Metabolites and Genes behind Cardiac Metabolic Remodeling in Mice with Type 1 Diabetes Mellitus

Figures

• 

  Figure 1. Validation of T1DM phenotype in the Akita mice. (A) Fasting blood glucose levels were measured by glucometer in WT and Akita mice. Values are mean ± SEM. n = 6. (B) Serum insulin levels were measured by ELISA method in WT and Akita mice. n = 5–6. Values are mean ± SE. Statistical analysis was conducted via a paired Student’s t-test. *** p < 0.001; * p < 0.05 (the exact p-value was 0.01).

• 

  Figure 2. LC-MS metabolomics analysis of Akita compared to WT hearts. (Ai,Aii) Top predicted canonical metabolic pathways disrupted in the Akita heart with a –log(p-value) > 1.3 based on metabolite IPA analysis. (B) Heatmap indicating percent change of individual metabolites associated with the listed pathways in the Akita and WT hearts based on the expression log ratio calculated from individual metabolite expression; p-value < 0.05. n = 4.

• 

  Figure 3. Differential expression of key metabolites involved in the tricarboxylic acid (TCA) cycle in the Akita heart by Ingenuity Pathway Analyses (IPA) of metabolites. (A) TCA cycle metabolite intermediates in Akita versus WT mice. Line indicates no change between groups. *FADH levels were undetected. (B) Heatmap indicating percent change of single metabolite levels associated with the TCA cycle in Akita and WT mice. Analysis conducted with Ingenuity Pathway Analysis software. n = 4.

• 

  Figure 4. Genomics analysis of the Akita compared to WT hearts from next generation sequencing. (A) Based on Ingenuity Pathway Analysis (IPA) of genes obtained from next-generation sequencing showing the top predicted canonical metabolism pathways disrupted in the Akita heart with a –log(p-value) > 1.3. (B) Heatmap showing percent change in genes associated with the aforementioned metabolic pathways in the Akita heartbased on the expression log ratio calculated from FPKM values. All genes associated with the predicted pathways had a p-value < 0.05. n = 2.

• 

  Figure 5. Independent and synergistic effects of genes and metabolites on metabolic remodeling in the Akita heart. (A) Overlap in genomic and metabolomic pathway analyses between disrupted metabolism pathways in the Akita heart. A total of 108 and 30 canonical metabolic pathways are predicted to be affected based on metabolomic and genomic analysis, respectively. There is an overlap of 15 disrupted metabolism pathways in the Akita heart between both analyses. (B) List of canonical metabolism pathways with a –log(p-value) > 1.3 commonly disrupted in both the genomic and metabolomic analysis in the Akita versus WT heart. Analysis was conducted with Ingenuity Pathway Analysis software. n= 2–4.

• 

  Figure 6. Characterization of the Akita mouse model. (A) Akita mice have a G-to-A point mutation in their Insulin 2 (Ins2) gene, causing the severing of disulfide from a cysteine-to-tyrosine substitution. (B) Genotyping of Akita and littermate/sibling, normoglycemic WT mice. WT mice present a single band representing an allele at 140 base pairs (bp), while Akita mice demonstrate two bands at 140 and 280 bp, indicating the presence of two alleles.

• 

  Figure 7. Overall study design for metabolomic and genomic analysis of heart tissue of Akita. Tissue was isolated specifically from the heart’s left ventricle (LV). LV tissue was prepared for liquid chromatography-mass spectrophotometry (LC-MS), and metabolites were analyzed. Ingenuity Pathway Analysis (IPA) was used to determine the association of metabolites with metabolic pathways. LV tissue was also used for RNA isolation and next-generation sequencing to evaluate gene expression. IPA was performed on differentially expressed genes to evaluate their roles in disrupted metabolic pathways. Both metabolomic and genomic analyses were compared to identify key pathways behind the regulation of metabolic remodeling in the Akita heart.