MYC regulates the unfolded protein response and glucose and glutamine uptake in endocrine resistant breast cancer

Abstract

Background: About 70% of all breast cancers are estrogen receptor alpha positive (ER+) and are treated with antiestrogens. However, 50% of ER + tumors develop resistance to these drugs (endocrine resistance). In endocrine resistant cells, an adaptive pathway called the unfolded protein response (UPR) is elevated that allows cells to tolerate stress more efficiently than in sensitive cells. While the precise mechanism remains unclear, the UPR can trigger both pro-survival and pro-death outcomes that depend on the nature and magnitude of the stress. In this study, we identified MYC, an oncoprotein that is upregulated in endocrine resistant breast cancer, as a regulator of the UPR in glucose-deprived conditions.

Methods: ER+ human breast cancer cell lines (LCC1, LCC1, LY2 and LCC9) and rat mammary tumors were used to confirm upregulation of MYC in endocrine resistance. To evaluate functional relevance of proteins, siRNA-mediated inhibition or small molecule inhibitors were used. Cell density/number was evaluated with crystal violet assay; cell cycle and apoptosis were measured by flow cytometry. Relative quantification of glutamine metabolites were determined by mass spectrometry. Signaling molecules of the UPR, apoptosis or autophagy pathways were investigated by western blotting.

Results: Increased MYC function in resistant cells correlated with increased dependency on glutamine and glucose for survival. Inhibition of MYC reduced cell growth and uptake of both glucose and glutamine in resistant cells. Interestingly, in glucose-deprived conditions, glutamine induced apoptosis and necrosis, arrested autophagy, and triggered the unfolded protein response (UPR) though GRP78-IRE1α with two possible outcomes: (i) inhibition of cell growth by JNK activation in most cells and, (ii) promotion of cell growth by spliced XBP1 in the minority of cells. These disparate effects are regulated, at different signaling junctions, by MYC more robustly in resistant cells.

Conclusions: Endocrine resistant cells overexpress MYC and are better adapted to withstand periods of glucose deprivation and can use glutamine in the short term to maintain adequate metabolism to support cell survival. Our findings reveal a unique role for MYC in regulating cell fate through the UPR, and suggest that targeting glutamine metabolism may be a novel strategy in endocrine resistant breast cancer.

Figures

Figure 1

MYC expression is elevated in antiestrogen resistant breast cancerin vitroandin vivo.A, Basal MYC-luciferase activity is 4.24-fold (SE = 0.10) higher in LY2 and LCC2 (estrogen independent but responsive; antiestrogen resistant) and 6.67-fold (SE = 0.09) higher in LCC9 (estrogen independent and non-responsive; antiestrogen resistant) compared with LCC1 (estrogen independent but responsive; antiestrogen sensitive); see Experimental Procedures. ANOVA, p < 0.001; *p < 0.05 for MYC promoter activation in indicated cells compared with LCC1 cells. B, Western blot shows increased expression of MYC protein in LCC9 cells compared to LCC1 cells while MAX protein levels did not change; actin was used as a loading control. C, Immunohistochemical (IHC) MYC staining show increased protein levels (brown) in LCC9 compared with LCC1 xenografts; for negative controls, antibody diluents without MYC antibody were used. D, DMBA-induced rat mammary gland tumors with acquired resistance to TAM show increased levels MYC protein levels (brown) compared to sensitive (or de novo resistant) tumors.

Figure 2

MYC promotes survival in antiestrogen resistant cells.A, Western blot, reduced MYC in LCC9 cells at 48 h with MYC siRNA compared to control siRNA. Actin is a loading control. B, Quantitation of MYC in in LCC9 cell show 60% reduction in MYC siRNA transfected cells compared with control siRNA. C, MYC siRNA interacted additively (RI = 1.11) with ICI in inhibiting cell number in LCC1 but not in LCC9 cells. Bars,mean ± SE of relative cell number (normalized to vehicle controls) for a representative experiment performed in sextuplicate. ANOVA, p < 0.001; *p < 0.05 for treatment versus control for respective cell lines. ^, p < 0.05 for LCC1 versus LCC9 cells with MYC siRNA + ICI. D, LCC9 cells showed increased sensitivity to 10058-F4compared with LCC1 cells at 48 h. Points, mean of cell number; bars, ±SE. E, 10058-F4 or ICI alone or the combination for 48 h inhibit cell number in LCC1. In LCC9 cells, RI = 1.51, suggest a modest synergistic interaction between ICI and 10058-F4; ANOVA, p < 0.001; *p < 0.05 for treatment versus control for respective cell lines. ^, p < 0.05 for ICI + 10058-F4 versus 10058-F4. F, Western blot show decrease in MYC, MAX, BCL2 and an increase in cleaved CASP7 , with 10058-F4 (MI: MYC inhibitor) or ICI + 10058-F4 (C: combination) compared with vehicle (V) alone or with ICI alone (I) treatment (48 h). G, Annexin V-FITC (apoptosis) in LCC1 and LCC9 cells with vehicle, ICI , 10058-F4 , or ICI + 10058-F4 (combination). ANOVA, p < 0.001; *p < 0.05 for indicated treatment versus vehicle control for respective cells lines. Paclitaxel, a positive control for apoptosis. H, Dot plots showcells positive for annexin-V-FITC (x-axis) and propidium iodide (PI; y-axis).

Figure 3

Combination of MYC inhibitor and antiestrogen increased G1 cell cycle arrest in endocrine resistant cells.A, Top, ICI (100 nM), 10058-F4 (25 μM), or the combination significantly increased percentage of cells in G1 arrest and reduced percentage of cells in S phase in LCC1 (p < 0.001). Bottom, Representative cell count plots for propidium iodide (PI) in LCC1 cells are shown. B, Top, Only the combination of ICI and 10058-F4 induced significant increase in G1 arrest in LCC9 cells (p < 0.001). Bottom, Representative cell count plots for PI in LCC9 cells are shown. Graphs represent data that are presented as the mean ± SE for three independent experiments. ANOVA, p < 0.001; *p < 0.001 for indicated treatment versus vehicle control.

Figure 4

Increased dependence on glutamine and glucose in antiestrogen resistant cells.A, Schematic for glutamine metabolism: glutamine is converted to glutamate by the mitochondrial enzyme, glutaminase (GLS); the reverse reaction is catalyzed by glutamate-ammonia ligase (GLUL). Glutamine is an essential substrate for the biosynthesis of proline. B-D, Relative quantification of glutamine, glutamate, and proline by UPLC-QqQLIT showed a significant increase in glutamate (p = 0.002) and proline levels (p = 0.032) in LCC9 cells when compared with LCC1 control cells; six biological replicates from each cell line was used and levels of each respective metabolite was normalized to total protein levels in each sample. E, Uptake of glucose is significantly increased in LCC9 cells compared with LCC1 cells under basal conditions (p < 0.05). Relative cellular metabolites and glucose uptake were compared using Student’s t test F-G, Inhibition of MYC using siRNA significantly deceased (F) glutamine (p < 0.05) and (G) glucose uptake (p = 0.011) in LCC9 compared with LCC1 cells. ANOVA, p < 0.001. H, Inhibition of MYC with siRNA decreased protein levels of transporters of glutamine (ASCT2/SLC1A5), glutamate (EAAT2/SLC1A2) and glucose (GLUT1/SLC2A1) in LCC9 cells. Western blot shown is representative of three independent experiments.

Figure 5

Glutamine and glucose metabolism is increased in antiestrogen resistant cells.A-B, LCC9 cells were significantly more sensitive to (A) compound-968, an inhibitor of GLS/GAC, and to (B) STF-31, an inhibitor of GLUT-1. Bars represent the mean ± SE of relative number (normalized to vehicle control) for a single representative experiment performed in sextuplicate. ANOVA, p ≤ 0.001; *p < 0.05 for LCC9 versus LCC1 for indicated concentrations. C, Cells were treated with compound-968 (20 μM), STF-31 (5 μM), ICI (100 nM), or the indicated combinations for 48 h. Bars represent the mean ± SE of relative cell number (normalized to vehicle controls) for a single representative experiment performed in sextuplicate. ANOVA, p < 0.001; *p < 0.05 for LCC9 versus LCC1 for indicated treatments. D, Knockdown of GLS levels with siRNA in LCC9 cells showed significant decrease in cell number within 24 h compared with that in LCC1 cells. ANOVA, p = 0.03; *p ≤ 0.05 for LCC9 GLS siRNA compared with LCC1 GLS siRNA. E, Western blot showing decreased levels of GLS in both cell lines; actin was used as a protein loading control. F, Right, LCC9 ells were treated with 10058-F4 (25 μM), or vehicle for 48 h; left, transfected with MYC or control siRNA for 48 h. Knockdown of MYC increased GLS/GAC levels and decreased GLUL levels. G, siRNA mediated MYC knockdown showed increase in GLS and a decrease in GLUL levels in LCC2 and LY2 cells.

Figure 6

MYC expression increases sensitivity to glucose and glutamine deprivation.A-B, Overexpression of MYC in LCC1 cells significantly increased (p < 0.01) (A) and knockdown of MYC in LCC9 cells (B) significantly decreased cell number in the absence of glucose and glutamine (p ≤ 0.001). (C-F) LCC1 and LCC9 cells were grown in complete (12 mM glucose; 2 mM glutamine), incomplete (no glucose; no glutamine), glucose only (12 mM glucose; no glutamine), and glutamine-only (2 mM glutamine; no glucose) for 72 h. Changes in cell growth rates were determined by normalizing cell numbers measurements at 24 h, 48 h, and 72 h to cell numbers measurements at 0 h. At 72 h, LCC9 cells showed significantly higher growth rate compared to LCC1 in complete media. However, growth rate was significantly reduced for LCC9 in incomplete media when compared with LCC1 cells. In glucose-only media (at 72 h), LCC1 and LCC9 cells did not show an increase in cell growth. In glutamine-only media, LCC9 cells showed a significant decrease in cell number relative to LCC1 cells. Dashed line denotes change in scales between the graphs. Bars represent the mean ± SE of relative number (normalized to vehicle control) for a single representative experiment performed in sextuplicate. G, Knockdown of MYC in LCC9 cells reduced sensitivity to incomplete media, as seen in B, and also reduced inhibition of cell number in the presence of 2 mM glutamine in glucose-deprived conditions. ANOVA, p < 0.001; p ≤ 0.01 for LCC9-MYC siRNA versus LCC9-control siRNA for indicated treatment. Bars represent the mean ± SE of relative number (normalized to vehicle control) for a single representative experiment performed in sextuplicate.

Figure 7

Glutamine induces apoptosis and arrests autophagy via the UPR in glucose-deprived conditions.A, Significantly higher levels of apoptosis were seen in LCC9 compared with LCC1 cells following treatment with 2 or 4 mM glutamine at 48 h. ANOVA, p < 0.05; *p < 0.05 for LCC9 versus LCC1 for indicated treatment. B, Time-course, 0, 24 and 48 h, analysis of the autophagosome-associated proteins LC3II (marker for autophagosome formation or enlargement) and p62/SQSTM1 (marker for autophagosome activity, degradation of cargo). Increased formation of autophagosomes but arrested cargo degradation was seen within 24 h in both LCC1 and LCC9 cells in glutamine only media (and in no-glucose + no glutamine) conditions at 24 and 48 h but not in glucose-only (or in glucose + glutamine) media. C, In presence of 2 or 4 mM glutamine at 48 h, LCC9 cells showed increased levels of MYC and MAX and LC3II but no change in SQSTM1/p62. D, Cellular levels of total reactive species (RS) was significantly elevated in LCC9 compared to LCC1 cells in incomplete media (ANOVA, p < 0.001; *p < 0.05 for LCC9 versus LCC1 with no glucose + no glutamine).

Figure 8

Glutamine in glucose-deprived conditions activates the UPR.A, Cells were plated at 70% confluence. 24 hr later, media was changed to 0, 2, or 4 mM glutamine alone or in presence of 12 mM glucose. Western blot analysis showed Increased levels of GRP78, IRE1a, phospho-JNK, CHOP and decreased levels of BCL2 were present in LCC1 (right) and LCC9 (left) cells in glutamine-only conditions. MYC protein levels were highest when both glucose and glutamine are present; MYC is undetectable when these metabolites are absent in the media. MYC expression in the presence of glutamine-only, but not in presence of glucose-only, conditions correlated with increased expression of UPR proteins. B, Knockdown of MYC for 24 h was followed by media change to either glucose + glutamine, glucose-only, glutamine-only or no glucose + no glutamine conditions for another 48 h. Western blots analysis showed that a decrease in MYC protein levels correlated with an increase in the UPR proteins IRE1α and phospho-JNK(Thr183/Ty4185), and the autophagosome formation marker LC3II, and the autophagosome cargo degradation marker p62/SQSTM1. GRP78 was also increased in glucose + glutamine, glucose-only and no glucose + no glutamine conditions but robust expression of GRP78 in glutamine-only conditions was not affected by MYC siRNA. Total levels of JNK did not change.

Figure 9

UPR in glutamine-only conditions can lead to both pro-survival and pro-death outcomes. Effect of transfection of siRNA targeting GRP78, IRE1α, XBP1(s), and MYC for 24 h; or JNK inhibition with a small molecule inhibitor (SP600125) on growth in either glucose + glutamine or glutamine-alone media. Western blot (48 h); A-C, GRP78. D-F, IRE1α. G-I, JNK. J-L, XBP1. M-O, MYC. Inhibition of GRP78 did not significantly further affect cell numbers in glutamine-only conditions in both LCC1 and LCC9 cell lines, A. Western blot analysis of total GRP78 protein are shown in both cell lines in different conditions, B-C. Knockdown of IRE1α, D-F and XBP1, J-L, significantly increased inhibition of cell growth in glutamine-only conditions in both cell lines. However, inhibition of JNK with SP600125 significantly decreased the inhibition of cell growth in glutamine-only conditions, G-I. Also, knockdown of MYC, M-O, significantly decreased inhibition of cell growth in glutamine-only conditions. Overall, MYC may have facilitate an IRE1α-XBP1 pathway to promote cell survival during glutamine-only conditions, and an IRE1α-phospho-JNK pathway to promote cell death in this condition. ANOVA, p ≤ 0.001; *p < 0.05 for respective cell lines transfected with indicated siRNA (or treated with SP600125, for JNK) compared with control siRNA (or vehicle alone, for JNK) in glutamine-only conditions.

Figure 10

MYC confers metabolic flexibility in antiestrogen resistant cells.A, Rate of cell growth was significantly reduced in LCC9Gln cells compared with LCC9 control cells (p ≤ 0.001). Cell numbers at 72 h were compared using Student’s t test. B, MYC, MAX and GLUL protein levels were reduced, while GLS/GAC was increased, in LCC9Gln cells compared with control. C, Schematic diagram illustrating the role of MYC in regulating glutamine metabolism in complete (right; basal; with glucose and glutamine) and in glutamine-only conditions (left; glutamine but no glucose). MYC regulates glutamine, glutamate, and glucose uptake through transporters, ASCT2, EAAT2 and GLUT1, respectively, under normal conditions. In glucose-deprived conditions, glutamine metabolism triggers the UPR and induces cell death (inducing apoptosis and arresting autophagy) via a MYC-regulated IRE1α-JNK-CHOP in the short-term (72 h), and also promotes cell survival, through a IRE1α-XBP1(s); the surviving cells grow at a slower rate of cell proliferation (A), at >72 h. Dashed line denotes presence of intermediate metabolites/proteins that are not addressed in this study.