ABSTRACT
Hepatocellular carcinoma (HCC) is the most common form of liver cancer worldwide. Increasing evidence suggests that mitochondria play a central role in malignant metabolic reprogramming in HCC, which may promote disease progression. To comprehensively evaluate the mitochondrial phenotype present in HCC, we applied a recently developed diagnostic workflow that combines high-resolution respirometry, fluorometry, and mitochondrial-targeted nLC-MS/MS proteomics to cell culture (AML12 and Hepa 1-6 cells) and diethylnitrosamine (DEN)-induced mouse models of HCC. Across both model systems, CI-linked respiration was significantly decreased in HCC compared to nontumor, though this did not alter ATP production rates. Interestingly, CI-linked respiration was found to be restored in DEN-induced tumor mitochondria through acute in vitro treatment with P1, P5-di(adenosine-5') pentaphosphate (Ap5A), a broad inhibitor of adenylate kinases. Mass spectrometry-based proteomics revealed that DEN-induced tumor mitochondria had increased expression of adenylate kinase isoform 4 (AK4), which may account for this response to Ap5A. Tumor mitochondria also displayed a reduced ability to retain calcium and generate membrane potential across a physiological span of ATP demand states compared to DEN-treated nontumor or saline-treated liver mitochondria. We validated these findings in flash-frozen human primary HCC samples, which similarly displayed a decrease in mitochondrial respiratory capacity that disproportionately affected CI. Our findings support the utility of mitochondrial phenotyping in identifying novel regulatory mechanisms governing cancer bioenergetics.
ABSTRACT
Human disease pathophysiology commonly involves metabolic disruption at both the cellular and subcellular levels. Isolated mitochondria are a powerful model for separating global cellular changes from intrinsic mitochondrial alterations. However, common laboratory practices for isolating mitochondria (e.g., differential centrifugation) routinely results in organelle preparations with variable mitochondrial purity. To overcome this issue, we developed a mass spectrometry-based method that quantitatively evaluates sample-specific percent mitochondrial enrichment. Sample-specific mitochondrial enrichment was then used to correct various biochemical readouts of mitochondrial function to a 'fixed' amount of mitochondrial protein, thus allowing for intrinsic mitochondrial bioenergetics, relative to the underlying proteome, to be assessed across multiple mouse tissues (e.g., heart, brown adipose, kidney, liver). Our results support the use of mitochondrial-targeted nLC-MS/MS as a method to quantitate mitochondrial enrichment on a per-sample basis, allowing for unbiased comparison of functional parameters between populations of mitochondria isolated from metabolically distinct tissues. This method can easily be applied across multiple experimental settings in which intrinsic shifts in the mitochondrial network are suspected of driving a given physiological or pathophysiological outcome.
