Facioscapulohumeral Muscular Dystrophy is Associated With Altered Myoblast Proteome Dynamics

Figures

• 

  Fig. 1. Dynamic proteome profiling of two pairs of FSHD and UASb myoblasts.A, experimental design and workflow for sample preparation and analysis. B, characteristics of FSHD and UASb donors, including shortened length of the 4q D4Z4 repeat array (bold red) and muscle score using the Medical Research Council (MRC) scale where 5/5 is maximum strength. Data were retrieved from Homma et al. (21). C, matrix correlation of abundance data, n = 4420 proteins, using Pearson’s correlation coefficient. Bi, biceps; Del, deltoid.

• 

  Fig. 2. Differences in protein abundance between FSHD and UASb myoblasts.A, volcano plot comparing the Log2 Fold-Difference (FSHD/UASb) protein abundance plotted against the −Log10p value (n = 4420). Colored data points represent proteins more abundant (red, log2 Diff. > 1), less abundant (blue, log2 Diff. < −1), or stable (grey, log2 Diff. < 1 and > −1) in FSHD compared to UASb myoblasts. The dashed horizontal line shows a threshold of statistical significance (p < 0.05). Gene ontology (GO) analysis of cellular component and biological process in proteins (B) depleted (log2 Diff. < −1) or (C) enriched (log2 Diff. > 1) in FSHD compared to UASb myoblasts. GO terms were ranked by −log10 (q value) and the number of proteins included in each GO term reported alongside each entry. Each bar chart colour scale represents the level of GO enrichment. STRING protein interaction network for proteins (D) depleted (log2 Diff. < −1) or (E) enriched (log2 Diff. > 1) in FSHD compared to UASb myoblasts. The STRING interaction network was generated using a minimum interaction score of 0.7 and the interaction network was clustered using k-means clustering. Proteins without interaction partners were omitted…

• 

  Fig. 3. Differences in protein turnover rate between FSHD and UASb myoblasts.A, volcano plot comparing the Log2 fold-difference (FSHD/UASb) in protein turnover plotted against the −Log10p value (n = 2324 proteins). Colored data points represent proteins with greater (red, log2 Diff. > 1), lesser (blue, log2 Diff. < −1), or stable (grey, log2 Diff. < 1 and > −1) rates of turnover in FSHD compared to UASb myoblasts. Dashed horizontal line denotes a significance threshold of p < 0.05. B, gene ontology (GO) analysis of cellular component and biological process for proteins exhibiting a lesser turnover rate in FSHD compared to UASb (log2 Diff. < −1). GO terms were ranked by -log10 (q value) and the number of proteins included in each GO term reported alongside each entry. Each bar chart colour scale represents the level of GO enrichment. C, density plot of log10 transformed fractional protein turnover rate (FTR, %/h) in FSHD and UASb myoblasts. The vertical black line in each density plot indicates the median for each sample, whereas the red vertical line indicates (0 log10) represents an FTR of 1%/h. No difference in FTR profile (one-way ANOVA, p = 0.37) were detected between FSHD and UASb…

• 

  Fig. 4. Mitochondrial proteins are more abundant but exhibit slower protein turnover rates in FSHD myoblasts.A, scatter plot comparing the differences in the Log2 Fold-Difference (FSHD/UASb) between protein abundance (x-axis) and protein FTR (y-axis). Rug plots display the distribution of individual data both in X and Y axis. B, STRING interaction network of proteins more abundant (log2 Diff. > 0.6) and slower FTR (log2 Diff. < −0.6) in FSHD myoblasts. The STRING interaction network was generated using a minimum interaction score of 0.7 and the interaction network was clustered using k-means clustering.

• 

  Fig. 5. Dysregulation of mitochondrial protein is common in FSHD despite variation between families. Principal Component Analysis (PCA) on (A) protein abundance and (B) FTR data in FSHD and UASb myoblasts of Family #15 and family #16. Venn diagrams of abundance (C) and FTR (D) data illustrating the number of strongly regulated (log2 Diff. > 1 or < −1) proteins common to family #15 and family #16. Scatter plots illustrating family-specific Log2 Fold-Differences (FSHD/UASb) in protein abundance (x-axis) and protein FTR (y-axis). Black circles represent proteins strongly regulated (log2 Diff. > 1 or < −1) in abundance and FTR, only mitochondrial proteins were annotated (green circles). Blue circles in (F; family #16A/16U) highlight nine mitochondrial proteins that were strongly regulated in family #15 (E).

• 

  Fig. 6. Dysregulation of mitochondrial proteins tends to be reversed by 2′-MOE antisense oligomer treatment targeting DUX4.A, experimental design of a subsequent independent study including treatment of FSHD and UASb myoblasts with a 2′-MOE modified antisense oligomer targeting exon three of DUX4. Scatter plots (B–E) compare the Log2 Fold-Difference between FSHD and UASb (x-axis) and the Log2 Fold-Difference in 2′-MOE treated FSHD myoblasts against vehicle control for abundance data of family #15 (B) or family #16 (C) and FTR data of family #15 (D) or family #16 (E). Rug plots display the distribution of data on x-axis and y-axis. A linear regression line with 95% confidence intervals was drawn in each figure panel. The negative slope of the regression line indicates a counter effect of 2′-MOE modified antisense oligomer treatment against FSHD.

• 

  Fig. 7. Accumulation of “older” less viable mitochondrial proteins may contribute to the pathophysiology of FSHD. FSHD myoblasts exhibit a greater protein abundance but a slower turnover rate of mitochondrial respiratory complex subunits and mitochondrial ribosomal subunits, which may indicate an accumulation of ‘older’ less viable mitochondrial proteins compared to UASb myoblasts. This may contribute to the reduced respiratory function specifically observed in Complex I as recently shown by Heher et al. (8) in DUX4 expressing iDUX4 myoblasts. Impaired mitochondrial function is proposed as one of the major pathophysiological mechanisms of FSHD. Created with BioRender.com.