Activation of GATA4 gene expression at the early stage of cardiac specification

Abstract

Currently, there are no effective treatments to directly repair damaged heart tissue after cardiac injury since existing therapies focus on rescuing or preserving reversibly damaged tissue. Cell-based therapies using cardiomyocytes generated from stem cells present a promising therapeutic approach to directly replace damaged myocardium with new healthy tissue. However, the molecular mechanisms underlying the commitment of stem cells into cardiomyocytes are not fully understood and will be critical to guide this new technology into the clinic. Since GATA4 is a critical regulator of cardiac differentiation, we examined the molecular basis underlying the early activation of GATA4 gene expression during cardiac differentiation of pluripotent stem cells. Our studies demonstrate the direct involvement of histone acetylation and transcriptional coactivator p300 in the regulation of GATA4 gene expression. More importantly, we show that histone acetyltransferase (HAT) activity is important for GATA4 gene expression with the use of curcumin, a HAT inhibitor. In addition, the widely used histone deacetylase inhibitor valproic acid enhances both histone acetylation and cardiac specification.

Keywords: GATA4; cardiac differentiation; cardiomyocytes; gene regulation; histone acetylation.

Figure 1

Valproic acid-enhanced cardiac differentiation. (A) P19 stem cells were treated with DMSO or increasing concentrations of valproic acid (VPA 0.5, 1, 2 mM) during EB formation, maintained in the tissue culture dishes for 3 additional days without treatments, and then stained for myosin heavy chain and cTnT. Quantification is presented as fractions of cells differentiated into cardiomyocytes relative to the total cell populations. Error bars represent the standard deviations of three independent experiments. (B) Shown are the representative microscopy images of the cells stained for cTnT (green). Hoechst was used to stain the DNA (blue) concomitantly (scale bars = 50 μm). (C) Western analysis of GATA4 protein expression and the levels of global H3 acetylation. The blots were then stripped and reprobed for β-tubulin as loading controls. Undifferentiated cells were included as a negative control. Shown are the cropped blot images representing indicated protein. (D) Occupancy of p300 at the GATA4 promoter (GATApro) and a control locus (GATActl) were examined by a real-time PCR based ChIP analysis. Quantification is presented as the fold variations of undifferentiated control.

Figure 2

Effects of curcumin on the differentiation of ES cells into cardiomyocytes. (A) ES cells were grown in hanging drops for 2 days and in suspension for 4 additional days to form the EBs. Treatment with curcumin (10 μM) was for days 0–2 (E), days 2–4 (M), or days 4–6 (L). The cells were then maintained on cover slips for another 4 days to allow for the development of cardiomyocytes. Quantification is presented as the percentage of cells positive for cTnT and myosin heavy chain (MyHC), respectively. Error bars represent the standard deviations of three independent experiments (^*p < 0.05). (B) Shown are the representative images of untreated cells stained for cTnT or myosin heavy chain (green) on day 10 of differentiation. Hoechst was used to stain the DNA (blue) concomitantly (scale bars = 50 μm). (C) Western analysis of GATA4 protein expression and the levels of global H3 acetylation. The blots were then stripped and reprobed for β-tubulin as loading controls. Undifferentiated ES cells and were used as the negative control. Shown are the cropped blot images representing indicated protein.

Figure 3

Effects of curcumin on the differentiation of P19 cells into cardiomyocytes. (A) P19 stem cells were treated with DMSO during EB formation. The addition of curcumin (10 μM) was during the early stage (days 0–2) or the late stage (days 2–4) of EB formation. The cells were then maintained in the tissue culture dishes for 3 additional days without any treatments, and stained for myosin heavy chain or cTnT (green), and with Hoeschst for the nuclei (blue) (scale bars = 50 μm). (B) Quantification of the myosin heavy chain positive cells is expressed as fractions of cardiomyocytes relative to the total cell populations. Error bars are the standard deviations of five independent experiments (^**p < 0.001). (C) Cells were treated with curcumin during EB formation and quantified as the percentage of cardiomyocytes relative to the DMSO control which is defined as 100%. Error bars are the standard deviations of three independent experiments. (D,E) The administration of curcumin was during the early stage (days 0–2) or the late stage (days 2–4) of EB formation. Quantification of the percentage of the myosin heavy chain (MyHC) or cTnT positive cells is presented in relation to the DMSO-alone control which is defined as 100%. Error bars are the standard deviations of three independent experiments. (F) The mRNA levels of GATA4, Tbx5, MEF2C, and Nkx2.5 were analyzed by real-time RT-PCR using the same batch of cDNA, with GAPDH as internal controls. Quantification is plotted as fold variations of the undifferentiated controls. Error bars represent the standard deviations of the triplicates from one representative experiment.

Figure 4

Effects of curcumin on cardiac protein expression. (A) P19 stem cells were treated with DMSO and the addition of curcumin (10 μM) was during the early (days 0–2) or the late stage (days 2–4) of EB formation. Western analysis was used to examine the levels of GATA4, MEF2C, and cTnT protein, and global H3 acetylation. The blots were then stripped and reprobed for β-tubulin as loading controls. Undifferentiated cells were used as the negative control. Shown are the cropped blot images representing indicated protein. (B) Quantification of the Western blots is expressed as fold variations in relation to the undifferentiated controls (mean ±SD, n = 3). (C) Quantification of cTnT blots is plotted as percentages of the DMSO control (mean ±SD, n = 3, ^**p < 0.005 relative to the DMSO control).

Figure 5

Occupancy of p300 at the GATA4 promoter at early stage of differentiation. (A) P19 cells were differentiated with DMSO and co-treatment of curcumin (10 μM) was during the first 2 days of EB formation. The cellular levels of H3 acetylation and p300 protein were analyzed by Western blotting on day 4. The blots were then stripped and reprobed for β-tubulin as loading controls. Undifferentiated cells were used as the negative control. Shown are the cropped blot images representing indicated protein. (B) Quantification of acetylated H3 blots is presented as fold variations of the undifferentiated control (mean ± SD, n = 3). (C) The levels of acetylated H3 at the GATA4 promoter were determined by the ChIP analysis. Quantification is presented as fold variations of the undifferentiated control. (D) Occupancy of p300 at the GATA4 promoter was examined in parallel. (E) Quantification of the p300 Western blots is presented as fold variations of the undifferentiated controls (mean ± SD, n = 3).

Figure 6

Shown is the schematic presentation of sequential activation of cardiac factors during the differentiation of stem cells into cardiomyocytes. Pathway involved in linage specification step is denoted with open gray arrows, whereas pathways involved in the differentiation step are denoted with open green arrows.