br Introduction br The early Warburg observations revealed
The early Warburg observations revealed that cancer Paxilline pre-ferably obtain the most of energy from glycolysis even in the aerobic condition (known as the Warburg eﬀect) . Cancer cells use onco-genes/oncometabolites to actively reprogram their metabolic ma-chinery to meet the increased anabolic demands for their abnormal growth [2, 3]. The importance of the Warburg eﬀect and other meta-bolic alterations, such as glutaminolysis, and lipogenesis, are also being increasingly recognized [4–8]. It is reported that such reprogrammed energy metabolism is one of the remarkable hallmarks of cancer .
In order to understand the complexity of cancer cell metabolism better, we investigated several small molecules that attack the specific metabolic weaknesses of cancer cells. Recently, various attractive mo-lecular targets, such as pyruvate kinase M2 and mutant IDH1/2, have been identified and their specific inhibitors have been developed [10–12]. Target-based approaches seem fruitful; however, small mole-cules obtained from in vitro assays are not always recapitulated in cel-lular systems. In contrast, cell-based phenotypic screening has multiple advantages [13–17]. In the present study, we focused on major meta-bolic properties (such as oxygen consumption rate (OCR) and
extracellular acidification rate (ECAR)) and protein expression levels during glycolysis. OCR and ECAR values are indicative of the eﬃcacy of oxidative phosphorylation (OXPHOS) and the amount of lactic acid, respectively, and thus, direct outcomes of changes in cellular bioener-getic systems, which include glycolysis, mitochondrial respiration, and other metabolic machineries. Formerly, OCR and ECAR could only be monitored independently using Clark electrodes  or a micro-physiometer [19, 20]; however, they can now be assessed with the Seahorse XFe96 analyzer, which can simultaneously measure these values in intact cells [21, 22]. Hence, the Seahorse XFe96 analyzer can be used to investigate compounds eﬀect on OCR and/or ECAR values. This helps to easily identify direct inhibitors of glycolysis/mitochon-drial respiration, as well as indirect inhibitors with unexpected modes of action.
ChemProteoBase is a unique target identification platform that we developed based on proteomic analysis using two-dimensional diﬀer-ence gel electrophoresis (2-D DIGE) . It contains changes in pro-teomes that are induced by various well-characterized compounds and can determine the targets of test compounds by profiling spot patterns. Recently, we reported on the use of targeted proteomics specific to metabolic pathways, such as glycolysis, to give a deep insight into the
Corresponding author at: Chemical Biology Research Group, RIKEN Center for Sustainable Resource Science, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan.
E-mail address: [email protected] (H. Osada). 1 These authors contributed equally to this work.
Y. Futamura et al.
biological eﬀect of collismycin A . It has been indicated that the inner mass of a cancer cell mass becomes hypoxic during the formation of the tumor. Furthermore, activation of hypoxia-inducible factor 1-alpha (HIF-1α) following exposure to hypoxia leads to upregulation of the expression of glycolytic enzymes at both transcriptional and pro-teomic levels [25, 26]. Iron chelators are able to induce a hypoxia-like eﬀect. Additionally, they induce HIF-1α activation by inhibiting the degradation of HIF-1α and upregulating the expression of glycolytic proteins [24, 27]. Moreover, environmental changes and drugs can induce intercellular proteomic changes in cancer cells.
Original natural product screening platforms, RIKEN Natural Products Depository (NPDepo) and a fraction library/NPPlot, have been previously constructed [28, 29]. The fraction library was created by systematically collecting chromatographically semi-purified samples of microbial fermentation broths/plant extracts and compiling NPPlot, a database of physico-chemical properties (retention time and UV/mass spectrometry data) of each detectable metabolite in the fractions. Using this system, we have comprehensively collected natural products with unique chemical structures and expanded the chemical space of NPDepo chemical library [30–32].
In the present study, we established a screening system for in-hibitors, searched for inhibitors of cancer-specific metabolism from NPDepo chemical library, and investigated the mechanisms of action of hit compounds using an in vitro reconstitution assay method for asses-sing mitochondrial respiration in digitonin-permeabilized cells.
2. Materials & methods