PEPCK-M recoups tumor cell anabolic potential in a PKC-ζ-dependent manner

Abstract

Background: Mitochondrial phosphoenolpyruvate carboxykinase (PEPCK-M; PCK2) is expressed in all cancer types examined and in neuroprogenitor cells. The gene is upregulated by amino acid limitation and ER-stress in an ATF4- dependent manner, and its activity modulates the PEP/Ca2+ signaling axis, providing clear arguments for a functional relationship with metabolic adaptations for cell survival. Despite its potential relevance to cancer metabolism, the mechanisms responsible for its pro-survival activity have not been completely elucidated. Methods: [U-13C]glutamine and [U-13C]glucose labeling of glycolytic and TCA cycle intermediates and their anabolic end-products was evaluated quantitatively using LC/MS and GC/MS in conditions of abundant glucose and glucose limitation in loss-of-function (shRNA) and gain-of-function (lentiviral constitutive overexpression) HeLa cervix carcinoma cell models. Cell viability was assessed in conjunction with various glucose concentrations and in xenografts in vivo. Results: PEPCK-M levels linearly correlated with [U-13C]glutamine label abundance in most glycolytic and TCA cycle intermediate pools under nutritional stress. In particular, serine, glycine, and proline metabolism, and the anabolic potential of the cell, were sensitive to PEPCK-M activity. Therefore, cell viability defects could be rescued by supplementing with an excess of those amino acids. PEPCK-M silenced or inhibited cells in the presence of abundant glucose showed limited growth secondary to TCA cycle blockade and increased ROS. In limiting glucose conditions, downregulation of PKC-ζ tumor suppressor has been shown to enhance survival. Consistently, HeLa cells also sustained a survival advantage when PKC-ζ tumor suppressor was downregulated using shRNA, but this advantage was abolished in the absence of PEPCK-M, as its inhibition restores cell growth to control levels. The relationship between these two pathways is also highlighted by the anti-correlation observed between PEPCK-M and PKC-ζ protein levels in all clones tested, suggesting co-regulation in the absence of glucose. Finally, PEPCK-M loss negatively impacted on anchorage-independent colony formation and xenograft growth in vivo. Conclusions: All in all, our data suggest that PEPCK-M might participate in the mechanisms to regulate proteostasis in the anabolic and stalling phases of tumor growth. We provide molecular clues into the clinical relevance of PEPCK-M as a mechanism of evasion of cancer cells in conditions of nutrient stress.