Fig. 1: Expansion of the LD reorganizes PM ultrastructure in vivo.
From: Plasma membrane remodeling determines adipocyte expansion and mechanical adaptability
a Examples of caveolae identification and classification. b PM-LD impacts density and complexity of caveolar domains. Scale bar: 100 nm. c Distance at caveolar coordinates (as calculated in Supplementary Fig. 1c.2) for different caveolae types. Sample size: 6908 single-pit, 6362 caveolae in rosettes, 8729 cavicles. Individual datapoints and boxplot outliers were removed in the representation for simplicity. Boxplot boxes represent first, second (center), and third quantile. Box whiskers expand to the full range of data. The representation shows the distribution of all caveolae in the sample, but statistical analysis was performed comparing median through-caveolae distance by cell with paired Wilcoxon sign rank test (sample size: 64 cells, p values: 3.9 × 10−8 for rosette and single-pit comparison, 1.7 × 10−7 for cavicle and single pit comparison). d–g Caveolae density measured as caveolae/µm of cell perimeter for total caveolae (d), caveolae in rosettes (e), cavicles (f) and single-pit caveolae (g). Sample size=64. p values shown for the null hypothesis slope=0. For d–g error bands show 95% confidence interval for the estimation of the mean (line). h Caveolae counts in regions of increasing PM-LD distance, normalized by the frequency of the regions in the sample. i–l Spatial concentration of caveolae measured as caveolae/µm2 for total caveolae (i), rosette caveolae (j), cavicles (k), and single-pits (l) as a function of median PM-LD distance. Each dot is an individual cell. Sample size = 64 cells. p values shown for the null hypothesis slope=0. For i–l error bands show 95% confidence interval for the estimation of the mean (line). p-value codes: *<0.05, **<0.01, ***<0.001. All tests are two-sided. Source data are provided as a SourceData.zip.
