Figure 4.

VAT lipids measured in morbid obese and lean control individuals (cohort 2)

(A) Volcano plot of the differentially abundant lipids in VAT between obese and lean control subjects. In addition to the lipid classes measured in the other samples, triglycerides (Tg) were measured. Colored dots highlight significant up- or down-regulated individual lipid species (fold change ≥1.5 and p < 0.05, Welch’s corrected).

(B) Lipid class repartition (PC, PE, PI, CL, HexCer, Cer, and SM) in human VAT from lean control and obese individuals (in mol %).

(C and D) Relative PC level changes (mol %) in VAT collected from lean control and obese individuals, clustered according to the chain length (C), and represented according to the degree of saturation (D). LC (long Chain C28-34), VLC (Very Long Chain C38-44).

(E and F) Relative PI level changes (mol %) in VAT, represented according to the nature of the fatty acid linkage [diacyl or monoacyl (lyso)] (E) and the degree of saturation (F).

(G) Relative SM level changes (mol %) in VAT collected from lean control and obese individuals, clustered according to the chain length.

(H and I) Association between the relative levels of ceramides detected in sera (H) or VAT (I), and the BMI of the subjects (serum n = 21, VAT n = 17, Spearman correlation, for the serum R = 0.412, p = 0.057, for the VAT R = 0.315, p = 0.2). Statistics for (B–G) are unpaired Student’s t test p values. Data from obese (n = 11) and control (n = 7) individuals are represented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01; ∗∗∗p < 0.001.