Uncoupling of Proliferative Capacity from Developmental Stage During Directed Cardiac Differentiation of Pluripotent Stem Cells

Abstract

Lineage-specific differentiation of human-induced pluripotent stem cells (hiPSCs) into cardiomyocytes (CMs) offers a patient-specific model to dissect development and disease pathogenesis in a dish. However, challenges exist with this model system, such as the relative immaturity of iPSC-derived CMs, which evoke the question of whether this model faithfully recapitulates in vivo cardiac development. As in vivo cardiac developmental stage is intimately linked with the proliferative capacity (or maturation is inversely correlated to proliferative capacity), we sought to understand how proliferation is regulated during hiPSC CM differentiation and how it compares with in vivo mouse cardiac development. Using standard Chemically Defined Media 3 differentiation, gene expression profiles demonstrate that hiPSC-derived cardiomyocytes (hiPSC-CMs) do not progress past the equivalent of embryonic day 14.5 of murine cardiac development. Throughout differentiation, overall DNA synthesis rapidly declines with <5% of hiPSC-CMs actively synthesizing DNA at the end of the differentiation period despite their immaturity. Bivariate cell cycle analysis demonstrated that hiPSC-CMs have a cell cycle profile distinct from their non-cardiac counterparts from the same differentiation, with significantly fewer cells within G1 and a marked accumulation of cells in G2/M than their non-cardiac counterparts throughout differentiation. Pulse-chase analysis demonstrated that non-cardiac cells progressed completely through the cell cycle within a 24-h period, whereas hiPSC-CMs had restricted progression with only a small proportion of cells undergoing cytokinesis with the remainder stalling in late S-phase or G2/M. This cell cycle arrest phenotype is associated with abbreviated expression of cell cycle promoting genes compared with expression throughout murine embryonic cardiac development. In summary, directed differentiation of hiPSCs into CMs uncouples the developmental stage from cell cycle regulation compared with in vivo mouse cardiac development, leading to a premature exit of hiPSC-CMs from the cell cycle despite their relative immaturity.

Keywords: cardiac differentiation; cell cycle; induced pluripotent stem cells; proliferation; stem cell derived cardiomyocytes.

FIG. 1.

Emergence of cardiac lineages during directed hiPSC differentiation. (A) Overview of directed hiPSC cardiac differentiation. (B) Lineage identity was monitored over the course of directed differentiation of human iPSCs by qRT-PCR for loss of pluripotency genes (black) and expression of mesoderm (blue) and cardiac genes (red). (C) Cardiac lineage was quantified over the course of differentiation using flow cytometry analysis of complementary antibodies for cTnT and cTnI (no statistically significant differences were observed between antibodies). Values represent means ± SEM of a minimum of four independent cell lines. hiPSC, human-induced pluripotent stem cells; qRT-PCR, quantitative real-time polymerase chain reaction; SEM, standard error of the mean.

FIG. 2.

Lineage specification during directed hiPSC cardiac differentiation in comparison to mouse cardiogenesis. (A) Pluripotent, mesoderm, cardiac progenitor, and cardiac lineage-specific genes were quantified by RT-PCR throughout hiPSC cardiac differentiation. Values are generated from three hiPSC cell lines in triplicate. The data set was then processed and displayed using the “ClustVis” heatmap function. (B) Data for the murine homologs of the genes examined in A were assembled from [14], processed and displayed using the “ClustVis” heatmap function.

FIG. 3.

EdU incorporation during directed hiPSC cardiac differentiation. Percent of cells that stained positive by flow cytometry for EdU incorporation (yellow squares), cardiac markers (red), or double positive for cardiac markers and EdU (black). Values represent mean ± SEM of a minimum of five independent cell lines. P-values calculated for percent of EdU+ cells (Kruskal–Wallis, non-parametric analysis of variance, and repeat measures corrected by Dunn's tests). EdU, 5-ethynyl-2′-deoxyuridine.

FIG. 4.

Quantitative cell cycle analysis of directed hiPSC cardiac differentiation. Quantification of cell cycle distributions from bivariate analysis of DNA content versus EdU in all live cells (A), cTnT+ cardiac cells (B), and cTnT- non-cardiac cells (C) from hiPSC directed cardiac differentiation. Values represent means of a minimum of five hiPSC lines per time point; statistical significance was determined using a one-way ANOVA and a Holm–Sidak multiple-comparison test. ANOVA, analysis of variance.

FIG. 5.

EdU pulse chase quantitation of cell division during directed hiPSC cardiac differentiation. To examine the progression of EdU+ cTnT+ cardiac (A) and cTNT-non-cardiac cells (B) through the cell cycle, cells on Day 20 of differentiation were pulsed with EdU for 1 h, returned to standard growth media and then cells were collected at 0, 8, and 24 h of recovery. Values represent mean ± SEM, n = 7. *P < 0.01 between cardiac and non-cardiac, early S-Phase cells and F1 cells. ^#P < 0.01 between cardiac and non-cardiac, late S-Phase cells. Statistical significance was determined using a one-way ANOVA and a Holm-Sidak multiple-comparison test. E, early; L, late; M, mid S-Phase.

FIG. 6.

Gene expression of cell cycle regulatory genes during directed hiPSC cardiac differentiation in comparison to mouse development. Positive (A) and negative (B) cell cycle regulatory gene expression. Data were generated from three hiPSC cell lines in triplicate. Data for the murine homologs of the genes examined in hiPSC were assembled from [14]. The data set was then processed and displayed using the “ClustVis” heatmap function.