Introduction
70% of all breast cancers are estrogen receptor α positive (ER+) and are treated with endocrine therapies (antiestrogens or aromatase inhibitors) that disrupt the ER function. The antiestrogens Tamoxifen (TAM) antagonizes estrogen binding to the ER while ICI 182,780 (ICI; Faslodex/Fulvestrant) targets ER for degradation. Despite their clear clinical activity, 50% of ER + tumors never respond or eventually develop resistance to antiestrogens [
1,
2]. Understanding the molecular basis of endocrine resistance is a prerequisite to finding new interventions to resistance in the clinic.
c-MYC (hereafter referred to as MYC) is a transcription factor that is frequently deregulated in human cancers. MYC contributes to cancer progression through its involvement in several cellular functions including cell cycle progression, proliferation, differentiation, and apoptosis [
3‐
6]. MYC is overexpressed in 30-50% of high-grade breast tumors [
7,
8]. Activation of MYC is implicated in hormone-independence
in vitro and endocrine resistance in patients [
9], and it is predictive of a shorter time to recurrence following adjuvant TAM therapy [
10]. The oncogenic activity of MYC depends on its ability to dimerize with MAX [
11,
12]. Thus, agents that disrupt MYC-MAX heterodimers might be useful in treating some antiestrogen resistant breast cancers.
MYC controls several genes that regulate glycolysis and glutaminolysis [
13,
14]. Both normal and cancer cells use glucose and glutamine to generate energy (ATP), produce raw materials for the synthesis of amino acids, fatty acids, and nucleosides, and maintain redox balance. However, rapidly growing cancer cells demand higher levels of substrates for macromolecule synthesis and for maintaining redox balance [
15,
16]. Whether MYC can regulate cellular metabolism in antiestrogen resistant cancers, and whether this is a key component of this phenotype, remain unknown.
We describe how MYC upregulation in ER + antiestrogen resistant breast cancer cells increases dependency on glucose and glutamine but enables cell survival in glucose-deprived conditions by increasing dependency on glutamine. We show that glutamine in glucose-deprived conditions triggers the UPR through glucose-regulated protein-78 (GRP78/HSP5A/BiP) and inositol-requiring enzyme-1α (IRE1α/αRΝ1), and simultaneously, activates both pro-death and pro-survival pathways by increasing c-Jun N-terminal kinase (JNK) activation and spliced X-box protein-1 XBP1(s), respectively. While this UPR promotes apoptosis in most resistant cells in the short-term (72 h), in the longer term (>72 h), cell survival is promoted through cellular adaption to glutamine-only conditions in a minority of the cells that show adjusted MYC levels. Thus, safely targeting glutamine metabolism is a promising strategy to treat MYC-driven antiestrogen resistant breast cancer.
Discussion
MYC is a target of estrogen signaling in breast cancer cells [
26] that can control diverse aspects of cancer cell survival including cellular metabolic reprogramming [
47‐
49]. Activation of MYC has been linked to acquired antiestrogen resistance in human breast tumors [
10] and poor clinical outcome [
50]. Our findings show that MYC-driven pro-survival signaling in antiestrogen resistant breast cancer is partially dependent on proteins that control the cell cycle and apoptosis. While rapid drug metabolism limits the efficacy of 10058-F4 as an antitumor agent for solid tumors [
28], its use
in vitro showed that inhibiting MYC in antiestrogen resistant breast cancer cells confirmed the essential role of MYC activation in driving this phenotype. Metabolically stable small-molecule inhibitors of MYC hold significant promise as new agents to treat some drug resistant breast tumors.
MYC is an important regulator of glutamine and glucose metabolism [
51]. Antiestrogen resistant breast cancer cells with higher MYC activation showed increased sensitivity to small molecule inhibitors of glutaminolysis and glycolysis (Figure
5C), but did not re-sensitize these cells to antiestrogens. Thus, activation of these metabolic pathways in resistant cells may be independent of ER-mediated signaling. Increased levels of glutamate and proline in antiestrogen resistant breast cancer cells imply an essential role for glutamine metabolism in sustaining cell survival. Glutamate, which is converted from glutamine by GLS, is an essential substrate for many cellular processes including for the formation of the antioxidant glutathione (GSH), feeding into the tricarboxylic acid (TCA) cycle via its metabolism to α-ketoglutarate (α-KG), indirect generation of NADPH for the synthesis of fatty acids and nucleotides, and a key source of the ammonia that is required for acid–base homeostasis [
52,
53]. Conversely, a steady supply of glutamine is essential for cancer cells to modify proteins by O-linked N-acetylglucosamine (O-GlcNAc) through the hexosamine biosynthesis pathway. MYC can regulate global O-GlcNAc modification of proteins in rat fibroblast cells [
54]. A fraction of glutamine is also used as the nitrogen donor for the
de novo synthesis of purines and pyrimidines, needed to match the demands of nucleic acid production during cell proliferation, the rate of which is often greater in drug resistant cancer cells [
52,
55]. Regulation of the GLS/GAC-GLUL system by MYC in antiestrogen resistant cells may, therefore, be essential to maintain and/or drive the resistant phenotype. MYC regulation of GLS and GLUL in antiestrogen resistant breast cancer cells was unexpected. While in prostate cancer cells, MYC knockdown was shown to decrease GLS and increase GLUL protein levels [
56], in our antiestrogen resistant breast cancer cell models (LCC9, LCC2, and LY2) we observed the reverse effect – MYC knockdown increased GLS and decreased GLUL protein levels (Figure
5E and F).
The UPR pathway is an evolutionarily conserved adaptive pathway coupled to endoplasmic reticulum stress that is upregulated in antiestrogen resistant breast cancer [
37]. Previously, we have shown that GRP78, a member of the HSP70 family of proteins, is overexpressed in antiestrogen resistant breast cancer cells and tumors and promotes their survival [
23]. To date, it is unclear how the UPR regulates cellular metabolism or vice versa. Our findings show that GRP78, IRE1α, phospho-JNK and XBP1(s) are robustly upregulated in antiestrogen resistant ER + breast cancer cells in the presence of glutamine but absence of glucose (Figure
8A). While blocking JNK activation significantly reduced inhibition of cell growth in glutamine-only conditions, knockdown of XBP1 significantly increased the inhibition of cell growth (Figure
9). MYC directly inhibited phospho-JNK in glutamine-only conditions (Figure
8B). JNK or stress activated protein kinases (SAPK) belong to the MAPK family of proteins [
57] and can directly contribute to pro-apoptotic signaling by phosphorylating and inactivating BCL2. In contrast, MYC inhibited IRE1α expression similarly in all four conditions of glucose and glutamine availability. Thus, regulation of JNK by MYC may reflect a mechanism to regulate the UPR under specific cellular stresses.
JNK can regulate MYC through phosphorylation [
58] and can associate with and mediate MYC ubiquitination and degradation [
59]. Moreover, in HeLa and HEK293 cells, MYC knockdown decreased LC3II levels and decreased formation of autophagosomes by inhibiting JNK [
60]. In our endocrine resistant breast cancer cell models, MYC inhibition increased both JNK activation and LC3II levels, with an associated increased inhibition of cell growth in glutamine-only conditions (Figure
8B; Figure
9M). Further studies are needed to investigate how MYC controls stress signaling mediated through JNK and cell death pathways. Autophagosome formation and the accumulation of p62/SQSTM1 (Figure
7B) can trigger cell death through apoptosis during cellular stress [
34], likely reflecting the inability to use autophagosome content degradation to feed intermediate metabolism. Thus, cellular metabolic status are clearly important in triggering specific MYC-mediated functions. Within a tumor, cancer cells can experience glucose deprivation due to an inadequate vasculature [
61] or drug treatment [
62]. Short-term inhibition of glycolysis may initiate UPR-mediated responses that subsequently induce apoptosis in most cells but can also promote survival in a small fraction of cells until an adequate energy supply becomes available to enable both cell survival and proliferation. Indeed, in bortezomib-induced cell death, MYC has been shown to bind to pro-apoptotic BCL2 proteins, NOXA and BIM, and cooperate with EGR1 [
63]. Thus, MYC induced cell death in cancer cells warrants further elucidation.
Increased activation of MYC in antiestrogen resistant cells is also associated with their increased dependence on glutamine and glucose for cell survival. However, the presence of glutamine in glucose deprived conditions initiated an UPR-mediated pathway that killed most cells via apoptosis but allowed the survival of a small minority. In LCC9Gln cells, which survived in media containing glutamine but no glucose, MYC levels were reduced and GLS/GAC levels were increased when compared with the parental antiestrogen resistant LCC9 cells. These adaptations may ensure the appropriate balance between the levels of glutamine versus glutamate needed for the cells to survive in glucose-deprived conditions. Glutamine alone can sustain survival of a small cell population in the absence of glucose, albeit with a significantly decreased rate of cell proliferation (Figure
10A). Molecular characterization of the multiple passages of LCC9Gln versus parental cells is underway and will help elucidate the MYC-mediated and UPR-regulated adaptive pathway.
Excessive systemic energy demand in cancer can lead to cachexia, which affects a large number of cancer patients and results in the progressive loss of muscle and adipose tissue mass [
64]. To date, it is unclear how therapeutic interventions can safely alter the energy demand of cancer cells within tumors without necessarily inducing additional metabolic problems for the host. While a tumor-to-liver Cori cycle is implicated in meeting glucose demands, a tumor-to-muscle cycle is implicated in meeting the glutamine demands of growing tumors [
52,
64,
65]. In addition, fibroblasts in the tumor stroma can also supply tumor cells with glutamine [
66]. As cancer progresses to a more aggressive, metastatic, drug resistant phenotype, the potential to induce cachexia likely also increases. Understanding the adaptation of cellular metabolism associated with drug resistant disease may offer new interventions to address this co-morbidity evident in many advanced cancers.
MYC expression is deregulated in various cancer types. Our findings show that antiestrogen resistant breast cancer cells express higher levels of MYC protein compared with sensitive cells, and elevated MYC levels correlate with increased sensitivity to deprivation of glutamine and glucose. While the levels of glutamine metabolites are higher in resistant cells, MYC regulates GLS/GAC and GLUL to meet the demands of the resistant phenotype, particularly during periods of glucose deprivation/insufficiency. Thus, glutamine metabolism may allow cancer cells to adapt to changes in glucose availability by re-programming existing pathways through MYC and the UPR. Safely targeting the glucose or glutamine pathway and/or the UPR could offer novel strategies to treat antiestrogen resistant breast cancer.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
ANSH and RC contributed to concept design, helped conceive the study, and coordinated the overall project. ANSH also drafted and edited the manuscript with RC. KLC carried out the immunohistochemistry. JLSR, AEE, DMD and COBF carried out the cell-based assays. AW and LHC conceived and carried out the animal studies. All authors read and approved the final manuscript.