Doxorubicin induces cardiotoxicity in a pluripotent stem cell model of aggressive B cell lymphoma cancer patients

Fig. 1. Analysis of human myocardium of ACT patients and isolated ACT cardiac fibroblasts. A Representative myocardium of non-failing donor (NF) and ACT patients (ACT) after trichrome staining. Scale bar: 100 µm. B Quantification of trichrome staining; NF: n = 5; ACT: n = 5 (biological replicates). 4 technical replicates from different tissue slices for each biological replicate were used. C mRNA expression of fibrosis-associated genes (CTGF, MMP9) in human ACT myocardium compared to NF by qPCR. Statistical analysis was performed using Student’s t test. p values: P > 0.05 are defined as statistically significant. Bars indicate mean values ± SEM. NF: 4–5; HF: 5. D mRNA and protein expression of SERCA2a. Representative Western Blots for SERCA2a in NF and ACT myocardium are shown. NF: n = 4; HF: n = 3. GAPDH was used as an internal control. Statistical analysis was performed using Student’s t test. p values: p > 0.05 are defined as statistically significant. Bars indicate mean values ± SEM. E NF cardiac fibroblasts (cFB-NF), cardiac fibroblasts from the ACT patient (cFB-ACT), and skin FB of ACT patient (sFB-ACT) show typical FB morphology. Scale bar: 100 µm. F Proliferation capacity over…

Fig. 2. Assessment of in vitro DOX-induced cardiotoxicity in patient-specific iPSC CM. A, B DOX resorption and retention in iPSC CM. Intracellular DOX levels were investigated with HPLC in regard to DOX concentration (24 h treatment) (A), treatment time (1 µM DOX) (B), ACT patient: n = 5 differentiations from 2 patients. Control: n = 3 differentiations from 2 controls. C–E Effect of DOX on sarcomeric regularity in iPSC CM. C Immunofluorescence staining visualized α-actinin protein expression and sarcomeric organization. DOX treatment impaired sarcomeric regularity. Scale bars: 50 μm. D, E Quantification of DOX-treated (24 h) sarcomeric regularity using Fast Fourier Transform algorithm. Sample numbers: 60 control iPSC CM from 4 differentiations, 90 patient iPSC CM from 6 differentiations. F, G Effect of DOX on iPSC CM cell death. Annexin V/PI apoptosis tests showed that the number of apoptotic and dead cells rose with increasing DOX levels after 24 h (F) and 72 h (G). 10 control CM differentiations, 10–11 patient CM differentiations. Statistical analysis was performed using 2-way ANOVA analysis and Sidak’s or Dunnett’s multiple comparison, or mixed-effects analysis. Bars indicate mean…

Fig. 3. DOX-dependent mitochondrial disturbance. A MitoSpy staining of the mitochondrial structure under DOX (0.5 µM) treatment after 24 and 72 h in ACT-iPSC CM. The mitochondrial shape is damaged under DOX that is also demonstrated on the skeletonized stage (Mitochondrial Network Analysis software tool). Scale bars: 50 µm. B Example of double staining of MitoSpy (red) and cardiac marker α-actinin (green). Scale bar: 50 µm. C, D Quantification of mitochondrial branch length (C) and network (D). The branch length and network decrease in ACT patients under DOX treatment for 24 h (0.25, 0.5 and 1 µM) and 72 h (0.5 µM). Sample numbers: 18 control-iPSC CM from 2 differentiations, 14–26 patient iPSC CM from 3 differentiations. Different colored dots indicate the different control lines (green: control 1) and patient lines (black: ACT2 and red: ACT3) used. Statistical analysis was performed using 2-way ANOVA analysis with Tukey’s multiple comparison. Bars indicate mean values with SEM. *p < 0.05, **p < 0.01. E Representative transmission electron micrographs of ACT-iPSC CM under DOX treatment for 72 h (0.5 µM) and untreated showing interfibrillar mitochondria (red arrows) and striated and…

Fig. 4. DOX-dependent ROS production in iPSC CM and cFB. A The amount of H2O2 in the supernatant of iPSC CM was measured with the Amplex Red assay after 24 h DOX treatment. ACT patient: n = 20 differentiations from 3 patients. Control: n = 12 differentiations from 2 controls. Different colored dots indicate the different control lines (green: control 1, grey: control 2) and patient lines (blue: ACT1, black: ACT2 and red: ACT3) used. B H2O2 in the supernatant of ACT patient and control cFB, and patient skin FB (patient sFB) was measured with the Amplex Red assay after 24 h DOX treatment. n = 3 biological replicates (samples of 3 different passages were measured of control cFB, patient cFB, and patient sFB. Statistical analysis was performed using 2-way ANOVA analysis and Sidak’s or Dunnett’s multiple comparison test. Data are shown as mean with SEM. *p < 0.05, **p < 0.01, ***p < 0.001. *Above single bars indicate statistical significances to untreated (0 µM DOX) conditions of the same group

Fig. 5. DOX decreases the force of contraction in engineered human myocardium (EHM). A Beating frequency of control and ACT patient EHM increases after DOX treatment (0.25 µM). B The generation of the contractile force of control and ACT patient EHM is Ca2+-dependent, and impeded by DOX treatment (0.25 µM). C Maximal force of contraction of control and ACT patient EHM is lowered after incubation with 0.25 µM DOX. D The DOX-induced relative change of maximal contractile force is greater in ACT patient EHM compared to control EHM. E Maximal force of contraction of control EHM and ACT patient EHM normalized to cross-sectional area (CSA). F DOX-induced change in the maximal force of control and ACT patient EHM normalized to CSA. A–F Sample numbers: 12 control EHM from 2 differentiations, 17 ACT patient-EHM from 3 differentiations. Different colored dots indicate the different control lines (green: control 1, grey: control 2) and patient lines (blue: ACT1, black: ACT2 and red: ACT3) used. Statistical analysis was performed using Student’s t test or 2-way ANOVA analysis and Sidak’s multiple comparison test. Mean with SEM. *p < 0.05, **p < 0.01, ***p < 0.001. *Above single bars indicate…

Fig. 6. DOX disturbs Ca2+ homeostasis. A, B Effects of DOX (0.25 µM and 5 µM) on Ca2+ transient amplitude (A) and Ca2+ transient decay time (B) in iPSC CM of control and ACT patients. C–G DOX-induced relative changes in the amount of protein expression and phosphorylation of SERCA2a (C), NCX (D), CaMKIIδ (E), RYR2, RYR2-S2808, RYR2-S2814 (F), PLN, PLN-S16p, PLN-T17p (G) in iPSC CM of control and ACT patients after 24 h of DOX treatment. Different colored dots indicate the different control lines (green: control 1, grey: control 2) and patient lines (blue: ACT1, black: ACT2 and red: ACT3) used. H Representative Western blots of Ca2+-associated proteins in DOX-treated iPSC CM of control and ACT patients. Sample numbers: A, B 180 control-iPSC CM from 9 differentiations from 2 controls, 180 ACT patient-iPSC CM from 9 differentiations from 3 patients. C–G 4 control-iPSC CM differentiations, 6 ACT patient iPSC CM differentiations. Statistical analysis was performed using 2-way ANOVA with Dunnett’s multiple comparison test. Mean with SEM. *p < 0.05, **p < 0.01, ***p < 0.001. *Above single bars indicate statistical significances to untreated (0 µM DOX) conditions of the same group

Fig. 7. Effects of DOX on cellular electrophysiology and proarrhythmic triggers. A Representative confocal line scans (Fluo-4) showing diastolic sarcoplasmic reticulum Ca2+ sparks and B mean sarcoplasmic reticulum Ca2+ leak (norm. to untreated) after DOX treatment (0.25 µM, 24 h) in ACT patients (untreated (0 DOX): n = 79 cardiomyocytes /4 differentiations; 0.25 µM DOX: n = 79/4) versus control (untreated (0 DOX): n = 120/6; 0.25 µM DOX: n = 120/6). p values were calculated using Mann–Whitney test. C Original unstimulated recordings (whole-cell current-clamp) of iPSC CM during 10 s rest after a series of 30 elicited action potentials at 1 Hz showing delayed afterdepolarizations (DADs) after treated with DOX for 24 h. D Incidence of DADs/min of iPSC CM after DOX treatment for 24 h in patients (Untreated: n = 9 cells/3 differentiations; DOX: n = 13/3) compared to control (untreated: n = 8/2; DOX: n = 8/2). E Incidence of DADs/min of iPSC CM after DOX treatment for 24 h in patients (Untreated: n = 8 cells/2 differentiations; DOX: n = 13/2) compared to iPSC-CM from patients treated with DOX (0.25 µM) and AIP (1 µM), both for 24 h (n = 8/2; DOX: n = 6/2). F Original recordings (whole-cell…

Fig. 8. DOX-dependent differential gene expression in both ACT patient EHM and control EHM. A–F For control EHM, 5–6 independent EHM from 1 cell line of control patient 1 and 2, and for ACT patient EHM, 6–7 independent EHM from 1 cell line of each of the 3 ACT patients (ACT1, ACT2, ACT3) were analyzed. A DOX-induced differentially expressed genes (DEGs) in ACT patient EHM and control EHM compared to untreated conditions illustrated in a Venn diagram. B PCA plot of control and ACT patient EHM samples used for the analysis. C Normalized counts of DOX-dependent upregulated DEGs (BAX, SYNM, SIRT1). D Heat map of 9 structural genes that are significantly downregulated after DOX treatment in control and ACT patient EHM. Normalized counts of DOX-dependent downregulated TOP2B and QKI. E Normalized counts of ACT-specific, DOX-dependent upregulated DEGs (SET7, EIF2AK4, EGFR) as well as downregulated DEGs (SMYD1, PRDM16). F Significant enriched GO terms after DOX in control EHM compared to untreated conditions according to ClueGo Cytoscape plugin. The data of differential expression are based on all up-regulated genes. Data are generated by GO cellular component, GO biological process or KEGG…

Fig. 9. Individuals with CD20 + B-cell lymphoma who had received high doses of DOX and suffered cardiac dysfunction were studied and compared to control-iPSC CM from cancer survivors without cardiac symptoms. In cellular studies, ACT-iPSC CM were persistently more susceptible to DOX toxicity including augmented, mitochondrial dysregulation, increased reactive oxygen species, increased apoptotic events, higher beating rate, and disorganized myofilament structure. ACT cells exhibited hyperactive CaMKIIδ and hyperphosphorylation of PLN-T17 and RYR2-S2814 but lowered SERCA2a expression. Ca2+ transient amplitude of ACT-iPSC CM was reduced compared to control cells, and diastolic sarcoplasmic reticulum Ca2+ leak was DOX-dependently increased. Prolonged action potentials in ACT CM suggest a cellular link to arrhythmogenic events and contractile dysfunction especially found in ACT engineered human myocardium. The CaMKII-inhibitor AIP rescues the DAD incidence in ACT-iPSC CM, confirming the contribution of the CaMKII to DOX-dependent altered electrophysiology, proarrhythmic triggers and ACT development. In contrast, control CM upregulated SERCA2a expression in a DOX-dependent manner, without…