Steroid receptor coactivator 1 is an integrator of glucose and NAD+/NADH homeostasis

Abstract

Steroid receptor coactivator 1 (SRC-1) drives diverse gene expression programs necessary for the dynamic regulation of cancer metastasis, inflammation and gluconeogenesis, pointing to its overlapping roles as an oncoprotein and integrator of cell metabolic programs. Nutrient utilization has been intensely studied with regard to cellular adaptation in both cancer and noncancerous cells. Nonproliferating cells consume glucose through the citric acid cycle to generate NADH to fuel ATP generation via mitochondrial oxidative phosphorylation. In contrast, cancer cells undergo metabolic reprogramming to support rapid proliferation. To generate lipids, nucleotides, and proteins necessary for cell division, most tumors switch from oxidative phosphorylation to glycolysis, a phenomenon known as the Warburg Effect. Because SRC-1 is a key coactivator responsible for driving a hepatic gluconeogenic program under fasting conditions, we asked whether SRC-1 responds to alterations in nutrient availability to allow for adaptive metabolism. Here we show SRC-1 is stabilized by the 26S proteasome in the absence of glucose. RNA profiling was used to examine the effects of SRC-1 perturbation on gene expression in the absence or presence of glucose, revealing that SRC-1 affects the expression of complex I of the mitochondrial electron transport chain, a set of enzymes responsible for the conversion of NADH to NAD(+). NAD(+) and NADH were subsequently identified as metabolites that underlie SRC-1's response to glucose deprivation. Knockdown of SRC-1 in glycolytic cancer cells abrogated their ability to grow in the absence of glucose consistent with SRC-1's role in promoting cellular adaptation to reduced glucose availability.

Figure 1.

SRC-1 is responsive to nutrient status in more invasive cells. a and b, Different cancer cell lines were grown in the absence or presence of glucose. Whole-cell lysates were then subjected to immunoblot analysis for the indicated proteins. c, A549 cells were grown in the absence or presence of glucose for the indicated time. Whole-cell lysates were then subjected to immunoblot analysis for the indicated proteins. d, A549 cells were grown in the absence or presence of glucose. The mRNA expression of p160 family members was measured by qPCR.

Figure 2.

The glucose responsiveness of SRC-1 is reversible and is not dependent on ROS but is dependent on nutrient availability. a, A549 cells were grown in the absence or presence of glucose. These cells were then split and grown for 4 passages in the corresponding medium. b, A549 cells passaged without glucose were subsequently changed into medium in the absence or presence of glucose. c, A549 cells were grown in the absence or presence of glucose, with a ROS inhibitor N-acetylcysteine (NAC) or a vehicle control. d, A549 cells were grown under the indicated metabolite treatment. Where indicated, cells were treated with 4.5 g/L glucose, 110 mg/L pyruvate, 4.5 g/L fructose, or 50 mM 2-DG (2-deoxyglucose) for 24 hours. Following all treatments whole-cell lysates were then subjected to immunoblot analysis for the indicated proteins.

Figure 3.

SRC-1 protein accumulation is related to cellular localization and thereby protein degradation. a, A549 cells were grown in the absence or presence of glucose. For each nutrient condition, the cells were subjected to dimethylsulfoxide (DMSO) or MG132 treatment (10 μM). Whole-cell lysates were then subjected to immunoblot analysis for the indicated proteins. b, A549 cells were grown in the absence or presence of glucose for 24 hours. For each nutrient condition, the cells were subjected to immunofluorescence. Scale bars, 10 μm. c, A549 cells were grown in the presence or absence of glucose for 10 or 24 hours. Cells were subsequently fractioned for the cytoplasmic and nuclear fractions and then subjected to immunoblot analysis for the indicated proteins. DAPI, 4′,6-diamidino-2-phenylindole.

Figure 4.

Microarray analysis and validation of SRC-1's role in reacting to glucose deprivation in A549 cells. Fourteen hours after glucose deprivation, (a) RNA was harvested for qPCR quantification of SRC-1 mRNA levels and (b) whole-cell lysates were then subjected to immunoblot analysis for the indicated proteins. c, Microarray analysis of glucose-responsive genes in response to nontargeting siRNA (siCntrl) or siSRC-1. A549 cells with siCntrl or siSRC-1 were grown in the absence or presence of glucose, RNA was harvested for microarray analysis 14 hours after glucose deprivation. d, Comparison of differential glucose-responsive genes from the microarray analysis with siCntrl or siSRC-1. Values used for the scatter plot were generated by the following equation: value = (RNA levels of siCntrl with glucose − RNA levels of siCntrl without glucose) − (RNA levels siSRC-1 with glucose − RNA levels siSRC-1 without glucose) / (RNA levels siCntrl with glucose − RNA levels siCntrl without glucose). e, Comparison of mitochondrial electron transport chain complex I genes from the microarray analysis. f, Ingenuity Pathway Analysis under glucose deprivation of nontargeting siRNA and siSRC-1. g, A549 cells with siCntrl or siSRC-1 were grown in the absence or presence of glucose. RNA was harvested for qPCR quantification of ND6 mRNA expression 14 hours after glucose deprivation. h, A549 cells were grown under high-glucose DMEM, with or without 10 mM NAM or no-glucose DMEM for 24 hours. DNA was harvested and mtDNA was quantified. *, P < .05; **, P < .01; and ***, P < .005.

Figure 5.

Effect of metabolic inhibitors on SRC-1 protein levels. a, A549 cells were grown in the absence or presence of glucose. Cells grown under glucose-containing medium were then treated with rotenone, oligomycin A, or 2-deoxyglucose (2-DG) at different concentrations. b, A549 cells grown in glucose-containing medium were treated with rotenone, carbonyl cyanide 3-chlorophenylhydrazone (CCCP), or combinations of both compounds as indicated. c, A549 cells grown in glucose-containing medium were treated with NAM or FK866 at the indicated concentration. d, PC-3 cells were grown in glucose-containing medium and treated with rotenone, oligomycin A, or 2-DG at the indicated. e, PC-3 cells grown in glucose-containing medium were treated with NAM at the indicated concentrations. For all treatments described, whole-cell lysates were then subjected to immunoblot analysis for the indicated proteins. f, NADH measurement of A549 cells grown in the absence or presence of glucose.

Figure 6.

SRC-1 expression levels are strongly correlated to complex I subunit mRNAs but not complex II subunits mRNAs. a, The mRNA expression profile of SRC-1 and complex I in a number of cancer types. Samples were clustered from low to high SRC-1 levels (top to bottom). b, The mRNA expression profile of SRC-1 and complex II in a number of cancer types. Samples were clustered from low to high SRC-1 levels (top to bottom). Green represents low expression, and red represents high expression.

Figure 7.

SRC-1 is essential for survival or growth in response to glucose deprivation. a, A549 cells, treated with siControl or siSRC-1, were grown in the absence or presence of glucose. Images of cells were obtained 24 hours after treatment. b, Cells from Figure 1a were stained using crystal violet, resuspended with sodium dodecyl sulfate, and quantified. **, P < .01. c, A549 cells were grown in the presence of glucose, with gossypol and/or NAM as indicated. The attached cells were stained with crystal violet 24 hours after treatment. DMSO, dimethylsulfoxide; 2-DG, 2-deoxyglucose.