Deletion of the mitochondrial chaperone TRAP-1 uncovers global reprogramming of metabolic networks

Abstract

Reprogramming of metabolic pathways contributes to human disease, especially cancer, but the regulators of this process are unknown. Here, we have generated a mouse knockout for the mitochondrial chaperone TRAP-1, a regulator of bioenergetics in tumors. TRAP-1(-/-) mice are viable and showed reduced incidence of age-associated pathologies, including obesity, inflammatory tissue degeneration, dysplasia, and spontaneous tumor formation. This was accompanied by global upregulation of oxidative phosphorylation and glycolysis transcriptomes, causing deregulated mitochondrial respiration, oxidative stress, impaired cell proliferation, and a switch to glycolytic metabolism in vivo. These data identify TRAP-1 as a central regulator of mitochondrial bioenergetics, and this pathway could contribute to metabolic rewiring in tumors.

Figure 1. Characterization of TRAP-1^−/− mice

(A) Map of the gene trapping construct and position of the β-galactosidase cassette inserted in the mouse TRAP-1 locus. (B) PCR genotyping of TRAP-1 wild type (WT, +/+), heterozygous (+/−) and homozygous (−/−) mice. The position of WT and LacZ alleles is indicated. (C) The indicated tissue extracts from WT or TRAP-1 knockout (KO) mice were analyzed by Western blotting. (D) WT or TRAP-1 KO mice were analyzed for changes in body weight at the indicated ages (mo, months). Data are represented as box plots. Ages <3 mo (top), WT, n=16; KO, n=15; Ages>12 mo (bottom), WT, n=16; KO, n=26. *, p=0.033–0.022. (E) Heatmap of changes in protein expression and/or phosphorylation in liver extracts isolated from WT or TRAP-1 KO mice (A–C, replicates; D, average), as determined by Reverse Phase Protein Array (RPPA). Blue, downregulated; red, upregulated. Only statistically significant changes (p<0.05) are shown. (F) Liver extracts from WT or TRAP-1 KO mice were analyzed by Western blotting. M, male; F, female.

Figure 2. TRAP-1 regulation of mitochondrial oxidative phosphorylation

(A) Mitochondria isolated from WT or KO hepatocytes were analyzed for Complex I activity. The quantification of the slope (s) per each reaction is indicated. Representative experiment. (B) Quantification of Complex I specific activity in WT or KO mitochondria. Mean±SEM (n=3). ns, not significant (p=0.67). (C, F, H) Representative experiments of mitochondrial Complex II (C), Complex II+III (F) or Complex IV (H) activity in mitochondria isolated from WT or KO hepatocytes. The quantification of the slope (s) per each reaction is indicated. (D, G, I) Quantification of mitochondrial Complex II (D), Complex II–III (G) or Complex IV (I) specific activity in WT or KO hepatocytes. Data per each mitochondrial Complex activity were normalized against citrate synthase activity. Mean±SD (n=3). ns, not significant (p=0.66). *, p=0.027–0.033. (E) Representative experiment of mitochondrial Complex II activity in the presence or absence of mitochondrial Hsp90 inhibitor, Gamitrinib (Gam) The slope of individual curves is as follows, WT-Gam, s=−0.0005; WT+Gam, s=−0.0004; KO-Gam, s=−0.0005; KO+Gam, s=−0.0005. (J) WT or TRAP-1 KO hepatocytes (top) or mouse embryonic fibroblasts (MEFs) (bottom) were analyzed for oxygen consumption. Data were normalized for cell number by direct counting and cell viability by a fluorescence reporter. Mean±SEM of replicates of a representative experiment. *, p=0.015–0.036.

Figure 3. Glycolytic reprogramming induced by TRAP-1 deficiency

(A and B) Mitochondria isolated from WT or TRAP-1 KO hepatocytes (top) or MEFs (bottom) were analyzed for glucose consumption (A) or lactate generation (B). Data were normalized for cell number by direct cell counting and cell viability by a fluorescence reporter. Mean±SEM of replicates. ***, p=0.0001–0.0009; **, p=0.0057. (C) Whole body PET/CT images of a WT (left) and TRAP-1 KO (right) mouse obtained 1 h post ¹⁸F-FDG injection. Three times greater uptake of radioactivity in the liver of the TRAP-1 KO mouse represents increased glucose consumption. SUV, standardized uptake values. (D) Quantification of hepatic standardized uptake values (SUVs) of ¹⁸F-FDG as determined by PET/CT analysis in WT or TRAP-1 KO mice. Each point corresponds to an individual mouse. *, p=0.03 by Kolmogorov-Smirnov two-sample test. (E) Mitochondria isolated from WT or TRAP-1 KO hepatocytes (left) or MEFs (right) were analyzed for ATP production. Mean±SEM of replicates. **, p=0.0025. (F and G) WT or TRAP-1 KO MEFs were transfected with vector or Myc-TRAP-1 cDNA, and analyzed by Western blotting (F) or glucose consumption (G). Mean±SEM of replicates of a representative experiment. ***, p<0.0001.

Figure 4. TRAP-1 deficiency induces low level oxidative damage and cell cycle defects

(A) WT or TRAP-1 KO MEFs were analyzed for ROS production by flow cytometry. Data are expressed as mean geometric fluorescence. Mean±SEM of replicates of a representative experiment. ***, p=0.0001. (B and C) WT or TRAP-1 KO MEFs were analyzed for γH2AX reactivity by fluorescence microscopy (B), and quantified as normalized mean fluorescence (C). Nuclei were stained with DAPI. None, untreated. Etop, etoposide. Magnification, x60. ***, p=0.0003. (D) WT or TRAP-1 KO MEFs were analyzed for cell proliferation at the indicated time intervals by direct cell counting. Mean±SEM of replicates. Representative experiment. (E) WT or TRAP-1 KO MEFs were stained with propidium iodide and analyzed for DNA content by flow cytometry. The percentage of cells in the various cell cycle transitions is indicated. (F) WT or TRAP-1 KO MEFs were analyzed by Western blotting.