br Aerobic glycolysis a hallmark of progression and
Aerobic glycolysis, a hallmark of progression and metastasis, has not been associated with antioxidant effects or with BACH1 (Hanahan and Weinberg, 2011). We found that BACH1 binds to the Hk2 and Gapdh promoters after antioxidant administration, stimulating their expression and that of several other glycolytic genes, and increases glucose uptake, glycolysis rates, and lactate secretion. Moreover, in human lung cancer datasets, BACH1 is co-expressed with HK2. Interestingly, NRF2’s ability to increase tumor progression has been linked to glycolysis (Mit-suishi et al., 2012; Rojo de la Vega et al., 2018; Sayin et al., 2017; Zhang et al., 2018). We found that NRF2 activation, like antioxi-dant supplementation, increases Hk2 and Gapdh transcription, stimulates glycolysis, and increases migration of naive lung cancer cells in a BACH1-dependent fashion. Thus, a possible advantage for cancer cells harboring NRF2/KEAP1 mutations is activation of BACH1-induced glycolysis.
How aerobic glycolysis drives cancer cell invasion has been studied for decades, and, although there is no unifying hypothesis, several explanations have been proposed (Elia et al., 2018; Liberti and Locasale, 2016; Payen et al., 2016). By-products of glycol-ysis, such as methylglyoxal, stimulate cell motility by activating the epithelial-mesenchymal transition. Extracellular acidification by lactate triggers apoptosis of neighboring healthy cells and se-lects for low pH-resistant cancer cells; lactate also controls tumor cell adhesion to the extracellular matrix by modulating integrin-collagen interactions and by stimulating the secretion of matrix-degrading enzymes. Furthermore, aerobic glycolysis increases ATP levels; if this happens near filopodia, lamellipodia, and 846557-71-9 fibers, it could fuel the rapid cytoskeletal remodeling required for
(F) Migration of mTC and mTN cells (n = 3) incubated with 25 mM of the glycolysis inhibitor 3-bromopyruvate (3-BP).
(H) Migration of mTC and mTN cells incubated with 10 mM of the pyruvate dehydrogenase kinase inhibitor dicholoroacetate (DCA).
(I) Migration of cells incubated with 1 and 10 mM of the lactate secretion inhibitor AZD3965.
migration (De Bock et al., 2013). However, a simple explanation is that multiple mechanisms contribute to glycolysis-driven inva-sion. Although our results do not allow us to distinguish between these possibilities, they establish that antioxidant- and BACH1-induced invasion depends entirely on glycolysis and lactate secretion and is associated with high ATP production rates.
Our findings raise the possibility that targeting BACH1 or pro-teins up- or downstream might inhibit lung cancer metastasis. Such a strategy might be effective in patients with NRF2/ KEAP1 mutations, who would be expected to have tumors with high BACH1 levels. Targeting metastasis would make sense in conjunction with surgery or radiotherapy. One potential strat-egy would be to target glycolysis (Liberti et al., 2017; Porporato et al., 2011; Polanski et al., 2014). Indeed, the MCT-1 inhibitor AZD3965 blocked BACH1-induced lung cancer cell migration and reduced metastasis. The GAPDH inhibitor 3-BP reduced metastasis more efficiently than AZD3965 but has been associ-ated with severe side effects in some studies. Regardless, our results suggest that lung cancer cells with high BACH1 levels possess a strong metabolic liability—a finding that warrants further exploration.
Detailed methods are provided in the online version of this paper and include the following:
d KEY RESOURCES TABLE
d LEAD CONTACT AND MATERIALS AVAILABILITY d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Mice, Lung Cancer Experiments, and Antioxidant Administration
B Cell Culture and Regents d METHODS DETAILS
B Histology and Immunohistochemical Analyses B Immunohistofluorescence
B Migration and Invasion Assays B ROS Measurements
B GSH, NADH and NADPH Measurements B Free Heme Measurements
B Lentiviral Production and Transduction B Western Blotting
B Cell Viability Assay
B Extracellular Flux Measurements
B Total ATP, Glucose Uptake and Lactate Secretion
B RNA-Sequencing and Bioinformatics B Real-Time Quantitative PCR
B Chromatin Immunoprecipitation-Sequencing
B Luciferase Assays
d QUANTIFICATION AND STATISTICAL ANALYSIS d DATA AND CODE AVAILABILITY
Supplemental Information can be found online at https://doi.org/10.1016/j.
We thank O. Persson and E. Tu¨ksammel for help with mouse experiments, C. Karlsson for histology, S. Ordway for editing the manuscript, and The Feno core facility at the Department of Laboratory Medicine for histology. The study was supported by grants from the Knut and Alice Wallenberg Foundation, Sjo¨-berg Foundation, Strategic Research Program in Cancer at Karolinska Institu-tet, Center for Innovative Medicine, Swedish Cancer Society, Medical Research Council, and Children’s Cancer Fund to M.O.B.; The Swedish Soci-ety for Medical Research, Medical Research Council, and Wallenberg Center for Molecular and Translational Medicine to V.I.S.; and the Alex and Eva Wall-stro¨m Foundation to C.W. C.W. holds a Marie Slowdoska-Curie Individual Fellowship and a Swedish Cancer Society Fellowship.