Cardiomyocyte-specific deletion of GCN5L1 in mice restricts mitochondrial protein hyperacetylation in response to a high fat diet

Abstract

Mitochondrial lysine acetylation regulates several metabolic pathways in cardiac cells. The current study investigated whether GCN5L1-mediated lysine acetylation regulates cardiac mitochondrial metabolic proteins in response to a high fat diet (HFD). GCN5L1 cardiac-specific knockout (cKO) mice showed significantly reduced mitochondrial protein acetylation following a HFD relative to wildtype (WT) mice. GCN5L1 cKO mice did not display any decrease in ex vivo cardiac workload in response to a HFD. In contrast, ex vivo cardiac function in HFD-fed WT mice dropped ~ 50% relative to low fat diet (LFD) fed controls. The acetylation status of electron transport chain Complex I protein NDUFB8 was significantly increased in WT mice fed a HFD, but remained unchanged in GCN5L1 cKO mice relative to LFD controls. Finally, we observed that inhibitory acetylation of superoxide dismutase 2 (SOD2) at K122 was increased in WT (but not cKO mice) on a HFD. This correlated with significantly increased cardiac lipid peroxidation in HFD-fed WT mice relative to GCN5L1 cKO animals under the same conditions. We conclude that increased GCN5L1 expression in response to a HFD promotes increased lysine acetylation, and that this may play a role in the development of reactive oxygen species (ROS) damage caused by nutrient excess.

Figure 1

Characterization of mitochondrial acetylation in WT and GCN5L1 cKO animals. (A) Genotype and breeding scheme of WT and GCN5L1 cKO transgenic mice. (B) Diet and tamoxifen (Tam) injection schedule of WT and GCN5L1 cKO mice. (C, D) Overall acetylation of mitochondria isolated from cardiac tissues was greatly increased in HFD WT animals, which was significantly attenuated in HFD cKO mice. Values are expressed as means ± SD, n = 5, *P < 0.05 versus WT LFD, ^$P < 0.05 versus cKO LFD, ^#P < 0.05 versus WT HFD.

Figure 2

Impact of GCN5L1 deletion on physical features and cardiac workload ex vivo. (A, B) Exposure to a HFD led to similar significant increases in total body weight and heart weight: tibia length ratio in both WT and GCN5L1 cKO mice. (C) There was a significant, ~ 50% decrease in normalized ex vivo workload in WT HFD animals relative to WT LFD mice. This relative decrease in normalized ex vivo workload was not observed in GCN5L1 cKO mice. Values are expressed as means ± SD, n = 4–7, *P < 0.05 versus WT LFD, ^$P < 0.05 versus cKO LFD, ^#P < 0.05 versus WT HFD.

Figure 3

Acetylation status of mitochondrial electron transport chain proteins in response to a HFD. (A) Acetylated lysine immunoprecipitation from whole cardiac tissue showed an increase in the acetylation of NDUFB8 (Complex I) in WT HFD animals relative to GCN5L1 cKO mice under the same conditions. (B, C) There were no changes observed in the acetylation status of SDHB (Complex II) and UQCR2 (Complex III). Values are expressed as means ± SD, n = 4–5, *P < 0.05 versus WT LFD, ^#P < 0.05 versus WT HFD.

Figure 4

Impact of GCN5L1 deletion on ROS regulation and damage in response to a HFD. (A, B) Acetylation of superoxide dismutase 2 (SOD2) at K122 is significantly increased in whole cardiac tissue from WT animals in response to a HFD. There was no increase in SOD2 acetylation in GCN5L1 cKO mice under the same dietary conditions. (C) Cardiac lipid peroxidation in samples obtained from while cardiac tissue, a measure of ROS damage, is significantly lower in GCN5L1 cKO mice relative to WT mice under HFD conditions. Values are expressed as means ± SD, n = 4–8, *P < 0.05 versus WT LFD, ^#P < 0.05 versus WT HFD.