Selective serotonin reuptake inhibitors induce cardiac toxicity through dysfunction of mitochondria and sarcomeres

Figures

• 

  Fig. 1. SSRIs toxicity on differentiated cardiomyocytes. a The procedure of SSRIs treatment on differentiated cardiomyocytes. The LC50 levels of SSRIs toxicity on differentiated cardiomyocytes of H1-CMs (b) and hiPS-CMs (c). The ROS levels of differentiated H1-CMs (d) and hiPS-CMs (e) after five days of SSRIs exposure. n = 6 biological replicates of each group for (b–e). The ATP production of differentiated H1-CMs (f) and hiPS-CMs (g) after five days of SSRIs exposure. ATP production was assessed using n = 4 and n = 8 biological replicates per group for H1-CMs and hiPS-CMs, respectively. h–l The mitochondrial respiration levels of differentiated H1-CMs after five days of SSRIs exposure. h Graphical representation of the mitochondrial stress test, i basal respiration, j maximal respiration, k spare respiration capacity, l ATP-linked respiration. *indicates p < 0.05, **indicates p < 0.01, ***indicates p < 0.001, ****indicates p < 0.0001, ns indicates no significance. n = 4 biological replicates of each group for (h–l).

• 

  Fig. 2. SSRIs toxicity on differentiating cardiomyocytes. a The procedure of SSRIs treatment on differentiating cardiomyocytes. b The identification of NOAEL of three SSRIs during the entire differentiation process of H1-CMs. c The ATP production of differentiating H1-CMs after SSRIs exposure (n = 8 biological replicates). d The mitochondrial membrane potential of differentiating H1-CMs after SSRIs exposure (n = 7 biological replicates). e–i The mitochondrial respiration levels of differentiating H1-CMs after SSRIs exposure. e Graphical representation of the mitochondrial stress test. f basal respiration, g maximal respiration, h spare respiration capacity, i ATP respiration. *indicates p < 0.05, **indicates p < 0.01, ***indicates p < 0.001, ****indicates p < 0.0001. n = 5 biological replicates for DMSO and Sertraline groups; n = 7 for Fluoxetine and Paroxetine groups.

• 

  Fig. 3. Fluorescence staining of mitochondria and sarcomere of differentiating cardiomyocytes. a Fluorescence staining of mitochondria in H1-CMs with MitoTracker (green for mitochondria and blue for cell nucleus). b Immunofluorescence staining of sarcomere with the primary antibodies of cTNT and α-actinin in H1-CMs (green for α-actinin, red for cTNT, and blue for cell nucleus). The mean branch length (c) and branches per network (d) of mitochondria were calculated by Fiji/ImageJ with the MiNA plugin. n = 7 biological replicates of each group for (c and d). Fast Fourier transformation (FFT) analysis showed the sarcomere length (e) and the sarcomere organization score (f) of α-actinin signal distribution in the H1-CMs. *indicates p < 0.05, **indicates p < 0.01, ***indicates p < 0.001, ****indicates p < 0.0001, ns indicates no significance. n = 7 biological replicates of each group for (e and f).

• 

  Fig. 4. Transcriptomic analysis of SSRIs toxicity in the 2D model. a Principal component analysis for the samples. b Heatmap indicates the overall DEG expression pattern of all 4 groups of samples: C (control, DMSO), F (fluoxetine), P (paroxetine), and S (sertraline). c Comparison of the number of shared and distinct DEGs in three SSRIs compared to control. d–f Volcano plot showed the top altered genes after three SSRIs compared to control. Some representative genes are highlighted in blue in the plot. g–i GO-BP terms and DEGs relationship network in three SSRIs compared to control. Each yellow node (circle) represents distinct GO-BP terms, and the circle size indicates the number of differentially expressed genes in the term. The edges represent categories of GO-BP terms with different color lines.

• 

  Fig. 5. SSRIs toxicity on differentiating cardiac organoids. a The procedure of SSRIs treatment on differentiating cardiac organoids. b The identification of NOAEL of three SSRIs during the entire differentiation process of cardiac organoids. c The morphology changes of cardiac organoids derived from H1-hESCs. d–h The mitochondrial respiration levels of SSRIs toxicity on differentiating cardiac organoids. d Graphical representation of the mitochondrial stress test. e Basal respiration, f maximal respiration, g spare respiration capacity, h ATP-linked respiration. *indicates p < 0.05, **indicates p < 0.01, ***indicates p < 0.001. n = 4 biological replicates for all groups.

• 

  Fig. 6. Immunofluorescence staining of 3D model. a Immunofluorescence staining of sarcomere with the primary antibodies of CD31, WT1, cTNT, and α-actinin in 3D-model. b–e The fluorescence intensity was calculated by Fiji/ImageJ and normalized to cell counts. b CD31, c cTNT, d WT1, e α-actinin. * indicates p < 0.05. ns indicates no significance. n = 3 biological replicates for all groups.

• 

  Fig. 7. Transcriptomic analysis of SSRIs toxicity in the 3D model. a Principal component analysis for the samples. b Heatmap indicates the overall DEG expression pattern of 2 groups of samples: P1000 (paroxetine, 1000 nM) and P0 (paroxetine, 0 nM). c Volcano plot showed the DEGs between paroxetine and control. Some representative genes are highlighted in blue in the plot. d, e GO-BP terms enrichment in paroxetine compared to control. f GO-BP terms and DEGs relationship network in paroxetine compared to control. Each yellow node (circle) represents distinct GO-BP terms, and the circle size indicates the number of differentially expressed genes in the term. The edges represent categories of GO-BP terms with different color lines. g The identification of RNA-seq was verified by qPCR. *indicates p < 0.05, **indicates p < 0.01, ***indicates p < 0.001. For qPCR, n = 6 biological replicates; for RNA-seq, n = 3.