Capacitation of human naïve pluripotent stem cells for multi-lineage differentiation

Abstract

Human naïve pluripotent stem cells (PSCs) share features with the pre-implantation epiblast. They therefore provide an unmatched opportunity for characterising the developmental programme of pluripotency in Homo sapiens Here, we confirm that naïve PSCs do not respond directly to germ layer induction, but must first acquire competence. Capacitation for multi-lineage differentiation occurs without exogenous growth factor stimulation and is facilitated by inhibition of Wnt signalling. Whole-transcriptome profiling during this formative transition highlights dynamic changes in gene expression, which affect many cellular properties including metabolism and epithelial features. Notably, naïve pluripotency factors are exchanged for postimplantation factors, but competent cells remain devoid of lineage-specific transcription. The gradual pace of transition for human naïve PSCs is consistent with the timespan of primate development from blastocyst to gastrulation. Transcriptome trajectory during in vitro capacitation of human naïve cells tracks the progression of the epiblast during embryogenesis in Macaca fascicularis, but shows greater divergence from mouse development. Thus, the formative transition of naïve PSCs in a simple culture system may recapitulate essential and specific features of pluripotency dynamics during an inaccessible period of human embryogenesis.

Keywords: Competence; Differentiation; Epiblast; Human embryo; Lineage specification; Pluripotent stem cell.

Fig. 1.

Naïve hPSCs do not respond to lineage differentiation cues. (A) Bright-field images of aggregates cultured for 7 days in N2B27, unstained and stained with Trypan Blue. Images were acquired with the same 10× objective and are shown in scale. (B) RT-qPCR analysis of marker expression after 14 days of aggregation culture. Here, and in all qPCR measurements, error bars represent s.d. of technical duplicates. (C) Flow cytometry analysis of intracellular marker staining after application of dual SMAD inhibition for neuroectoderm induction. SSC-A, forward scatter. (D) RT-qPCR analysis following dual SMAD inhibition. (E) Flow cytometry analysis of cell surface (CXCR4) and intracellular (SOX17) marker expression after application of definitive endoderm induction conditions. (F) RT-qPCR analysis in definitive endoderm induction conditions. (G) Cell counts in paraxial mesoderm induction conditions. Error bars are derived from two independent experiments. (H) RT-qPCR analysis following application of paraxial mesoderm induction protocol. EB, embryoid bodies; Un, undifferentiated.

Fig. 2.

Transition of naive hPSCs in N2B27 only. (A) Bright-field images of naïve hPSCs and cultures in N2B27 for 7 or 12 days, or in E8 for 14 days. Images were acquired with the same 10× objective and are shown in scale. (B) Colony counts after culture in N2B27 or E8 and replating at low density in naïve PSC conditions. (C) RT-qPCR analysis of pluripotency marker expression during culture in N2B27 or E8. Dashed lines indicate that the intermediate time points were not analysed. (D) Flow cytometry analysis of neuroectoderm and definitive endoderm markers following induction after 7 days in N2B27 or E8. SSC, side scatter. (E) Summary of efficiencies of differentiation after 7 days in N2B27 or E8, from multiple independent experiments. (F) RT-qPCR analysis of meso-endodermal marker expression after 7 days in N2B27 or E8. (G) Immunostaining for TBXT and NANOG on day 7 in the indicated conditions.

Fig. 3.

Transition of naïve hPSCs is facilitated by XAV939. (A) Bright-field images of conventional hPSCs and naive hPSCs before and after culture in N2B27 plus XAV939 for 14 days. Images were acquired with the same 10× objective and are shown in scale. (B) Cell counts during transition in XAV939; error bars represent s.d. from three independent experiments. (C) Colony counts after culture in XAV939 and replating at low density in naïve PSC conditions. (D) RT-qPCR assay of pluripotency gene expression during transition in XAV939. Dotted lines indicate expression in conventional H9 hES cells. (E) Immunofluorescent staining for pluripotency markers during transition in XAV939. H9 conventional hPSCs provide reference staining. (F) Quantification of intensity of immunostaining. Mean intensity of cells stained with the secondary antibodies only was subtracted from the measurement of each cell that was stained with specific antibodies. The resulting values of intensity above control are represented as boxplots.

Fig. 4.

Differentiation competence of hPSCs after formative transition in XAV939. (A-C) Neuroectoderm induction after transition in XAV939, parallel treatment of conventional hPSCs and naïve hPSCs after capacitation. Neuroectoderm markers were examined using flow cytometry (A), immunostaining (B) and RT-qPCR (C). (D-F) Definitive endoderm induction after transition in XAV939. Markers were examined using immunostaining (D), flow cytometry (E) and RT-qPCR (F). (G,H) Differentiation to paraxial mesoderm after transition in XAV939. Markers were analysed using RT-qPCR (G) and immunostaining for TBX6 and CDX2 (H). DE, definitive endoderm; NE, neuroectoderm; PM, paraxial mesoderm; SSC-A, side scatter; Un, undifferentiated (conventional and capacitated hPSCs that were not induced to differentiation).

Fig. 5.

Global transcriptome analysis during formative transition. (A) PCA of HNES1 and cR-H9EOS during formative transition. For each cell line, three independent experiments are represented. (B) Heat maps showing expression of variable genes assigned to five dynamic clusters, derived separately for HNES1 and cR-H9-EOS. (C) Numbers of protein coding genes, and of transcription factors, co-factors and epigenetic remodellers (TF/coF/REM), within the five dynamic clusters. (D) Comparison of gene content of clusters between HNES1 and cR-H9-EOS; 61.7% of genes belong to the same clusters and 83.9% belong to the same or similar clusters. (E) Heat map of RNAseq expression values for selected genes during formative transition.

Fig. 6.

Comparison of gene expression during formative transition of hPSCs in vitro and M. fascicularis embryonic epiblast in utero. (A) Soft clustering analysis of variable genes during developmental progression of mouse and macaque epiblast in utero, and transition of hPSCs in vitro; average level of gene expression per cluster is indicated. (B) Comparison of clusters of variable genes between progression of mouse or macaque epiblast in utero and hPSCs in vitro. (C) Expression of selected genes during progression of mouse and macaque epiblast in utero and hPSCs in vitro. (D) PCA of macaque epiblast single cells at different stages of development in utero and hPSC populations during formative transition in vitro. (E,F) Fractions of similarity of hPSCs during formative transition to embryonic stages EPI, post-E and post-L of the macaque embryo: 3D plot (E) and 2D projections (F).