Molecular pathways behind acquired obesity: Adipose tissue and skeletal muscle multiomics in monozygotic twin pairs discordant for BMI

Figures

• 

  Figure 1. KEGG pathway enrichment analysis of differentially expressed genes in co-twins (A) The heatmap presents pathways that are significantly different in both tissues. (B) The heatmap presents significantly altered pathways in adipose tissue (n = 49 twin pairs). (C) The heatmap presents significantly altered pathways in skeletal muscle (n = 44 twin pairs). The direction and significance of each pathway was based on the gene set as a whole while considering the p value and fold change for each gene. KEGG pathways with FDR p < 0.05 are shown; blue indicates significantly downregulated pathways in the heavier co-twins, whereas red indicates upregulated pathways. Pathways are grouped according to their biological function. AT, adipose tissue; SkM, skeletal muscle; ECM, extracellular matrix.

• 

  Figure 2. Heatmaps showing the differential expression of individual genes in four mitochondrial pathways among co-twins The heatmaps show the downregulation in transcription levels for the majority of genes involved in the central mitochondrial pathways in AT (n = 49 twin pairs) and SkM (n = 44 twin pairs). The log2-fold differences indicated for the genes are based on the KEGG pathways. The color in the heatmaps reflects the differential expressions associated with the heavier co-twins, where blue indicates downregulation and red indicates upregulation. Asterisks indicate statistically significant differential expression: ∗∗∗FDR p < 0.001, ∗∗FDR p < 0.01, ∗FDR p < 0.05, §nominal p < 0.05. TCA, tricarboxylic acid; BCAA, branched-chain amino acids.

• 

  Figure 3. Reporter metabolite representation for heavier co-twins The reporter metabolites algorithm marks the regions during metabolism around which significant transcriptional changes occur in AT (n = 49 twin pairs) and SkM (n = 44 twin pairs). Reporter metabolites are obtained using the p values calculated from the comparison of heavier co-twins with leaner co-twins. A) Reporter metabolites for the associated subcellular compartments and divided based on the direction (down/up) of the related gene sets (nominal p < 0.05) in the heavier co-twins. (B) Reporter metabolites associated with key mitochondrion-related pathways were compared between AT and SkM (nominal p < 0.05). The heatmap shows the downregulated gene sets in the heavier co-twins that are significant in at least one of the tissues. In AT, no upregulated gene sets were associated with these pathways. In SkM, the gene set associated with citrate [c] was upregulated (data not shown). ER, endoplasmic reticulum; [c], cytosol; [m], mitochondrion; [p], peroxisome.

• 

  Figure 4. Metabolites and proteins among co-twins in AT and SkM (A–D) The direction (log2-fold changes) of the TCA cycle and pyruvate metabolite levels in the heavier co-twins (A), log2-fold changes of OXPHOS metabolites (B), log2-fold changes in the BCAA cycle metabolite levels (C), and log2-fold changes of the glycolysis protein levels in SkM (D). AT (yellow, n = 47 twin pairs); SkM (red, n = 40 twin pairs for metabolomics, n = 48 pairs for proteomics). ∗∗nominal p < 0.01, ∗nominal p < 0.05.

• 

  Figure 5. Within-pair associations of four mitochondrial and three inflammatory pathways with clinical outcome measures (A) Standardized coefficients (βs) in linear mixed models with the delta clinical outcome between co-twins as the dependent variable, the delta for the mitochondrial and inflammatory pathways scores as the fixed effect, and family ID as a random effect and adjusted for sex, age, and diabetes status in AT (n = 49 twin pairs). (B) Standardized coefficients (βs) in linear mixed models with the delta clinical outcomes between co-twins as the dependent variable, the delta for the mitochondrial and inflammatory pathways scores as the fixed effect, and family ID as a random effect and adjusted for sex, age, and diabetes status in SkM (n = 44 twin pairs). Error bars denote 95% confidence intervals. BMI, body mass index; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TAG, triacylglycerol; HOMA-IR, homeostatic model for the assessment of insulin resistance; CRP, C-reactive protein.

• 

  Figure 6. Summary of key metabolic differences in AT and SkM tissue metabolism in acquired obesity Results from AT (transcriptomics and metabolomics) and SkM (transcriptomics, proteomics, and metabolomics) are included. Significant differences between co-twins from proteomics are indicated by P and metabolomics by M (nominal p < 0.05). The color of the arrows reflects the differential expressions associated with the heavier co-twins, where blue indicates downregulation, and red indicates upregulation. OXPHOS, oxidative phosphorylation; BCAA, branched-chain amino acids.