br These authors contributed equally Lead Contact Correspond
7These authors contributed equally 8Lead Contact *Correspondence: [email protected] (V.I.S.), [email protected] (M.O.B.)
For tumors to progress efficiently, cancer 1236188-16-1 must overcome barriers of oxidative stress. Although die-tary antioxidant supplementation or activation of endogenous antioxidants by NRF2 reduces oxidative stress and promotes early lung tumor progression, little is known about its effect on lung cancer metas-tasis. Here, we show that long-term supplementation with the antioxidants N-acetylcysteine and vitamin E promotes KRAS-driven lung cancer metastasis. The antioxidants stimulate metastasis by reducing levels of free heme and stabilizing the transcription factor BACH1. BACH1 activates transcription of Hexokinase 2 and Gapdh and increases glucose uptake, glycolysis rates, and lactate secretion, thereby stimulating glycolysis-dependent metas-tasis of mouse and human lung cancer cells. Target-ing BACH1 normalized glycolysis and prevented antioxidant-induced metastasis, while increasing endogenous BACH1 expression stimulated glycol-ysis and promoted metastasis, also in the absence of antioxidants. We conclude that BACH1 stimulates glycolysis-dependent lung cancer metastasis and that BACH1 is activated under conditions of reduced oxidative stress.
Cancer cells produce high levels of reactive oxygen species (ROS) through alterations in signaling and metabolic path-ways. For decades, the belief that ROS stimulate tumor initia-tion and progression has prompted healthy people and cancer patients alike to supplement their diets with antioxidants;
however, many randomized clinical trials have disproved this strategy (Klein et al., 2011; Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group, 1994; van Zandwijk et al., 2000). Instead, it has become clear that ROS create barriers to tumor progression that antioxidant supplements can help cancer cells overcome (Le Gal et al., 2015; Gorrini et al., 2013; Piskounova et al., 2015; Sayin et al., 2014; Wang et al., 2016).
In the absence of exogenous antioxidants, cancer cells main-tain redox homeostasis by expressing endogenous antioxi-dants, many of which are regulated by the redox-sensitive tran-scription factor nuclear factor (erythroid-derived 2)-like 2 (NFE2L2, NRF2) and its negative regulator Kelch-like ECH-associated protein 1 (KEAP1) (DeNicola et al., 2011; Singh et al., 2006). Although ROS reduction is often beneficial for tumor cells, specific forms of ROS can stimulate pro-tumori-genic signaling pathways by oxidizing and inactivating tumor suppressor proteins and activating oncogenes by oxidizing DNA (Finkel, 2011; Sabharwal and Schumacker, 2014; Wein-berg et al., 2010). The anti- and pro-tumorigenic roles of ROS complicate the understanding of redox regulation in cancer cells, and these topics need to be studied further (Chio and Tu-veson, 2017).
Recent studies provide compelling evidence that ROS have anti-tumorigenic roles and consequently that both exogenous and endogenous antioxidants act in a pro-tumorigenic fashion. For example, antioxidant supplementation of the diet accelerates tumor progression in endogenous and pa-tient-derived xenograft-based mouse models of lung cancer and malignant melanoma (Le Gal et al., 2015; Piskounova et al., 2015; Sayin et al., 2014; Wang et al., 2016). The first of these studies showed that supplementing the diet with
the antioxidants N-acetylcysteine (NAC) and vitamin E accel-erates primary lung tumor progression in Kras2LSL/+ (K) mice
(Sayin et al., 2014). The antioxidants reduced ROS and DNA damage in tumors and eliminated the expression of p53—a tumor suppressor normally activated by DNA damage. These
Figure 1. The Antioxidants NAC and Vitamin E Increase Lung Cancer Metastasis Independently of p53
(A) Schematic showing that Kras2LSL/+ (K) and Kras2LSL/+;Trp53fl/fl (KP) mice were allowed to inhale a low dose of Cre-adenovirus. One week later, NAC (1 g/L) was administered in the drinking water and vitamin E (VitE, 0.5 g/kg) in the chow.
(B) Percentage of K mice with lymph node metastases. Numbers in bars indicate numbers of mice.
(C) Immunohistochemical staining for pro-surfactant protein C (pro-SPC) in a normal lymph node of a control K mouse (upper) and in an enlarged metastatic lymph node from an NAC-treated K mouse (lower). Scale bar, 50 mm. (D) Left, percentage of K mice with distant metastases at necropsy. Right, photos of metastases on kidney (top) and liver (bottom).
(E) Percentage of KP mice with lymph node metastases.
(F) Left, percentage of KP mice with thoracic metastases. Right, photo of thoracic rib cage metastases.