Oxidized LDL‐dependent pathway as new pathogenic trigger in arrhythmogenic cardiomyopathy

Figures

• 

  Figure 1. ACM patients show high plasma oxLDL, cardiac lipid peroxidation, and oxLDL receptor levels oxLDL plasma concentration in ACM patients and HC (n = 36; Mann–Whitney test).oxLDL plasma concentration in mutated ACM (n = 7) and non‐ACM relatives, carriers of the same causative mutation (n = 9; Mann–Whitney test).oxLDL plasma concentration in ACM patients carriers of a PKP2 mutation and ACM patients carriers of other desmosomal or non desmosomal mutations, or gene elusive (n = 10 vs. n = 26; one‐way ANOVA).Representative images of MDA immunostaining (green) on ACM and HC ventricular tissue sections and relative quantification (n = 4 biological replicates; two‐tailed Student’s t‐test). Nuclei are counterstained with Hoechst33342 (blue).Representative images of CD36 immunostaining (green) on ACM and HC ventricular tissue sections and relative quantification (n = 4 biological replicates; two‐tailed Student’s t‐test). Nuclei are counterstained with Hoechst33342 (blue). Data information: mean ± SEM. *P < 0.05; **P < 0.01; and ***P < 0.001. Source data are available online for this figure.

• 

  Figure 2. Plasma oxLDL cut‐off value of 86 ng/ml defines two patient subpopulations with different severity of ACM phenotypes ROC curve shows the capability of plasma oxLDL concentration in discriminating ACM vs. HC subjects (data as in Fig 1A). The red dot shows the point where the difference between sensitivity and specificity is minimized; this value, corresponding to the oxLDL cut‐off = 86 ng/ml, was used to part the whole ACM cohort in two subpopulations.Left panels: representative images of three cases of ACM in steady‐state free procession sequences at cardiac MRI. On the left, a case with RV dilation, regional bulging of the RV wall without fat infiltration, and oxLDL levels below the cut‐off; in the center and on the right, two cases with fat infiltration in the right or left ventricle wall (white arrow), respectively, and oxLDL levels above the cut‐off. Right panel: quantification of the myocardial fat mass of the two ACM subpopulations (above or below the cut‐off) whose MRI was available for re‐analysis (n = 14 oxLDL < 86 ng/ml, n = 25 oxLDL > 86 ng/ml; two‐tailed Student’s t‐test).Frequency of patients showing RV dysfunction (defined as in (31)) in the two ACM cohort subgroups…

• 

  Figure 3. ACM C‐MSC show elevated oxidative stress, PPARγ, and CD36 expression, and, during adipogenic differentiation, CD36 levels and functionality Left panel: exemplificative FACS analysis of the FITC emission of unstained HC C‐MSC (autofluorescence; white), HC C‐MSC in GM treated with DCF (gray), ACM C‐MSC in GM treated with DCF (red), and HC cells in GM treated with 2 mmol/l H2O2 as positive control (blue). Right panel: mean DCF emission of HC and ACM C‐MSC in GM (n = 5 biological replicates; two‐tailed Student’s t‐test).Left panels: representative images of MDA immunostaining (green) on ACM and HC C‐MSC in GM. Nuclei are counterstained with Hoechst33342 (blue). Right panel: image quantification (n = 4 biological replicates ACM, n = 5 biological replicates HC; two‐tailed Student’s t‐test).GSH/GSSG ratio quantification in HC and ACM C‐MSC cultured in GM (n = 8 biological replicates).Top panel: representative images of Western blot analysis of proteins extracted from ACM and HC C‐MSC cultured in GM, hybridized with anti‐CD36 and anti‐PPARγ antibodies. Immunostaining of the housekeeping GAPDH is shown for normalization. Bottom panel: d.a. of PPARγ (n = 18 biological replicates) and CD36…

• 

  Figure 4. ACM C‐MSC lipogenesis is increased with oxLDL or 13HODE and prevented by NAC treatment Left panels: representative images of ORO staining on ACM and HC C‐MSC in AM supplemented or not with 150 µg/ml oxLDL. Middle panel: image quantification (n = 11 biological replicates; two‐way ANOVA). Right panels: representative images of Western blot of CD36, PPARγ, and GAPDH protein expression of ACM and HC C‐MSC protein extracts in AM supplemented or not with 150 µg/ml oxLDL (n = 5 biological replicates) and d.a. normalized on the housekeeping GAPDH (two‐way ANOVA).Left panels: representative images of ORO staining on ACM and HC C‐MSC in AM supplemented or not with 20 μg/ml 13HODE, 5 mmol/l NAC, or both. Right panel: image quantification (n = 13 biological replicates; two‐way ANOVA).Left panel: representative images of Western blot of CD36, PPARγ, and GAPDH protein expression of ACM and HC C‐MSC protein extracts in AM supplemented or not with 20 μg/ml 13HODE, 5 mmol/l NAC, or both (n = 8 biological replicates). Right panels: d.a. normalized on the housekeeping GAPDH (two‐way ANOVA). Data information: mean ± SEM. *P < 0.05, **P < 0.01, and ***P < 0.001. Source data are available online for…

• 

  Figure 5. CD36 silencing reduces lipid accumulation and PPARγ antagonism reduces CD36 expression and function in ACM C‐MSC Left panel: representative images of Western blot of CD36, PPARγ, and GAPDH expression of protein extracts of ACM C‐MSC treated with scramble siRNA or CD36 siRNA. Right panels: d.a. normalized on the housekeeping protein GAPDH (n = 7 biological replicates; two‐tailed Student’s t‐test).Left panels: representative images of ORO staining on ACM C‐MSC treated with scramble siRNA or CD36 siRNA and cultured in AM with 150 µg/ml oxLDL. Right panel: image quantification (n = 7 biological replicates; two‐tailed Student’s t‐test).Left panel: correlation between CD36/GAPDH and PPARγ/GAPDH d.a. Right panel: correlation between CD36/GAPDH d.a. and ORO lipid accumulation. Regression lines, their equations, R 2, and correlation P values are shown (X‐Y correlation).Left panel: representative images of internalization of oxLDL (red) in ACM C‐MSC treated with scramble siRNA or CD36 siRNA, cultured in AM and subjected to 10 μg/ml DiI‐oxLDL treatment. Right panel: quantification of the DiI fluorescence normalized on nuclei number (n = 3 biological replicates, arbitrary units; two‐tailed…

• 

  Figure 6. HFD increases plasma oxLDL, cardiac adipose substitution, PPARγ and CD36 expression, and lipid peroxidation, and impairs RV function in ACM mice Left panel: plasma total cholesterol in the different lipoprotein fractions of WT and Pkp2+/− mice fed a CD (green line) or a 3‐month HFD (blue line; n = 7). Right panel: plasma total cholesterol quantity in fractions 29–32, corresponding to low‐, very‐low‐, and intermediate‐density lipoproteins, of WT and Pkp2+/− mice fed a CD (green line) or a 3‐month HFD (blue line; n = 7; Mann–Whitney test).Plasma concentration of oxLDL in WT and Pkp2+/− mice samples, fed a CD and a 3‐month HFD (n = 6; two‐tailed Student’s t‐test).Left panels: representative images of ORO staining of HFD‐fed WT and Pkp2+/− cardiac sections. Right panel: quantification of ORO positive area percentage (n = 10). For comparison, quantification of ORO positive area of cardiac sections of CD‐fed WT and Pkp2+/− mice (n = 9; two‐way ANOVA) is shown (Appendix Fig S3).Representative images of PPARγ (green) immunostaining on HFD‐fed WT and Pkp2+/− mice cardiac sections (n = 10; two‐way ANOVA). Quantification of the PPARγ staining in HFD is shown relative to the values of CD (n…

• 

  Figure 7. Atorvastatin prevented ACM phenotype in HFD‐fed Pkp2+/− mice Left panel: plasma total cholesterol in the different lipoprotein fractions of WT and Pkp2+/− mice fed a 3‐month HFD (blue line) or a 3‐month HFD+atorvastatin (red line). Right panel: plasma total cholesterol quantity in fractions 29–32, corresponding to low‐, very‐low‐, and intermediate‐density lipoproteins, of WT and Pkp2+/− mice fed a 3‐month HFD (blue line; n = 7) or a 3‐month HFD+atorvastatin (red line; n = 9; two‐tailed Student’s t‐test).oxLDL plasma concentration in WT and Pkp2+/− mouse samples, fed a 3‐month HFD plus atorvastatin (n = 8), compared to HFD (as in Fig 6B; two‐tailed Student’s t‐test).Left panel: representative images of ORO staining of cardiac sections of Pkp2+/− mice, fed a 3‐month HFD plus atorvastatin. Right panel: quantification of the percentage of ORO positive area (n = 9) is compared to that in HFD (as in Fig 6C; two‐tailed Student’s t‐test).Representative images of PPARγ immunostaining (green) on cardiac sections of Pkp2+/− mice fed a 3‐month HFD plus atorvastatin (n = 9). Quantification is compared to the values of Pkp2+/− in HFD and relative to WT in CD (as in Fig 6D; two‐tailed Student’s…