AMPK activator-treated human cardiac spheres enhance maturation and enable pathological modeling

Abstract

Background: Cardiac pathological outcome of metabolic remodeling is difficult to model using cardiomyocytes derived from human-induced pluripotent stem cells (hiPSC-CMs) due to low metabolic maturation.

Methods: hiPSC-CM spheres were treated with AMP-activated protein kinase (AMPK) activators and examined for hiPSC-CM maturation features, molecular changes and the response to pathological stimuli.

Results: Treatment of hiPSC-CMs with AMPK activators increased ATP content, mitochondrial membrane potential and content, mitochondrial DNA, mitochondrial function and fatty acid uptake, indicating increased metabolic maturation. Conversely, the knockdown of AMPK inhibited mitochondrial maturation of hiPSC-CMs. In addition, AMPK activator-treated hiPSC-CMs had improved structural development and functional features-including enhanced Ca²⁺ transient kinetics and increased contraction. Transcriptomic, proteomic and metabolomic profiling identified differential levels of expression of genes, proteins and metabolites associated with a molecular signature of mature cardiomyocytes in AMPK activator-treated hiPSC-CMs. In response to pathological stimuli, AMPK activator-treated hiPSC-CMs had increased glycolysis, and other pathological outcomes compared to untreated cells.

Conclusion: AMPK activator-treated cardiac spheres could serve as a valuable model to gain novel insights into cardiac diseases.

Keywords: AMPK; Cardiomyocytes; Maturation; Metabolic regulation; Pathological modeling; Stem cells.

Fig. 1

AMPK activators improve metabolic maturation of 3D hiPSC-CMs. A Experimental design and cell morphology. Cardiac spheres were generated on day 5 and treated with AMPK activators on day 14 for 7 days before the assessments of cardiomyocyte maturation. B Flow cytometry analysis of cTnT and α-actinin on day 14 (n = 3 cultures). C High-content imaging analysis of NKX2-5 and α-actinin by ArrayScan on day 14 (n = 4 cultures). D High-content imaging analysis of NKX2-5 and α-actinin by ArrayScan on day 21 (n = 4 cultures). E ATP content (n = 5 cultures). F Mitochondrial membrane potentials analyzed by ArrayScan of TMRM (n = 8 cultures). G Flow cytometry analysis of MitoTracker Red (n = 3 cultures). H High-content imaging analysis of TOM20 by ArrayScan (n = 8 cultures). I Flow cytometry analysis of TOM20 (n = 3 cultures). J Ratios of mtDNA to nDNA (n = 3 cultures). K Measurement of OCR and quantification of mitochondrial functional parameters including basal respiration, maximal respiration, spare respiratory capacity, non-mitochondrial respiration and ATP production (n = 4 cultures). L Fatty acid uptake (n = 4). M Glucose concentration (n = 4 cultures). N Measurement of OCR and quantification of fatty acid oxidation (the amount of OCR derived from fatty acid oxidation). O Measurement of ECAR and quantification of glycolysis (n = 4 cultures). Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 and ****P < 0.0001 (one-way ANOVA). 2-DG, 2-Deoxy-d-glucose; A100, A-769662 at 100 µM; A200, A-769662 at 200 µM; AA, Activin A, BMP4, Bone morphogenetic protein 4; cTnT, Cardiac troponin T; DMSO, Dimethyl sulfoxide; E10, EX229 at 10 µM; E50, EX229 at 50 µM; ECAR, Extracellular acidification rate; ETO, Etomoxir; FCCP, Carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone; LPL, Lipoprotein lipase; MFI, Mean fluorescence intensity; mt-CO2, Mitochondrially encoded cytochrome c oxidase II; mtDNA, Mitochondrial DNA; ND1, Mitochondrially encoded NADH dehydrogenase 1; nDNA, Nuclear DNA; NKX2-5, NK2 homeobox 5; OCR, Oxygen consumption rate; Oligo, Oligomycin; Rot/Ant, Rotenone/antimycin A; SDHA, Succinate dehydrogenase complex flavoprotein subunit A and TMRM, Tetramethylrhodamine, methyl ester

Fig. 2

AMPK inhibition using Compound C inhibits the metabolic maturation of hiPSC-CMs. Measurements of relative A ATP content, B TMRM and C TOM20. D, E Measurement of OCR and quantification of basal respiration, maximal respiration, ATP production, spare respiratory capacity and non-mitochondrial respiration. The effect of EX229 (D) and A-769662 (E) on mitochondrial function was abolished with Compound C. All measurements were normalized to cell counts (n = 4 cultures). Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 and ****P < 0.0001 (one-way ANOVA). A100, A-769662 at 100 µM; A200, A-769662 at 200 µM; DMSO, Dimethyl sulfoxide; E10, EX229 at 10 µM; E50, EX229 at 50 µM; FCCP, Carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone; OCR, Oxygen consumption rate; Oligo, Oligomycin; Rot/Ant, Rotenone/antimycin A and TMRM, Tetramethylrhodamine, methyl ester

Fig. 3

siRNA knockdown of AMPK inhibits the metabolic maturation of hiPSC-CMs. A Relative mRNA levels of PRKAA1 and PRKAA2 in hiPSC-CMs transfected with siRNA at doses 50, 100 and 200 μM compared with control siRNA after 72 h (n = 3 cultures). B Relative mRNA levels of PRKAA1 and PRKAA2 in hiPSC-CMs transfected with siRNA at 50 μM compared with control siRNA after 7 days (n = 3 cultures). Measurements of relative C ATP content (n = 5), D TMRM (n = 6 cultures), E TOM20 (n = 4 cultures) and F fatty acid uptake capacity (n = 3 cultures) of AMPK siRNA transfected hiPSC-CMs treated with DMSO or E10 compared with control siRNA transfected hiPSC-CMs treated with DMSO or E10. (H) Relative ratio of mtDNA and nDNA. G Mitochondrial function measured in DMSO- or E10-treated hiPSC-CMs transfected with control or AMPK siRNA. All measurements were normalized to cell counts (n = 4 or 5 cultures). Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 and ****P < 0.0001 (one-way ANOVA). DMSO, Dimethyl sulfoxide; E10, EX229 at 10 µM; FCCP, Carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone; LPL, Lipoprotein lipase; mt-CO2, Mitochondrially encoded cytochrome c oxidase II; mtDNA, Mitochondrial DNA; ND1, Mitochondrially encoded NADH dehydrogenase 1; nDNA, Nuclear DNA; OCR, Oxygen consumption rate; Oligo, Oligomycin; PRKAA1, Protein kinase AMP-activated catalytic subunit alpha 1; PRKAA2, Protein kinase AMP-activated catalytic subunit alpha 2; Rot/Ant, Rotenone/antimycin A; SDHA, Succinate dehydrogenase complex flavoprotein subunit A; siRNA, Small interfering RNA and TMRM, Tetramethylrhodamine, methyl ester

Fig. 4

AMPK activation improves structural and functional maturation of hiPSC-CMs. A Structural analysis of sarcomere length, cell size, perimeter and length to width ratio. n = 105–112 cells. B Representative Ca²⁺ transient traces from DMSO- and AMPK activator-treated hiPSC-CMs and quantification of Ca²⁺ transient characteristics (n = 52–70 cells). C Representative action potential traces and quantification of dF/dT, time to peak, normalized amplitude, AP50 and AP80 (n = 55–95 cells). D Representative image and heatmap (left) depicting time-averaged magnitude of all motion and tracing (middle) of average beating speed followed by the analysis of contraction velocity and relaxation velocity of 3D hiPSC-CMs (n = 19–28 spheres). Beating activities were recorded at 0.5 Hz. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 and ****P < 0.0001 (one-way ANOVA). 3D, Three-dimensional; A100, A-769662 at 100 µM; A200, A-769662 at 200 µM; AP50, Action potential duration measured at 50%; AP80, Action potential duration measured at 80%; DMSO, Dimethyl sulfoxide; E10, EX229 at 10 µM and E50, EX229 at 50 µM

Fig. 5

AMPK activation alters the gene expression profile in hiPSC-CMs identified by RNA-seq. Volcano plots portray log₂ (fold change) vs. negative log₁₀(adjusted P value) for differentially expressed genes (DEGs) in A E10-treated hiPSC-CMs and B tissues from human left ventricle (LV) compared with DMSO control (n = 3 cultures or tissue samples). C Heatmap showing common DEGs in E10-treated hiPSC-CMs and LV compared with DMSO-treated hiPSC-CMs. D Venn diagram showing the number of upregulated and downregulated DEGs in E10-treated hiPSC-CMs versus DMSO-treated hiPSC-CMs and LV versus DMSO-treated hiPSC-CMs. E Upregulated and downregulated GO terms in E10-treated hiPSC-CMs and LV compared with DMSO-treated hiPSC-CMs. Heatmaps of DEGs in E10-treated hiPSC-CMs vs. DMSO-treated hiPSC-CMs associated with F mitochondrial function, G fatty acid metabolism and H cardiac muscle contraction. E10, EX229 at 10 µM; LV, Heart tissue samples from pediatric left ventricle and RNA-seq, RNA-sequencing

Fig. 6

AMPK activation alters the expression of proteins in hiPSC-CMs identified by proteomics analysis. A Volcano plot illustrating proteins with statistically significant abundance differences between E10-treated and DMSO-treated hiPSC-CMs (n = 3 cultures). B Upregulated GO terms of biological processes. n refers to the number of proteins found in each GO term. C Upregulated pathways based on KEGG analysis and upregulated GO terms of cellular compartment (CC) and molecular function (MF). n refers to the number of proteins found in each pathway/GO term. D Protein interaction network of the upregulated proteins that participate in the oxidative phosphorylation pathway. Heatmaps showing the key upregulated proteins in E10-treated hiPSC-CMs versus DMSO-treated hiPSC-CMs associated with E mitochondrial function and F fatty acid metabolism. G Venn diagram showing the number of overlapping upregulated and downregulated genes (DEGs) and proteins (DEPs) in E10-treated hiPSC-CMs versus DMSO-treated hiPSC-CMs. DMSO, Dimethyl sulfoxide and E10, EX229 at 10 µM

Fig. 7

Metabolomics analysis identifies enhanced energy-related metabolic pathways in E10-treated hiPSC-CMs. A Three-dimensional principal component analysis (PCA) score plot based on metabolite abundance measured as ion peak intensities showing a significant separation between E10-treated and DMSO-treated hiPSC-CMs (n = 6 cultures). B Volcano plot illustrating discriminatory metabolites between E10-treated and DMSO-treated hiPSC-CMs. C Hierarchical clustering using the top 600 discriminatory features. D Enriched metabolic pathways using the 84 discriminatory metabolites. Significantly affected pathways were identified based on negative Log₁₀ (P value < 0.05) and comprising of ≥ 3 metabolites per pathway. Comparison of E intermediates in glycolysis pathway, F lipoamide in TCA cycle, intermediates in carnitine shuttle including, G long-chain and H medium- or short-chain acyl-carnitines between E10-treated and DMSO-treated hiPSC-CMs. Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 and ****P < 0.0001 (two-sided unpaired t-test)

Fig. 8

E10-treated hiPSC-CMs enhance pathological modeling. E10-treated and immature (without E10 treatment) hiPSC-CMs were subjected to pathological stimulation (H + ISO) and analyzed for pathological outcomes. A Expression of ADRB1 (n = 3 cultures). B Representative kinetics of ECAR and quantification of glycolysis, glycolytic capacity and glycolytic reserve (n = 4 cultures). C Expression of genes associated with glycolysis and D biosynthesis of triacylglycerol CD36, PLIN2 and HSL (n = 3 cultures). E Quantitative analysis of Nile red staining by ArrayScan (n = 5 cultures). F Relative ATP content (n = 7 cultures), G relative cell viability (n = 5 cultures), H cell count (n = 3 cultures) and I caspase 3/7 activity (n = 6 cultures). Data are presented as mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 and ****P < 0.0001 (two-sided unpaired t-test). 2-DG, 2-Deoxy-d-glucose; CD36, cluster of differentiation 36; E10, EX229 at 10 µM; ECAR, Extracellular acidification rate; GAPDH, Glyceraldehyde-3-phosphate dehydrogenase; H, Hypoxia; ISO, Isoproterenol; HSL, Lipase E, hormone-sensitive type; LDHA, Lactate dehydrogenase A; Oligo, Oligomycin; PKM2, Pyruvate kinase M1/2; PLIN2, Perilipin 2 and SLC2A1, Solute carrier family 2 member 1