphogenetic protein (BMP) and Dox- or BMP-treated BT+ cells of the SOX17/TFAP2C/BLIMP1 clones, BMP and Dox-treated BT+ cells of the SOX17/TFAP2C clones, and BMP-treated BT+AG+ PGCLCs of the parental line. The transcriptome data were used and the values are shown in log2(RPM + 1). See Table S1 for the samples analyzed. (B) The transgene expression levels in the designated hiPSC clones stimulated with (+) or without (-) Dox (1.0 µg/ml) for 24 h. Two clones for MSX2/SOX17/TFAP2C and GATA3/SOX17/TFAP2C, one clone for GATA2/SOX17/TFAP2C, and four clones for GATA3/SOX17 expression were examined. To quantify the transgene expression levels by qRT-PCR, primer pairs for rabbit rBGpA or for the transgene coding sequences were used, and the ?Ct was calculated from the average Ct value of two housekeeping genes, RPLP0 and PPIA (set as 0). (C) Heat map representation of the expression levels of the indicated genes in the designated hiPSC clones stimulated with (+) or without (-) Dox (1.0 µg/ml) for 24 h. To quantify the expression levels of the transgenes or endogenous genes, primer pairs for rBGpA or the 3' untranslated regions were used, respectively, and the ?Ct was calculated from the average Ct value of two housekeeping genes, RPLP0 and PPIA (set as 0). The color coding is as indicated. (D) Phase-contrast images of iMeLCs of the designated clones. No apparent morphological differences were seen among the clones. Representative images of at least two independent experiments are shown. Bars, 50 µm. (E) Bright-field and fluorescence (AG/BT) images, and FACS analyses for BTAG expression in d2/d4 floating aggregates of the GATA3/SOX17/TFAP2C (left) and GATA3/SOX17 (right) clones induced by 0 (top) or 1.0 µg/ml (bottom) of Dox. Representative data for six (GATA3/SOX17/TFAP2C) and two (GATA3/SOX17) experiments are shown. Bars, 200 µm. (F) BT+AG+ cell induction efficiency (%) (left), BT+AG+ cell numbers per aggregate (middle), and total cell numbers per aggregate (right) of the GATA2/SOX17/TFAP2C and GATA3/SOX17/TFAP2C clones. Mean values (n = 3 for each line) are shown as red bars. (G) The principal component analysis plots of the cells (iMeLCs, Dox-induced d2/d4 BT+ cells) derived from the GATA3/SOX17 clone (squares), overlaid with the indicated cells derived from the parental clone (circles with pale color). See Table S1 for the samples analyzed. The color coding is as indicated.

Fig 5: Transcriptome analysis of the effects of the transcription factor expression.(A) (left) Principal component analysis of the expression of the 481 genes characterizing human primordial germ-cell–like cell specification shown in reference 13. Red (first quadrant): genes for PGCLC specification; cyan (fourth quadrant): genes for endoderm/mesoderm specification; yellow (second and third quadrants): genes for pluripotency (right). Heat map representation of the expression of the 481 genes in the parental hiPSCs, iMeLCs, bone morphogenetic protein (BMP)4-induced d1 whole aggregates, and d2/d4 BT+AG+ cells, and in TFAP2C/BLIMP1, SOX17/TFAP2C, SOX17/BLIMP1, SOX17/TFAP2C/BLIMP1 clone-derived, Dox-induced d2 BT+ cells. Representative genes are annotated, and the correlation coefficient values of each induction with the d4 BT+AG+ cells are shown. The color coding is as indicated. (B, C) Expression dynamics of the genes up-regulated in BMP-stimulated (black, BT+AG+) compared with Dox-stimulated (red, BT+) d2 cells (B) (Note that TFAP2A, HAND1, CDX2, HAPLN1, and MSX2 were highly up-regulated also at d1 [Fig 2H]) or those up-regulated in Dox-stimulated (red) compared with BMP-stimulated (black) SOX17/TFAP2C/BLIMP1 clone-derived d1 iMeLC aggregates (C). The bars indicate the mean value of each time point, and the ranks of the genes ordered by the fold changes between BMP- and Dox-stimulation are shown. See Table S1 for the samples analyzed. (D, E) Gene ontology terms (blue) and KEGG pathways (pink) enriched in differentially expressed genes between BMP and Dox- and Dox-stimulated SOX17/TFAP2C clone-derived cells at d2 (BMP and Dox: BT+AG+; Dox: BT+). (D, E) Representative genes up-regulated in BMP and Dox- (D) or Dox- (E) stimulations and P-values are shown.