Human pluripotent stem cell culture density modulates YAP signaling

Abstract

Human pluripotent stem cell (hPSC) density is an important factor in self-renewal and differentiation fates; however, the mechanisms through which hPSCs sense cell density and process this information in making cell fate decisions remain to be fully understood. One particular pathway that may prove important in density-dependent signaling in hPSCs is the Hippo pathway, which is regulated by cell-cell contact and mechanosensing through the cytoskeleton and has been linked to the maintenance of stem cell pluripotency. To probe regulation of Hippo pathway activity in hPSCs, we assessed whether Hippo pathway transcriptional activator YAP was differentially modulated by cell density. At higher cell densities, YAP phosphorylation and localization to the cytoplasm increased, which led to decreased YAP-mediated transcriptional activity. Furthermore, total YAP protein levels diminished at high cell density due to the phosphorylation-targeted degradation of YAP. Inducible shRNA knockdown of YAP reduced expression of YAP target genes and pluripotency genes. Finally, the density-dependent increase of neuroepithelial cell differentiation was mitigated by shRNA knockdown of YAP. Our results suggest a pivotal role of YAP in cell density-mediated fate decisions in hPSCs.

Keywords: Cell density; Human pluripotent stem cells; Neural differentiation; Self-renewal; YAP.

Figure 1

YAP localization switched from nuclear to cytoplasmic as cell density increased. hPSCs were singularized and plated at 0.1, 0.2 and 0.4 × 10⁵ cell/cm². Cells were fixed and stained after 2, 3 and 4 days of culture. (A, B) Representative confocal images of the nuclear Hoechst stain and (A) YAP immunofluorescence and (B) phospho-YAP (Ser127) immunofluorescence at days 2 and 4 are shown. Scale bars represent 10 µm. (C, D) Pearson’s coefficients were calculated for 5 to 7 images and averaged for each condition to quantify the colocalization of YAP and phospho-YAP immunofluorescence with the Hoechst stain. Pearson’s coefficient of 1 represents complete colocalization, 0 represents no correlation and −1 represents negative correlation. (C) Colocalization of total YAP immunofluorescence and Hoechst. (D) Colocalization of phosphorylated YAP immunofluorescence and Hoechst. Error bars represent standard deviation. (• indicates p<0.05 compared to Day 2, •• indicates p<0.05 compared to 0.1 ×10⁵ cell/cm² seeding density on the same day)

Figure 2

YAP levels, localization and phosphorylation are modulated by hPSC density. hPSCs were singularized and plated at densities of 0.2 and 4.0 × 10⁵ cell/cm². Cells were harvested for western blot analysis after 2, 3 and 4 days of culture. (A) Western blot analysis of total YAP, Histone H3 (nuclear control) and GAPDH (cytoplasmic control) in nuclear and cytoplasmic protein extracts. (B) Representative western blot analysis of total YAP, phospho-YAP (Ser127), and β-actin (housekeeping control) in whole cell lysates. (C) Semi-quantification of the ratio of phosphorylated YAP (Ser127) to total YAP protein by densitometry analysis of western blots. Error bars represent standard deviation. (• indicates p<0.05 compared to 0.2 × 10⁵ cell/cm² seeding density on the same day) (D) Semi-quantification of intensities of total YAP protein to β-actin by densitometry analysis of western blots. Error bars represent standard deviation. (• indicates p<0.05 compared to Day 2, •• indicates p<0.05 compared to Day 3)

Figure 3

YAP transcriptional activity decreased as hPSC density increased. (A) Schematic of YAP transcriptional reporter construct. 4xGTIIC are 4 repeats of TEAD binding sites (ACATTCCA) that drive expression of NanoLuc. (B) hPSCs were singularized and plated at densities of 0.2 to 4.0 × 10⁵ cell/cm². After 3 days in culture, 100,000 cells were harvested for luciferase assays. Chemiluminescent signal was normalized to CellTiter-Glo signal. Error bars represent standard deviation. (• indicates p<0.05 compared to 0.2 ×10⁵ cell/cm² seeding density, •• indicates p<0.05 compared to 1.0 ×10⁵ cell/cm² seeding density) (C) After 3 days, RNA was extracted for end-point RT-PCR analysis of YAP target genes with TATA-box binding protein (TBP) expression serving as the housekeeping control. YAP, TAZ, POU5F1 (Oct4), NANOG and SOX2 expression were also analyzed. (D) H9 4xGTIIC-Nluc cells were singularized and plated at densities of 0.2 to 4.0 × 10⁵ cell/cm². The following day, cells were treated with 1 µM LatA or 2 µM LPA. After 24 hours, cells were harvested for luciferase assays and the chemiluminescent signal was normalized to cell count. Error bars represent standard deviation. (• indicates p<0.05 compared to no treatment condition at the same seeding density)

Figure 4

YAP knockdown decreased YAP levels and transcriptional activity. (A) Schematic of construct for dox-inducible shRNA knockdown of YAP expression. PH1-TetO is the Tet-responsive human H1 promoter. H9 hESC YAP knockdown and scrambled sequence shRNA (ishscram) cell clones were plated at 0.2 ×10⁵ cell/cm² and treated with 2 µM doxycycline (dox) for 3 days and (B) harvested for qPCR analysis of YAP expression (• indicates p<0.05 compared to ishscram control) or (C) harvested for western blot analysis of YAP protein levels. (D) After 3 days of dox treatment, cells were transfected with the YAP responsive 8xGTIIC firefly luciferase plasmid. Following an additional 24 hours of dox treatment, cells were harvested for a luciferase assay. (• indicates p<0.05 compared to no dox condition) (E) Cells were treated with dox for 3 days and harvested for PCR analysis of YAP target genes, YAP, TAZ, POU5F1 (Oct4), NANOG and SOX2.

Figure 5

High seeding density and YAP knockdown increased the conversion rate of hPSCs to PAX6+ neuroepithelial cells. (A) H9 hESCs were plated on Matrigel at densities of 0.5, 1.0, 1.5 and 4.0 × 10⁵ cell/cm² in E8 medium and neuroepithelial differentiation was initiated the next day by changing the medium to E6. Flow cytometry for the percentage of PAX6-positive cells was performed through day 6 of differentiation. (B) Flow cytometry for percentage Pax6-positive cells was performed through day 6 utilizing H9 hESC inducible YAP knockdown (ishYAP) and inducible scrambled sequence shRNA (ishscram) cell lines. Dox was added daily to fresh medium and compared to control cells with no dox added. Error bars represent standard deviation. (• indicates p<0.05 compared to no dox condition on the same-day)

Figure 6

Schematic model of the effects of hPSC density on YAP signaling. At low cell density (left), YAP is primarily localized to the nucleus where it associates with TEAD to regulate gene expression. At high cell density (right), YAP localizes to the cytoplasm where it is phosphorylated and destroyed.