FunHoP analysis reveals upregulation of mitochondrial genes in prostate cancer

Abstract

Mitochondrial activity in cancer cells has been central to cancer research since Otto Warburg first published his thesis on the topic in 1956. Although Warburg proposed that oxidative phosphorylation in the tricarboxylic acid (TCA) cycle was perturbed in cancer, later research has shown that oxidative phosphorylation is activated in most cancers, including prostate cancer (PCa). However, more detailed knowledge on mitochondrial metabolism and metabolic pathways in cancers is still lacking. In this study we expand our previously developed method for analyzing functional homologous proteins (FunHoP), which can provide a more detailed view of metabolic pathways. FunHoP uses results from differential expression analysis of RNA-Seq data to improve pathway analysis. By adding information on subcellular localization based on experimental data and computational predictions we can use FunHoP to differentiate between mitochondrial and non-mitochondrial processes in cancerous and normal prostate cell lines. Our results show that mitochondrial pathways are upregulated in PCa and that splitting metabolic pathways into mitochondrial and non-mitochondrial counterparts using FunHoP adds to the interpretation of the metabolic properties of PCa cells.

Fig 1. Up- and downregulated genes for various subcellular localizations in the cell.

All three datasets show that the number of upregulated genes (blue bars) is higher than the number of downregulated genes (orange bars) for proteins localized to mitochondria, whereas proteins with non-mitochondrial localization are more equally distributed.

Fig 2. The TCA cycle—mitochondrial version.

Fig 3. Glycolysis/gluconeogenesis—non-mitochondrial version.

Fig 4. Pyruvate metabolism–original version.

The FunHoP-expanded original pathway of Pyruvate metabolism is showing both up- and downregulated genes, independent of localization. The original picture changes when sorting the genes on localization into mitochondrial versus non-mitochondrial categories, and then splitting the pathway into a mitochondrial version (Fig 5) and a non-mitochondrial version (Fig 6).

Fig 5. Pyruvate metabolism–mitochondrial version.

Fig 6. Pyruvate metabolism–non-mitochondrial version.

Fig 7. Alanine, glutamine and glutamate metabolism–original version.

Fig 8. Alanine, glutamine and glutamate metabolism–mitochondrial version.

Fig 9. Alanine, glutamine and glutamate metabolism–non-mitochondrial version.