Escape from Pluripotency via Inhibition of TGF-β/BMP and Activation of Wnt Signaling Accelerates Differentiation and Aging in hPSC Progeny Cells

Abstract

Human pluripotent stem cells (hPSCs) represent a potentially valuable cell source for applications in cell replacement therapy, drug development, and disease modeling. For all these uses, it is necessary to develop reproducible and robust protocols for differentiation into desired cell types. However, differentiation protocols remain unstable and inefficient, which makes minimizing the differentiation variance among hPSC lines and obtaining purified terminally differentiated cells extremely time consuming. Here, we report a simple treatment with three small molecules-SB431542, dorsomorphine, and CHIR99021-that enhanced hPSC differentiation into three germ layers with a chemically transitional embryoid-body-like state (CTraS). Induction of CTraS reduced the innate differentiation propensities of hPSCs (even unfavorably differentiated hPSCs) and shifted their differentiation into terminally differentiated cells, particularly neurons. In addition, CTraS induction accelerated in vitro pathological expression concurrently with neural maturation. Thus, CTraS can promote the latent potential of hPSCs for differentiation and potentially expand the utility and applicability of hPSCs.

Keywords: aging; differentiation; disease model; induced pluripotent stem cells; pluripotency; stem cell biotechnology; stem cell differentiation.

Graphical abstract

Figure 1

Evaluation of Small Molecules for Enhancing hPSC Differentiation (A) Schematic representation of experiments for screening combinations of hPSC differentiation enhancers. (B) Representative images of SB-, DM-, and/or CHIR-treated hPSCs. Scale bars, 200 μm. (C) qPCR analysis of the indicated genes in hiPSCs cultured under the indicated conditions for 5 days compared with 36-day EB cultures (n = 3 independent experiments; mean ± SEM; ^∗p < 0.05, ^∗∗p < 0.01; versus untreated; Dunnett’s test). (D) Heatmap summary of the qPCR analysis shown in (C). (E) Schematic of experiments for the time course of the treatment with the three small molecules (SB, DM, and CHIR). (F) qPCR analysis for the indicated genes in hiPSCs cultured with SB, DM, and CHIR for the indicated days (n = 3 independent experiments; mean ± SEM; ^∗p < 0.05, ^∗∗p < 0.01; versus untreated; Dunnett’s test). (G) Heatmap summary of the qPCR analysis shown in (F). hPSC line used: 201B7. See also Figures S1 and S2.

Figure 2

Synergistic Inhibition of the GSK3, TGF-β, and BMP Signaling Pathways Enhanced the Differentiation State of the hPSCs (A) Hierarchical clustering analysis of the global gene expression profiles of untreated PSCs, CTraS PSCs, and EBs. (B) Gene Ontology analysis of transcripts that were up- and downregulated in CTraS PSCs compared with the transcript levels in untreated PSCs (a fold change difference of ±2.0). (C) Top 20 pathways associated with the genes that were differentially expressed in CTraS PSCs compared with those in untreated PSCs (a fold change difference of ±2.0). hPSC lines used, KA11 and eKA3. (D) Immunostaining of single-cell dissociated untreated PSCs and SB + DM + CHIR-treated PSCs for the indicated tridermal lineage markers. Scale bars, 100 μm. (E and F) Cell population and analysis using single-cell dissociated untreated PSCs and CTraS PSCs stained for the indicated markers (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). Intensity of markers (E) and numbers of cells (F) are shown. (G) Immunocytochemistry for the in vitro germ-layer assay and a representative image of EBs derived from untreated and CTraS PSCs. Scale bars, 200 μm. (H) Relative intensity of the indicated tridermal lineage markers in EBs induced from untreated or CTraS PSCs (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). hPSC lines used: 201B7, WD39, and KhES1. See also Figures S2 and S3.

Figure 3

hPSCs Were Rapidly Differentiated toward the Neural Cell Lineage via CTraS Induction (A) Overview of the culture protocol in this experiment. (B–D) Sphere formation analysis of NSs derived from untreated PSCs and CTraS PSCs; SOX1 expression (B), relationship between the number and size (C), and total number (D) of the NSs were analyzed on the indicated day (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Dunnett's test). Scale bar, 400 μm. (E) Representative images of NSs at day 10 derived from CTraS PSCs with antibodies targeting the indicated markers. Scale bar, 200 μm. (F) qPCR analysis of the indicated markers in untreated-PSC- and CTraS-PSC-derived NSs (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Dunnett's test). (G) Fold change in the endoderm and mesoderm gene expression levels of untreated-PSC- and CTraS-PSC-derived NSs (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). (H) Immunocytochemical analysis of NSs for A-P and D-V markers. The frequency of NSs containing immunopositive cells is shown as the percentage of total neurospheres (n = 3 independent experiments, mean ± SEM). Scale bar, 100 μm. hPSC lines used: 201B7, WD39, and KhES1. (I) Comparison of global gene expression profiles of untreated PSCs, CTraS PSCs, untreated NSs, and CTraS NSs. Principal component analysis of the gene expression data. Brown, untreated PSCs; red, CTraS PSCs; blue, untreated NSs; orange, CTraS NSs. (J) GO analysis of transcripts that were up- and downregulated in CTraS NSs compared with the transcript levels in untreated NSs (a fold change difference of ±2.0). (K) Top 20 GO terms associated with the upregulated genes in CTraS NSs compared with the levels in untreated NSs (a fold change difference of ±2.0). Red, developmental process and differentiation; green, neural lineage differentiation. hPSC lines used: KA11 and eKA3. See also Figure S4.

Figure 4

Efficient Generation of Functional Neurons Using Direct Neurosphere Conversion via CTraS (A) Overview of the culture protocol in this experiment. (B) Heatmap results derived from the qPCR analysis depicting the relative gene expression levels of the indicated markers. hNSC, hiPSC (201B7)-derived NSCs as described previously (Nori et al., 2011). (C) Representative images of terminally differentiated derivatives of NSs via CTraS PSCs or not with antibodies targeting the indicated markers. Scale bar, 100 μm. (D) Neural differentiation analysis quantifying the percentage of βIII-TUBULIN⁺MAP2⁺ cells (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). (E) Residual neural stem cell analysis quantifying the percentage of SOX1⁺ cells (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). (F) Cell population analysis of terminally differentiated derivatives of untreated PSCs and CTraS PSCs using dNS-based protocols. Scale bar, 50 μm. (G) SYNAPSIN1 expressions in CTraS and untreated neurons. Scale bar, 100 μm. (H) Neuronal maturation analysis indicated by the number of SYNAPSIN1⁺ puncta within βIII-TUBULIN⁺ neuronal cells at day 23 (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). (I) Representative image of cultured neurons on the 64-electrode array on day 23. Scale bar, 100 μm. (J and K) Electrophysiological analysis of neurons derived from derived from NSs via CTraS-PSC or not using a microelectrode array (MEA) recording system (n = 3 independent experiments; mean ± SEM; ^∗p < 0.05, ^∗∗p < 0.01; Student's t test). hPSC lines used: 201B7, WD39, and KhES1. See also Figure S5.

Figure 5

Neuronal Differentiation of Differentiation-Resistant hESC Lines via CTraS (A) Overview of the culture protocol for CTraS induction using KhESC lines. (B) Fluorescence intensities of the indicated tridermal lineage markers in untreated PSCs and CTraS PSCs (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). (C) Representative image of EBs and immunocytochemistry based on the in vitro three germ-layer assay using KhESC lines. Scale bar, 200 μm. (D) Heatmap results derived from the fluorescence intensity analysis depicting the relative protein expression levels of the indicated markers. The fluorescence intensity levels were normalized to the mean level of each marker in KhES1 cells. (E) Fluorescence intensities of the indicated tridermal lineage markers in differentiated EBs induced from untreated PSCs and CTraS PSCs. (F) Schematic representation of dNS-based neuronal differentiation protocols via CTraS (CTraS-dNS; CdNS) and experimental scheme. (G) SYNAPSIN1 and MAP2 expressions in CTraS and untreated human ES-derived neurons. Scale bar, 200 μm. (H) Sphere formation analysis of NSs derived from KhESC lines as reflected by the quantification of the number and size of the NSs using the indicated methods (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Student's t test). (I) Immunostaining of KhESC-derived neurons with antibodies targeting the indicated markers. Scale bar, 100 μm. (J and K) Neuronal differentiation and maturation analysis as quantified by the percentage of βIII-TUBULIN⁺MAP2⁺ cells (J) and the number of SYNAPSIN1⁺ puncta in βIII-TUBULIN⁺ neuronal cells (K) (n = 3 independent experiments; mean ± SEM; ^∗p < 0.05, ^∗∗p < 0.01; Student's t test). hPSC cells used: KhES1, KhES2, KhES3, KhES4, and KhES5.

Figure 6

Efficient and Rapid Neuronal Differentiation of Newly Established TiPSCs without Colony Selection (A) An overview of the culture protocol in this experiment. (B) Immunostaining with antibodies targeting the indicated markers of neurons derived from 30 newly established TiPSC. Scale bar, 100 μm. (C) Neuronal differentiation analysis, quantification of the number of neuronal differentiated SeV-TiPSC lines, and their βIII-TUBULIN⁺MAP2⁺ cell ratios at day 23 (n = 1). (D) Neuronal maturation analysis of SYNAPSIN1⁺ puncta in βIII-TUBULIN⁺ neuronal cells at day 23 (n = 1). hPSC lines used: SeV-TiPSC (#1–#30, total 30 lines).

Figure 7

CTraS Induction Accelerates Age-Associated Changes and Disease-Specific Phenotypes in a Neurodegenerative Disease Model (A) Overview of the culture protocol in this experiment. (B) TH and βIII-TUBULIN expression in neurons from CdNS-MD, CdNS, and control NS. Scale bar, 100 μm. (C) Neuron and dopaminergic neuron differentiation analysis (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01; Tukey's test). (D) CIII-Core I and βIII-TUBULIN expressions in neurons from CdNS-MD and control NS. Scale bar, 50 μm. (E) Mitophagy analysis of the CCCP/DMSO ratio of the CIII-Core I in βIII-TUBULIN⁺ cells (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01 control versus CdNS-MD; †p < 0.05, ††p < 0.01 healthy donor versus PARK2; Student's t test). (F) Stress vulnerability analysis based on the ratio of TH⁺ neurons after CCCP treatment (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01 control versus CdNS-MD; †p < 0.05, ††p < 0.01 healthy donor versus PARK2; Student's t test). (G) MAP2 immunostaining and CellROX fluorescence in neurons from CdNS-MD and control NS. Scale bars, 500 μm. (H) Oxidative stress analysis of PARK2 neurons and neurons from a healthy donor (n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01 control versus CdNS-MD; ††p < 0.01 healthy donor versus PARK2; Student's t test). (I) MAP2 and pα-syn expressions in neurons from CdNS-MD and control NS. Scale bar, 50 μm. (J) Quantitative analysis of pα-syn accumulation in neurons differentiated from iPSCs from a healthy donor and PARK2 iPSCs (n = 3 independent experiments; mean ± SEM; ††p < 0.01 healthy donor versus PA; Student's t test). (K) Dendrite length analysis of iPSC-derived neurons from a PARK2 patient compared with those from healthy donors (n = 3 independent experiments; mean ± SEM; ^∗∗P < 0.01; Kolmogorov-Smirnov test). Scale bar = 50 μm. (L) Cell viability analysis of iPSC-derived neurons from a PARK2 patient compared with those from healthy donors (n = 3 independent experiments; mean ± SEM; ^∗p < 0.05, ^∗∗p < 0.01 healthy donor versus PARK2; Student's t test). (M) Apoptotic cell population analysis of iPSC-derived neurons from a healthy donor and iPSC-derived neurons from PARK2 patients as gated on their TH intensities. (D–H, L: n = 3 independent experiments; mean ± SEM; ^∗∗p < 0.01 Control vs CdNS-MD; †p < 0.05, ††p < 0.01 healthy donor vs PARK2; Student's t-test). hPSC lines used: KA11, KA23, and eKA3 (healthy donor); PA1, PA9, and PA22 (PARK2-PA); PB2, PB18, and PB20 (PARK2-PB). See also Figures S6 and S7.