Hepatic Transcriptomics Reveals that Lipogenesis Is a Key Signaling Pathway in Isocitrate Dehydrogenase 2 Deficient Mice

Abstract

Mitochondrial nicotinamide adenine dinucleotide phosphate (NADP⁺)-dependent isocitrate dehydrogenase (IDH2) plays a key role in the intermediary metabolism and energy production via catalysing oxidative decarboxylation of isocitrate to α-ketoglutarate in the tricarboxylic acid (TCA) cycle. Despite studies reporting potential interlinks between IDH2 and various diseases, there is lack of effort to comprehensively characterize signature(s) of IDH2 knockout (IDH2 KO) mice. A total of 6583 transcripts were identified from both wild-type (WT) and IDH2 KO mice liver tissues. Afterwards, 167 differentially expressed genes in the IDH2 KO group were short-listed compared to the WT group based on our criteria. The online bioinformatic analyses indicated that lipid metabolism is the most significantly influenced metabolic process in IDH2 KO mice. Moreover, the TR/RXR activation pathway was predicted as the top canonical pathway significantly affected by IDH2 KO. The key transcripts found in the bioinformatic analyses were validated by qPCR analysis, corresponding to the transcriptomics results. Further, an additional qPCR analysis confirmed that IDH2 KO caused a decrease in hepatic de novo lipogenesis via the activation of the fatty acid β-oxidation process. Our unbiased transcriptomics approach and validation experiments suggested that IDH2 might play a key role in homeostasis of lipid metabolism.

Keywords: Idh2; bioinformatics; hepatic transcriptomics; knockout mouse; lipid metabolism.

Figure 1

The validation of Idh2 deletion in mice. The ablation of Idh2 was verified by tail DNA genotyping method (A), and by comparing IDH2 protein expression in liver tissues of wild-type (WT) and IDH2 KO mice (B).

Figure 2

Procedures of hepatic transcriptomics dataset. The analyses of transcriptomics dataset followed by computational analyses and qPCR validation: A flowchart (A), A volcano plot (B), and heat map (C) showed differentially expressed genes and the relationships among samples used, respectively.

Figure 3

Canonical pathway analysis using Ingenuity Pathway Analysis software. Enriched canonical pathways in IDH2 KO mice liver were listed. The green bars indicate the number of downregulated genes while the red portion in bars show the number of upregulated genes.

Figure 4

Biofunctions, networks, and upstream analyses using the Ingenuity Pathway Analysis (IPA). Enriched functional roles of networks in IDH2 KO mice are listed (A), and IPA Networks only related to lipid metabolism are integrated. The key transcripts at core position of the networks are highlighted with red color (B). Upstream regulators including Srebf1, Srebf2, and Scap, were predicted in the interactome image (C).

Figure 5

Functional analyses of differentially expressed genes by DAVID gene ontology (GO) enrichment analysis. The representative GO terms for biological processes (A) were utilized for DAVID functional annotation clustering of categories (B).

Figure 6

KEGG AMPK signaling. The AMPK signaling pathway was predicted as the most enriched pathway from the KEGG pathways analysis provided by the DAVID tools. The key player genes are highlighted with red color.

Figure 7

The validation of RNA sequencing results using quantitative PCR. The genes found at the core positions of the networks, upstream regulators, canonical pathway (TR/RXR), and found in the functional annotation clustering of DAVID tools were randomly selected and validated using qPCR analyses. The data were expressed as the means ± standard error of means (SEM; n = 5 for male, and n = 7 for female) (A). In addition to the validation, the representative genes of each lipid metabolic process were randomly selected and their expression levels were assessed. The genes responsible for fatty acid uptake (Cd36, and Slc27a1), fatty acid β-oxidation (Sirt1, Ppargc1, and Acox1), fatty acid esterification (Dgat2), and fatty acid desaturation/elongation (Scd1, and Elovl6) were quantified using qPCR analysis (B). * p < 0.05; ** p < 0.01; *** p < 0.001.