Examining the protective role of ErbB2 modulation in human-induced pluripotent stem cell-derived cardiomyocytes

Abstract

Human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) are being used as an in vitro model system in cardiac biology and in drug discovery (e.g., cardiotoxicity testing). Qualification of these cells for use in mechanistic investigations will require detailed evaluations of cardiomyocyte signaling pathways and cellular responses. ErbB signaling and the ligand neuregulin play critical roles in survival and functional integrity of cardiac myocytes. As such, we sought to characterize the expression and activity of the ErbB family of receptors. Antibody microarray analysis performed on cell lysates derived from maturing hiPSC-CMs detected expression of ∼570 signaling proteins. EGFR/ErbB1, HER2/ErbB2, and ErbB4, but not ErbB3 receptors, of the epidermal growth factor receptor family were confirmed by Western blot. Activation of ErbB signaling by neuregulin-1β (NRG, a natural ligand for ErbB4) and its modulation by trastuzumab (a monoclonal anti-ErbB2 antibody) and lapatinib (a small molecule ErbB2 tyrosine kinase inhibitor) were evaluated through assessing phosphorylation of AKT and Erk1/2, two major downstream kinases of ErbB signaling, using nanofluidic proteomic immunoassay. Downregulation of ErbB2 expression by siRNA silencing attenuated NRG-induced AKT and Erk1/2 phosphorylation. Activation of ErbB signaling with NRG, or inhibition with trastuzumab, alleviated or aggravated doxorubicin-induced cardiomyocyte damage, respectively, as assessed by a real-time cellular impedance analysis and ATP measurement. Collectively, these results support the expanded use of hiPSC-CMs to examine mechanisms of cardiotoxicity and support the value of using these cells in early assessments of cardiotoxicity or efficacy.

Keywords: ErbB signaling; human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs); nanofluidic proteomic immunoassay (NIA); protein expression; real-time impedance-based cell assay (RTCA) cardio system.

FIG. 1.

HiPSC-CMs in culture for 2 or 14 days after plating. (A) Microscopic images before fixation (top) and after, stained (bottom) for cTnI (red), myomesin (green), and nuclei (blue), demonstrate hypertrophic growth of hiPSC-CMs in culture. Bar represents 60μM. (B) Frequency distribution of cell surface area based on cTnI-stained cells (n > 6000 cells each from Day 2 and Day 14 cultures) shows an increased cell population with larger cell size in Day 14 cultures. (C) Comparison of cell surface area and the ratio of cell/nuclear area of cTnI-stained cells from Day 2 or Day 14 cultures. Each bar represents the mean ± SE of 20 wells. *p < 0.05 versus Day 2.

FIG. 2.

Enhancement of contractile properties in cultured hiPSC-CMs. (A) Screenshots of spontaneous beating cells taken from the same four wells on Day 2 and Day 14. Horizontal scale bar represents 3 s; vertical scale bar represents 0.06 arbitrary units (AU). (B) Comparison of beat amplitude and (C) myomesin staining intensity in cells from Day 2 or Day 14 cultures. Each bar represents the mean ± SE of 32 (B) or 20 (C) wells. *p < 0.05 versus Day 2.

FIG. 3.

Transmission electron microscopy (TEM) images of hiPSC-CMs demonstrate maturation to a more functional cardiac myocyte phenotype. (A) Low magnification (×1000) TEM images illustrate the presence of abundant mitochondria in hiPSC-CMs cultured for 2 or 14 days; well-defined sarcomeres are sparse in Day 2 cell cultures. (B) High magnification (×5000) TEM images show the disorganized myofilaments, the lack of Z-like dense structures and aligned sarcomeres in Day 2 hi-PSC-CMs, whereas the well-developed sarcomeres are frequently observed in Day 14 hi-PSC-CMs. (C) Underconstruction and well-established sarcomeres of hiPSC-CMs in Day 2 and Day 14 cultures, respectively. H: H-zone; M: mitochondria; MF: myofilaments; N: nucleus; S: sarcomeres; Z: Z-bands. (D) The area of mitochondria and sarcomeres relative to the total cell area was quantified using Definiens analysis software in low magnification (×1000) images randomly taken from Day 2 or Day 14 cultures of hiPSC-CMs. Bar graph depicts equal amounts of mitochondria in both Day 2 and Day 14 cultures, but significantly greater sarcomeres in Day 14 cultures. p < 0.05 versus Day 2; each data point represents mean ± SEM of 20 images per time point.

FIG. 4.

List of 50 selected protein kinases involved in the growth factor-activated or stress-activated pathways. Each data point represents the mean signal intensity in Day 14 hiPSC-CM cell lysates from three independent experiments using microarray protein analysis as described in the Materials and Methods section.

FIG. 5.

Expression of ErbB1, ErbB2, and ErbB4 receptors in hiPSC-CMs. Four-hundred microgram of total protein in 450 μl was loaded in the gel and each antibody marker was probed in duplicate lanes. GAPDH served as the loading control. (A) Representative Western blot. (B) Relative density of the bands normalized to GAPDH. Each data point represents the mean ± SE of three independent experiments.

FIG. 6.

Activation of downstream ErbB signaling proteins in hiPSC-CMs. Cells treated with 20-ng/ml neuregulin-1β (NRG) for either 10 or 30 min; untreated controls did not receive fresh media, whereas the time-matched controls (at 10 or 30 min) as well as the neuregulin-treated groups did receive fresh media at the time of treatment. Representative Western blot for AKT (A) and Erk1/2 (B); 10 μg total protein was loaded into each individual well of a 10-well mini-gel (50 μl per well). Blots were first probed for phosphorylated AKT and Erk1/2, then stripped and re-probed for total AKT, total Erk1/2, and GAPDH. Relative density of the AKT bands normalized to GAPDH (C). Relative density of the Erk1/2 bands normalized to GAPDH (D). Each data point represents the mean ± SEM of three independent experiments. *^#Statistically significant at p < 0.05: *pAKT 30-min control significantly less than untreated control; ^#pAKT 10-min and 30-min NRG treatment significantly greater than time-matched controls; *pErk1/2 10-min control significantly greater than untreated control; ^#pErk1/2 30-min NRG significantly greater than 30-min control.

FIG. 7.

Activation of multiple phosphorylation isoforms of AKT and Erk1/2 by neuregulin-1β (NRG). HiPSC-CMs were treated (or untreated) with fresh medium (time-matched vehicle control) or NRG 20 ng/ml (diluted in the fresh medium) for 10 or 30 min before lysed as described in the Materials and Methods section. Lysates were analyzed by the nanofluidic proteomic immunoassay (NIA) using antibodies raised against either the total protein or specific phosphorylated-site of AKT and Erk1/2. (A) Representative traces of nanofluidic proteomic immunoassay (NIA) analysis of 30-min treatment samples using anti-AKT2 or anti-pAKT(Ser473) to detect the non-phosphorylated or phosphorylated AKT. The vertical dashed line separates peaks of phosphorylated (pI < 5.40) or non-phosphorylated (pI > 5.40) ATK. Note the leftward shift and dramatic increases in phosphorylated peaks after NRG treatment. (B) Representative traces of NIA analysis of 30-min treatment samples using anti-Erk1/2 or anti-pErk/12(Thr202/Tyr204) to detect the non-phosphorylated or phosphorylated Erk1/2. Isoelectric peaks corresponding to the non-phosphorylated Erk1/2 and their mono- or dual-phosphorylated isoforms (pErk1/2 or ppErk1/2) are marked with vertical dashed lines. NRG treatment reduces the non-phosphorylated peak intensity of Erk1 and Erk2 and increases the peak intensity of phosphorylated isoforms, especially ppErk1 and ppErk2. The level of phosphorylation for AKT or Erk1/2 is quantified by either the phospho-/non-phosphorylated peak intensity ratio (C), or by the peak intensity of all phosphorylated isoform peaks (D). Each data represents the mean ± SE of three independent experiments. *p < 0.05 versus the time-matched control group. Isoelectric point (pI) is shown on the x-axis and chemiluminescence on the y-axis in (A) and (B).

FIG. 8.

Attenuation of neuregulin-1β (NRG)-activated phosphorylation of AKT and Erk1/2 by lapatinib (LPN) and trastuzumab (TZM). HiPSC-CMs were pretreated with lapatinib (1μM) or trastuzumab (1μM) for 30 min before changing to fresh media (vehicle control) or NRG 100 ng/ml (by adding 10X stock prepared in fresh media) for 30 min. Lysates were prepared and analyzed as described in the Materials and Methods section. (A) and (B) show representative traces of NIA analysis to illustrate the change of phosphorylated and non-phosphorylated peaks by NRG alone, or subsequently with LPN and TZM. (C) and (D) show quantification of AKT or Erk1/2 phosphorylation by the phospho-/non-phosphorylated peak intensity ratio, or by the peak intensity of all phosphorylated isoform peaks, respectively. Each bar represents the mean ± SE of three independent cell culture experiments. *p < 0.05 versus NRG treatment group. Isoelectric point (pI) is shown on the x-axis and chemiluminescence on the y-axis in (A) and (B).

FIG. 9.

ErbB2 expression knockdown by targeted siRNAs attenuates neuregulin-1β (NRG)-activated AKT and Erk1/2 phosphorylation. HiPSC-CMs were incubated with Lipofectamine/Opti-MEM (transfection vehicle), 100-nM scrambled or targeted siRNAs [control (ctrl) or ErB2 siRNAs] for up to 144 h. To test the effect of ErbB2 siRNA on NRG-activated AKT or Erk1/2 phosphorylation, cells at 72-h transfection were exposed to 100-ng/ml NRG for 30 min prior to lysing in RIPA buffer. (A) and (B) Representative Western blots of three independent experiments show diminished ErB2 expression and reduced phosphorylation of AKT and Erk1/2 in response to 100-ng/ml NRG treatment. Thirty microgram total protein was loaded into each individual well of a 10-well mini-gel (30 μl/well) (A); 10 μg total protein was loaded into each individual well of a 10-well mini-gel (30 μl/well) (B). Blots were first probed for phosphorylated AKT(Ser473) and Erk1/2(Thr202/Tyr204), then stripped and re-probed for total AKT and Erk1/2, and GAPDH. (C) Quantification of ErbB2 expression after 72- or 144-h transfection. ErbB2 expression relative to the protein loading (GAPDH) was normalized to the transfection vehicle group. (D) and (E) Quantification of NRG-induced AKT and Erk1/2 phosphorylation at 72-h transfection. Proteins detected by anti-pAKT(Ser473) or anti-pErk/12(Thr202/Tyr204) were expressed as a fraction of that detected by anti-total AKT or Erk1/2 antibodies. Each data point represents the mean ± SE of three independent experiments. *p < 0.05 compared with control siRNAs, ^‡p < 0.05 compared with 72-h ErbB2 siRNAs, ^#p < 0.05 compared with vehicle control exposed to NRG.

FIG. 10.

Neuregulin-1β (NRG) attenuates and trastuzumab (TZM) potentiates doxorubicin (Dox)-induced cardiomyocyte injury. (A) Representative screenshots of spontaneous beating cells taken from wells before or after exposure to Dox (1μM) or vehicle (0.1% DMSO) following pretreatment with NRG (100 ng/ml) or TZM (1μM) for 24 h. Horizontal scale bar: 3 s; vertical scale bar: 0.05 arbitrary units. (B) Real-time monitoring of cellular impedance for cells exposed to 0.3 or 1-μM Dox after pre-treatment with NRG (100 ng/ml) or TZM (1μM) for 24 h. Readings for each well were normalized to the baseline level prior to Dox application (marked by the arrow). For clarity, not all treatment groups are shown. Cellular impedance and intracellular ATP measured after 40-h exposure to Dox are plotted in (C) and (D). Data point for each concentration is the average of 9–12 wells from four to five separate experiments. RLU: relative light units; *p < 0.05 compared with Dox; ^#p < 0.05 compared with NRG 1μM + Dox.