Aspirin, the pro-drug of salicylate, is associated with reduced incidence of death from cancers of the colon, lung and prostate and is commonly prescribed in combination with metformin in individuals with type 2 diabetes. Salicylate activates the AMP-activated protein kinase (AMPK) by binding at the A-769662 drug binding site on the AMPK β1-subunit, a mechanism that is distinct from metformin which disrupts the adenylate charge of the cell. A hallmark of many cancers is high rates of fatty acid synthesis and AMPK inhibits this pathway through phosphorylation of acetyl-CoA carboxylase (ACC). It is currently unknown whether targeting the AMPK–ACC-lipogenic pathway using salicylate and/or metformin may be effective for inhibiting cancer cell survival. Salicylate suppresses clonogenic survival of prostate and lung cancer cells at therapeutic concentrations achievable following the ingestion of aspirin (<1.0 mM); effects not observed in prostate (PNT1A) and lung (MRC-5) epithelial cell lines. Salicylate concentrations of 1 mM increased the phosphorylation of ACC and suppressed de novo lipogenesis and these effects were enhanced with the addition of clinical concentrations of metformin (100 μM) and eliminated in mouse embryonic fibroblasts (MEFs) deficient in AMPK β1. Supplementation of media with fatty acids and/or cholesterol reverses the suppressive effects of salicylate and metformin on cell survival indicating the inhibition of de novo lipogenesis is probably important. Pre-clinical studies evaluating the use of salicylate based drugs alone and in combination with metformin to inhibit de novo lipogenesis and the survival of prostate and lung cancers are warranted.

INTRODUCTION

Daily aspirin ingestion reduces the risk of death from adenocarcinomas of the colon, lung and prostate [1,2]. Although the anti-tumour effects of aspirin are widely believed to involve its anti-inflammatory properties, most studies have found that other non-steroidal anti-inflammatory (NSAIDs) [35] or anti-platelet [1] therapies do not have the same anti-cancer properties indicating that other mechanisms may be important. Upon ingestion, aspirin is rapidly broken down into salicylate by carboxlyesterases, which enhances its relative effective circulating concentration and half-life [6]. We have shown that salicylate activates the AMP-activated protein kinase (AMPK) [7,8] a cellular energy sensor composed of an α catalytic subunit and regulatory β- and γ-subunits, through a mechanism requiring Ser108 of the AMPK β1-subunit. In the past decade AMPK has become a widely studied target for anti-cancer therapies due to its inhibitory effects on the mammalian target of rapamycin (mTOR) pathway which is commonly mutated and constitutively active in many adenocarcinomas [9]. AMPK inhibits mTOR through phosphorylation and inhibition of regulatory associated protein of mTOR (Raptor) [10] and through activating phosphorylation of tuberous sclerosis complex 2 (TSC2) [11], which inhibit Ras homolog enriched in brain-guanosine triphosphate (Rheb-GTP), an upstream activator of mTOR complex 1 (mTORC1) [12]. Salicylate has been shown to reduce the activity mTOR/S6 kinase in HCT116 colon cancer cells, an effect consistent with AMPK activation [1315]; however higher concentrations have been shown to be independent of AMPK activation [16]. The effects of aspirin/salicylate on AMPK and mTOR signalling in other cancer cell types have not yet been studied.

Like aspirin, metformin is a widely prescribed therapeutic that has been correlated with a life-long reduction in cancer development [1725]. Despite extensive research over the last decade and multiple potential targets, the exact mechanism by which metformin may have anti-neoplastic activity is still not known [25]. However, previous evidence suggests that effects may be linked to the inhibition of mitochondrial respiration [26] an idea consistent with the well-established action of metformin to activate AMPK by reducing the adenylate charge of the cell [27,28]. Importantly, this mechanism of AMPK activation is entirely distinct from the direct actions of salicylate on the AMPK β1-subunit [8]. A major caveat of many studies which have utilized metformin to inhibit cancer growth is that millimolar (mM) concentrations have been used, despite maximum concentrations observed clinically being 50–100 μM [25]. Although, an intensive investigation of metformin in cancer therapy is currently underway, strategies aimed at reducing the dose of metformin required for clinical effectiveness are important to maximize the anti-neoplastic activity of this well tolerated therapeutic. Combination therapy of metformin with salicylate may be a promising avenue to achieve this.

De novo synthesis of lipids is essential for rapid cell growth and is elevated in many tumours suggesting that inhibition of this pathway may prevent proliferation [29]. AMPK is a negative regulator of 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMG-CoA) and acetyl-CoA-carboxylase (ACC), the rate limiting enzymes for cholesterol and non-esterified fatty acid biosynthesis respectively [30,31]. We have recently demonstrated that in the liver metformin and the direct AMPK β1 activator A769662 inhibit de novo lipogenesis through AMPKβ1-dependent phosphorylation and inhibition of ACC [32]. Salicylate also inhibits de novo lipogenesis in the liver [33] but whether this effect is dependent on AMPK β1 has not yet been established. In prostate cancer AMPK expression is vital for inhibiting prostate cancer cell proliferation [34], an effect that may be related to the inhibition of de novo lipogenesis [35,36]. Whether salicylate and metformin reduce lipogenesis in cancer cells and if this is important for inhibition of proliferation/survival is not known.

In the current study, we demonstrate that salicylate dose-dependently reduces the clonogenic survival of prostate and lung adenocarcinomas with differing sensitivities (prostate > lung). The reductions in clonogenic survival occur at serum salicylate concentrations achievable through oral intake of aspirin and are associated with increases in ACC phosphorylation and the suppression of de novo lipogenesis. In mouse embryonic fibroblasts (MEFs) AMPK β1-subunit expression is vital for salicylate-induced ACC phosphorylation and inhibition of de novo lipogenesis. Salicylate induced inhibition of clonogenic survival in prostate and lung cancer is blunted by supplementing the media with fatty acids and/or mevalonate indicating that the suppression of de novo lipogenesis may be vital for reducing cell survival. Importantly, the effects of salicylate on cell survival, ACC phosphorylation and de novo lipogenesis are enhanced when used in combination with metformin. These data suggest that salicylate (aspirin) and metformin may work collaboratively to reduce the survival of prostate and lung cancer through the inhibition of de novo lipogenesis.

MATERIALS AND METHODS

Cell lines and treatments

Lung (A549, H1299) and prostate (PC3, 22RV-1) cancer cells as well as lung (MRC-5) and prostate (PNT1A) epithelial cells were obtained from the American Type Culture Collection (ATCC) MEFs from wild-type (WT) and AMPKβ1 knockout (KO) mice were generated from primary MEFs through spontaneous immortalization using sequential passaging. Immortalization was considered complete when the cells showed uniform morphology and rapid proliferation rate, which occurred typically at passages 20–24. MEFs and MRC-5 cells were grown in Dulbecco's modified Eagle's medium (DMEM) (Gibco) supplemented with 1% antibiotic–antimycotic and 10% FBS. Lung and prostate cancer as well as prostate epithelial cells were grown using Roswell Park Memorial Institute (RPMI) medium supplemented with 1% antibiotic–antimycotic (Gibco) and 10% FBS (Gibco). All cells were maintained at 37°C and were treated with the indicated concentrations of salicylate (Sigma) or metformin (Sigma) for 48 h unless stated otherwise.

Clonogenic assay

Cells were subjected to clonogenic assays as described [37]. In brief, approximately 500 cells were seeded into individual wells of a 12-well plate in triplicate and maintained at the indicated doses of salicylate. After 7 days, cells were fixed with crystal violet and viable colonies (>50 cells) were counted.

Immunoblotting

Cells were washed with PBS and lysed in ice-cold lysis buffer (20 mM Tris/HCl, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton-X100, 2.5 mM Na4O7P2, 1 mM Na3VO4 and 1 Roche inhibitor cocktail). Laemmli–SDS-sample buffer (4×) was then added and the samples were subsequently boiled. Either 30 (H1299 and MEFs) or 10 (22RV-1) μg of protein were separated by SDS/PAGE, transferred to PVDF or nitrocellulose membrane as specified and incubated with primary and horseradish peroxidase (HRP)-conjugated secondary antibodies (Cell Signaling).

Densitometry

Immunoblots were analysed using ImageJ software (McMaster University Biophotonics Lab). Densitometry values are expressed as a percentage of control.

Lipogenesis assay

Cells were treated with the indicated drugs for 48 h prior to the addition of radiolabelled 3H-sodium acetate (10 μCi/ml, PerkinElmer) and unlabelled sodium acetate (0.5 mM, Sigma) for 4 h. The cells were then washed with PBS, manually scraped and collected into 1.5 ml tubes and lipid extraction with chloroform–methanol was performed as described [32].

Statistical analysis

Unless otherwise indicated, results are expressed as mean ± S.E.M. Statistical analyses were performed using a two-way or one-way ANOVA when appropriate. For between treatments analysis Fisher-Least Significant Difference post-hoc test was used; for genotype analysis Sidak multiple comparison test was utilized. Significance was set at *P<0.05 using GraphPad Prism 6 software (La Jolla). For clonogenic and lipogenesis assays IC50 values were calculated using a non-linear regression model (with normalized slope), in GraphPad Prism 6 (La Jolla). Drug synergy was calculated using the combination index (CI)-isobologram theorem and CompuSyn software [38].

RESULTS

Salicylate dose-dependently inhibits clonogenic survival

At clinical concentrations of salicylate achievable through the intake of regular strength aspirin (<1.0 mM) [39,40] salicylate inhibited the survival of prostate (Figure 1A) and lung (Figure 1B) cancer cells by greater than 50%. In contrast, at concentrations of less than 1 mM salicylate had little effect on MRC-5 lung or PNT1A prostate epithelial cells (Supplementary Figure S1), suggesting the effects of salicylate on cancer cell survival were not related to general cell toxicity. We hypothesized that the inhibition of clonogenic survival in prostate and lung cancers involved activation of AMPK; therefore, we examined the phosphorylation of AMPKα (Thr172) and its downstream substrate ACC (Ser79). ACC is considered the most sensitive measure of cellular AMPK activity as it takes into account both the allosteric and the covalent regulation of AMPK [41]. In 22RV-1 prostate cancer cells only 5.0 mM salicylate increased Thr172 AMPK phosphorylation. In contrast, salicylate increased Ser79 ACC phosphorylation starting at concentrations as low as 1 mM (Figures 2A–2C). The lack of phosphorylation of AMPK but increases in ACC phosphorylation at concentrations < 5.0 mM is consistent with direct activation of AMPK via allosteric mechanisms involving the β1-subunit and independent of energy charge [7,8]. Treatment of 22RV-1 cells with 1 mM salicylate tended to reduce the phosphorylation of p70S6K (P = 0.063) and S6 (P = 0.058) relative to total p70S6K and S6 protein expression (Figures 2A, 2C and 2D). At higher concentrations of salicylate outside of the therapeutic window of exposure (3 and 5 mM) there were also reductions in the expression of β-actin and total p70S6K and S6 (Supplementary Figure 2A; Table 1) consistent with the high-degree of toxicity (Figure 1A). Interestingly, a similar reduction in the total expression of p70S6K and S6 has also been observed in prostate cancer cells treated chronically with the AMPK activator 5-Aminoimidazole-4-carboxamide ribonucleotide (AICAR) [42,43]. In H1299 lung cancer cells salicylate did not alter the phosphorylation of Thr172 AMPK (Figures 3A and 3B) but starting at concentrations of 1 mM it did increase the phosphorylation of ACC (Figures 3A and 3C) and lowered the phosphorylation of p70S6K (Thr389; Figures 3A and 3D) and S6 (Ser240/244; Figures 3A and 3E).

Table 1
Total protein expression relative to β-actin in 22RV-1 and H1299 cancer cells in a salicylate dose–response study (data calculated from Figures 2 and 3)
22RV-1H1299
Salicylate concentration (mM)Salicylate concentration (mM)
Control0.250.501.003.005.00Control0.250.51.003.005.00
AMPK 1.0 1.23±0.09 1.18±0.13 1.12±0.13 2.06* ± 0.36 1.90* ± 0.38 1.0 0.94±0.04 1.02±0.04 1.04±0.05 1.22±0.04 1.85**** ± 0.25 
ACC 1.0 1.83±0.49 1.74±0.49 1.19±0.23 2.37** ± 0.54 1.96* ± 0.48 1.0 0.84±0.15 0.94±0.11 0.86±0.15 0.74±0.15 0.69±0.21 
p70 S6K 1.0 1.12±0.02 1.15±0.08 1.07±0.09 1.67±0.54 1.99* ± 0.55 1.0 0.72±0.13 0.83±0.13 0.89±0.10 0.78±0.17 0.84±0.19 
S6 1.0 1.13±0.08 1.09±0.12 1.04±0.09 1.46±0.25 1.95* ± 0.63 1.0 0.97±0.10 1.03±0.07 0.90±0.06 0.80±0.06 0.66±0.16 
22RV-1H1299
Salicylate concentration (mM)Salicylate concentration (mM)
Control0.250.501.003.005.00Control0.250.51.003.005.00
AMPK 1.0 1.23±0.09 1.18±0.13 1.12±0.13 2.06* ± 0.36 1.90* ± 0.38 1.0 0.94±0.04 1.02±0.04 1.04±0.05 1.22±0.04 1.85**** ± 0.25 
ACC 1.0 1.83±0.49 1.74±0.49 1.19±0.23 2.37** ± 0.54 1.96* ± 0.48 1.0 0.84±0.15 0.94±0.11 0.86±0.15 0.74±0.15 0.69±0.21 
p70 S6K 1.0 1.12±0.02 1.15±0.08 1.07±0.09 1.67±0.54 1.99* ± 0.55 1.0 0.72±0.13 0.83±0.13 0.89±0.10 0.78±0.17 0.84±0.19 
S6 1.0 1.13±0.08 1.09±0.12 1.04±0.09 1.46±0.25 1.95* ± 0.63 1.0 0.97±0.10 1.03±0.07 0.90±0.06 0.80±0.06 0.66±0.16 

*P<0.05, **P<0.01, ****P<0.0001

Salicylate reduces colony survival in prostate and lung cancer cells

Figure 1
Salicylate reduces colony survival in prostate and lung cancer cells

(A) Approximately 5×102 prostate (22RV-1 and PC3) or (B) lung (A549 and H1299) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated (mM) concentrations of salicylate and allowed to grow for 7 days. These cells were then subsequently fixed and stained with crystal violet and viable colonies >50 cells were counted under a microscope. The results are expressed as the mean and S.E.M. relative to untreated controls from three independent experiments. Mean ± S.E.M. values of three independent experiments are shown. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 calculated using one-way ANOVA.

Figure 1
Salicylate reduces colony survival in prostate and lung cancer cells

(A) Approximately 5×102 prostate (22RV-1 and PC3) or (B) lung (A549 and H1299) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated (mM) concentrations of salicylate and allowed to grow for 7 days. These cells were then subsequently fixed and stained with crystal violet and viable colonies >50 cells were counted under a microscope. The results are expressed as the mean and S.E.M. relative to untreated controls from three independent experiments. Mean ± S.E.M. values of three independent experiments are shown. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 calculated using one-way ANOVA.

Salicylate activates AMPK and inhibits mTOR in 22RV-1 prostate cancer cells

Figure 2
Salicylate activates AMPK and inhibits mTOR in 22RV-1 prostate cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation relative to total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of 5–16 independent experiments are shown. *P<0.05, ***P<0.001, ****P<0.0001 relative to control (0 mM salicylate) calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Figure 2
Salicylate activates AMPK and inhibits mTOR in 22RV-1 prostate cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation relative to total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of 5–16 independent experiments are shown. *P<0.05, ***P<0.001, ****P<0.0001 relative to control (0 mM salicylate) calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Salicylate activates AMPK and inhibits mTOR in H1299 lung cancer cells

Figure 3
Salicylate activates AMPK and inhibits mTOR in H1299 lung cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of 5–12 independent experiments are shown. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 relative to control (0 mM salicylate) calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Figure 3
Salicylate activates AMPK and inhibits mTOR in H1299 lung cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of 5–12 independent experiments are shown. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 relative to control (0 mM salicylate) calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Salicylate inhibits de novo lipogenesis via an AMPK β1-dependent pathway and is important for inhibiting the clonogenic survival of prostate and lung cancer cells.

To examine the specificity of AMPK β1 for salicylate-induced increases in ACC phosphorylation and suppression of p70S6K/S6 we treated WT and AMPK β1 null MEFs. Salicylate dose-dependently increased AMPK and ACC phosphorylation in WT but not AMPK β1 null MEFs (Figures 4A–4C). At salicylate concentrations between 0.25 and 1.0 mM, the suppression of p70S6K occurred via an AMPK β1-dependent pathway (Figures 4A and 4D). In contrast, salicylate suppressed the phosphorylation of S6 in both WT and AMPK β1 null MEFs (Figures 4A and 4E). These data indicate that at clinically relevant concentrations salicylate increases the phosphorylation of ACC through an AMPK β1-dependent pathway.

Salicylate activation of AMPK and inhibition of mTOR in MEFs requires AMPK β1 expression

Figure 4
Salicylate activation of AMPK and inhibition of mTOR in MEFs requires AMPK β1 expression

Representative immunoblot using PVDF membrane (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Immunoblots were quantified using ImageJ software. Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of 4–6 independent experiments are shown. Indicated P-values are relative to control (0 mM salicylate) within the same genotype: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 within treatment group, #P<0.05, ##P<0.01, ###P<0.001 relative to same concentration of salicylate in WT controls using two-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Figure 4
Salicylate activation of AMPK and inhibition of mTOR in MEFs requires AMPK β1 expression

Representative immunoblot using PVDF membrane (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Immunoblots were quantified using ImageJ software. Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of 4–6 independent experiments are shown. Indicated P-values are relative to control (0 mM salicylate) within the same genotype: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 within treatment group, #P<0.05, ##P<0.01, ###P<0.001 relative to same concentration of salicylate in WT controls using two-way ANOVA. Phosphorylated and total proteins were measured on different gels.

AMPK phosphorylation of ACC is the rate-limiting step controlling de novo fatty acid synthesis [32]. Elevated rates of de novo lipogenesis are observed in lung and prostate cancers [29] therefore we hypothesized that salicylate may inhibit cell survival through the regulation of this pathway. Salicylate dose-dependently inhibited de novo fatty acid synthesis in both 22RV-1 prostate (Figure 5A) and H1299 lung (Figure 5B) cancer cells. The inhibition of de novo fatty acid synthesis correlated strongly with clonogenic survival in both cell lines suggesting a possible interrelationship (Figure 5C). In MEFs, the salicylate-induced inhibition of de novo lipogenesis was dependent on the expression of the AMPK β1 subunit at salicylate concentrations less than 3 mM (Figure 5D). At higher concentrations the inhibition of lipogenesis was independent of AMPK β1, consistent with previous reports in hepatocytes indicating that this effect is independent of ACC activity and is probably due to salicylate-induced uncoupling and activation of the citric acid cycle, which in turn starves ACC of substrate [44].

Salicylate-induced inhibition of cell survival in prostate and lung cancer involves the inhibition of de novo fatty acid and cholesterol synthesis

Figure 5
Salicylate-induced inhibition of cell survival in prostate and lung cancer involves the inhibition of de novo fatty acid and cholesterol synthesis

(A) 22RV-1 or (B) H1299 cells were treated with the indicated concentrations of salicylate for 48 h. 3H-acetate incorporation into cellular lipids was then evaluated. The results are expressed as nanomoles of acetate per milligram of protein per hour. **P<0.01 ***P<0.001 ****P<0.0001 relative to control as determined by one-way ANOVA analysis. (C) A linear regression analysis was conducted in 22RV-1 and H1299 cells comparing suppression of lipogenesis with clonogenic survival. (D) MEFs WT or AMPK β1−/− were treated with the indicated concentrations of salicylate for 48 h. 3H-acetate incorporation into lipids was measured. The results are expressed as nanomoles of acetate per milligram of protein per hour normalized to the control, *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 within genotype; #P<0.05, ##P<0.001 within treatment group as calculated by two-way ANOVA. 22RV-1 (E) or H1299 (F) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated (mM) concentrations of salicylate along with mevalonate/oleate or both and allowed to grow for 7 days. These cells were subsequently fixed and stained with crystal violet and viable colonies >50 cells were counted under a microscope. Legend conserved for E and F. The results are expressed as the mean and S.E.M. relative to treated vehicle from at least three independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 relative to salicylate treatment group as calculated by one-way ANOVA.

Figure 5
Salicylate-induced inhibition of cell survival in prostate and lung cancer involves the inhibition of de novo fatty acid and cholesterol synthesis

(A) 22RV-1 or (B) H1299 cells were treated with the indicated concentrations of salicylate for 48 h. 3H-acetate incorporation into cellular lipids was then evaluated. The results are expressed as nanomoles of acetate per milligram of protein per hour. **P<0.01 ***P<0.001 ****P<0.0001 relative to control as determined by one-way ANOVA analysis. (C) A linear regression analysis was conducted in 22RV-1 and H1299 cells comparing suppression of lipogenesis with clonogenic survival. (D) MEFs WT or AMPK β1−/− were treated with the indicated concentrations of salicylate for 48 h. 3H-acetate incorporation into lipids was measured. The results are expressed as nanomoles of acetate per milligram of protein per hour normalized to the control, *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 within genotype; #P<0.05, ##P<0.001 within treatment group as calculated by two-way ANOVA. 22RV-1 (E) or H1299 (F) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated (mM) concentrations of salicylate along with mevalonate/oleate or both and allowed to grow for 7 days. These cells were subsequently fixed and stained with crystal violet and viable colonies >50 cells were counted under a microscope. Legend conserved for E and F. The results are expressed as the mean and S.E.M. relative to treated vehicle from at least three independent experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 relative to salicylate treatment group as calculated by one-way ANOVA.

To directly test whether salicylate-induced inhibition of lipogenesis was important for reducing clonogenic survival within the therapeutic window of exposure achieved with aspirin intake (0.25–1.0 mM), we supplemented the media with mevalonate, and/or oleate which alleviate the need for cholesterol or fatty acid synthesis, respectively [35]. In 22RV-1 prostate cancer cells supplementation with mevalonate largely prevented the inhibitory effects of salicylate on clonogenic survival (Figure 5E). Supplementation of the media with oleate had more modest effects that were not additive with mevalonate (Figure 5E). In H1299 lung cancer cells mevalonate was without effect but oleate supplementation completely prevented salicylate-induced suppression of clonogenic survival (Figure 5F). These data indicate that the salicylate-induced suppression of lipogenesis, taking place at clinically relevant doses of the drug, is mediated via the AMPK β1 subunit and that the suppression of de novo cholesterol and fatty acid biosynthesis is an important mechanism contributing to the reduced survival of 22RV-1 prostate and H1299 lung cancer cells respectively.

Metformin and salicylate synergistically inhibit cancer cell proliferation

Many type 2 diabetics who take metformin for blood glucose control are also prescribed aspirin for cardioprotection [45]. To examine the interaction of these two commonly used medications, we analysed the dose–response relationship with the two drugs or their combination with clonogenic survival. We found that the IC50 for clonogenic survival was dramatically reduced in all cell types when metformin and salicylate were used in combination (Figure 6; Table 2). Further analysis of the IC50 curves using CompuSyn based software indicated that in prostate (PC3 and 22RV-1) and lung (A549 and H1299) cancer cells the effects of metformin and salicylate were either synergistic or additive (Table 2). In contrast, both prostate and lung epithelial cell lines were substantially more resistant to the anti-clonogenic effect of salicylate and metformin combination therapy (Supplementary Figure S1; Table 2). In addition there was no additivity or synergy towards inhibiting clonogenic survival in the prostate and epithelial cancer cells when used in combination (Supplementary Figure S1; Table 2).

Salicylate and metformin synergistically or additively reduce clonogenic survival of prostate and lung cancer cells

Figure 6
Salicylate and metformin synergistically or additively reduce clonogenic survival of prostate and lung cancer cells

PC3 (A), 22RV-1 (B), A549 (C) and H1299 (D) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated equivalent concentrations of salicylate and/or metformin and allowed to grow for 7 days. These cells were then subsequently fixed and stained with crystal violet and viable colonies >50 cells were counted under a microscope. The results are expressed as the mean and S.E.M. relative to untreated controls from at least three independent experiments. *P<0.05, **P<0.01 salicylate compared with combination calculated using two-way ANOVA.

Figure 6
Salicylate and metformin synergistically or additively reduce clonogenic survival of prostate and lung cancer cells

PC3 (A), 22RV-1 (B), A549 (C) and H1299 (D) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated equivalent concentrations of salicylate and/or metformin and allowed to grow for 7 days. These cells were then subsequently fixed and stained with crystal violet and viable colonies >50 cells were counted under a microscope. The results are expressed as the mean and S.E.M. relative to untreated controls from at least three independent experiments. *P<0.05, **P<0.01 salicylate compared with combination calculated using two-way ANOVA.

Table 2
Measured IC50 for colony survival of adenocarcinomas and epithelial cells treated with salicylate, metformin or combination (data calculated from Figure 6 and Supplementary Figure S1)
IC50 values (mM)
Cell lineSalicylateMetforminCombinationCombined interaction
PC3 0.67±0.07 1.15±0.07 0.30±0.04 Synergistic 
22RV-1 0.30±0.07 0.67±0.07 0.15±0.05 Additive 
H1299 0.40±0.07 0.98±0.10 0.28±0.07 Additive 
A549 0.97±0.14 0.42±0.13 0.29±0.10 Additive 
PNT1A 1.96±0.04 1.13±0.05 1.05±0.02 None 
MRC-5 3.76±0.08 1.74±0.02 1.57±0.07 None 
IC50 values (mM)
Cell lineSalicylateMetforminCombinationCombined interaction
PC3 0.67±0.07 1.15±0.07 0.30±0.04 Synergistic 
22RV-1 0.30±0.07 0.67±0.07 0.15±0.05 Additive 
H1299 0.40±0.07 0.98±0.10 0.28±0.07 Additive 
A549 0.97±0.14 0.42±0.13 0.29±0.10 Additive 
PNT1A 1.96±0.04 1.13±0.05 1.05±0.02 None 
MRC-5 3.76±0.08 1.74±0.02 1.57±0.07 None 

Subsequently, we examined the effects of clinically achievable concentrations of metformin (100 μM) with or without 1 mM salicylate on AMPK signalling and lipogenesis in 22RV-1 (Figure 7) and H1299 (Figure 8) cells. In 22RV-1 prostate cancer cells, metformin treatment did not affect the phosphorylation of AMPK or downstream substrates (ACC, S6 or p70S6K; Figures 7A–7E). Consistent with previous results, salicylate at 1 mM had modest effects on ACC, S6 and p70S6K phosphorylation but when combined with metformin these effects became more dramatic (Figures 7A–7E). Similar results were observed in H1299 lung cancer cells (Figures 8A–8E). Mirroring changes in ACC phosphorylation, metformin alone did not affect lipogenesis but when combined with salicylate it enhanced the inhibition of lipogenesis in both 22RV-1 (Figure 9A) and H1299 (Figure 9B) cells. In MEFs, the inhibition of lipogenesis by metformin and salicylate was dependent on the presence of the AMPK β1 subunit (Figure 9C). Importantly, the supplementation of the media with oleate and/or mevalonate blunted the suppressive effects of salicylate and metformin co-treatment on clonogenic survival in both 22RV-1 prostate (Figure 9D) and H1299 lung (Figure 9E) cancer cells.

Combined metformin and salicylate treatment potentiates AMPK activity and mTOR inhibition in 22RV-1 prostate cancer cells

Figure 7
Combined metformin and salicylate treatment potentiates AMPK activity and mTOR inhibition in 22RV-1 prostate cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Met=100 μM Metformin and Sal=1 mM Salicylate. Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of at least 10–16 independent experiments are shown. **P<0.01, ***P<0.001, ****P<0.0001 relative to control using one-way ANOVA; #P<0.05 between treatment groups calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Figure 7
Combined metformin and salicylate treatment potentiates AMPK activity and mTOR inhibition in 22RV-1 prostate cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Met=100 μM Metformin and Sal=1 mM Salicylate. Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of at least 10–16 independent experiments are shown. **P<0.01, ***P<0.001, ****P<0.0001 relative to control using one-way ANOVA; #P<0.05 between treatment groups calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Combined metformin and salicylate treatment potentiates AMPK activity and mTOR inhibition in H1299 lung cancer cells

Figure 8
Combined metformin and salicylate treatment potentiates AMPK activity and mTOR inhibition in H1299 lung cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Met=100 μM Metformin and Sal=1 mM Salicylate. Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of at least 7–12 independent experiments are shown. *P<0.05, **P<0.01, ***P<0.001 relative to control using one-way ANOVA; #P<0.05 between treatment groups calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Figure 8
Combined metformin and salicylate treatment potentiates AMPK activity and mTOR inhibition in H1299 lung cancer cells

Representative immunoblot (A) and densitometry for indicated phosphorylation and total AMPK (B), ACC (C) p70S6 kinase (D) and S6 (E). Met=100 μM Metformin and Sal=1 mM Salicylate. Densitometry values are expressed as percentage of control. Mean ± S.E.M. values of at least 7–12 independent experiments are shown. *P<0.05, **P<0.01, ***P<0.001 relative to control using one-way ANOVA; #P<0.05 between treatment groups calculated using one-way ANOVA. Phosphorylated and total proteins were measured on different gels.

Metformin augments salicylate-induced inhibition of lipogenesis

Figure 9
Metformin augments salicylate-induced inhibition of lipogenesis

22RV-1 prostate (A) or H1299 lung (B) cancer cells were treated with vehicle, salicylate (1 mM), metformin (0.1 mM) or both salicylate and metformin for 48 h. 3H-acetate incorporation into lipids was then evaluated. The results are expressed as nanomoles of acetate per milligram of protein per hour. ***P<0.001 ****P<0.0001 relative to control using one-way ANOVA; #P<0.05 between treatment groups as determined by one-way ANOVA analysis. (C) MEFS (WT or AMPK β1−/−) were treated with the vehicle, salicylate (1 mM), metformin (0.1 mM) or both salicylate and metformin for 48 h. 3H-acetate incorporation into lipids was measured. The results are expressed as nanomoles of acetate per milligram of protein per hour normalized to the control, ****P<0.0001 within genotype; ####P<0.0001 within treatment group as calculated by two-way ANOVA. Approximately 5×102 22RV-1 (D) or H1299 (E) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated (mM) concentrations of salicylate, 0.1 mM metformin or both along with mevalonate, oleate or mevalonate and oleate, the cells were allowed to grow for 7 days. These cells were then subsequently fixed and stained with Crystal Violet and viable colonies >50 cells were counted under a microscope. Legend conserved for D and E. The results are expressed as the mean and S.E.M. relative to untreated controls from at least three independent experiments. ***P<0.001, ****P<0.0001 all values were calculated using one-way ANOVA.

Figure 9
Metformin augments salicylate-induced inhibition of lipogenesis

22RV-1 prostate (A) or H1299 lung (B) cancer cells were treated with vehicle, salicylate (1 mM), metformin (0.1 mM) or both salicylate and metformin for 48 h. 3H-acetate incorporation into lipids was then evaluated. The results are expressed as nanomoles of acetate per milligram of protein per hour. ***P<0.001 ****P<0.0001 relative to control using one-way ANOVA; #P<0.05 between treatment groups as determined by one-way ANOVA analysis. (C) MEFS (WT or AMPK β1−/−) were treated with the vehicle, salicylate (1 mM), metformin (0.1 mM) or both salicylate and metformin for 48 h. 3H-acetate incorporation into lipids was measured. The results are expressed as nanomoles of acetate per milligram of protein per hour normalized to the control, ****P<0.0001 within genotype; ####P<0.0001 within treatment group as calculated by two-way ANOVA. Approximately 5×102 22RV-1 (D) or H1299 (E) cancer cells were seeded into 12-well plates. The following day, these cells were treated with the indicated (mM) concentrations of salicylate, 0.1 mM metformin or both along with mevalonate, oleate or mevalonate and oleate, the cells were allowed to grow for 7 days. These cells were then subsequently fixed and stained with Crystal Violet and viable colonies >50 cells were counted under a microscope. Legend conserved for D and E. The results are expressed as the mean and S.E.M. relative to untreated controls from at least three independent experiments. ***P<0.001, ****P<0.0001 all values were calculated using one-way ANOVA.

DISCUSSION

An important role for aspirin in suppressing colon cancer has been highly documented [5,15]; however reduced incidence of adenocarcinomas of the lung and prostate have also been noted in numerous studies (for detailed meta-analysis and review see [46]). We find that the clonogenic survival of prostate and lung cancer cells is impaired with salicylate treatment of cells at concentrations that are consistent with the ingestion of regular strength aspirin. Importantly, this effect is not observed in prostate and lung epithelial cell lines. Given the essential requirement for de novo lipogenesis in rapidly dividing cells and the elevated rates of de novo lipogenesis observed in many prostate and lung cancers [4750], we examined the effects of salicylate on this pathway. We find that salicylate at concentrations as low as 0.25 mM inhibited de novo lipogenesis in prostate and lung cancer cells and this was associated with the inhibition of clonogenic survival. In MEFs salicylate also inhibited lipogenesis and this effect was entirely dependent on the AMPK β1 subunit at concentrations of less than 1 mM. The finding that higher concentrations of salicylate inhibited de novo lipogenesis independently of AMPK is consistent with previous observations that high concentrations of salicylate can uncouple mitochondrial respiration [8] resulting in increased TCA cycle flux which reduces acetyl-CoA, thus limiting the substrate for ACC [44]. Importantly, we find that when the cancer cells were supplemented with oleate and/or mevalonate, the inhibitory effects of salicylate on cell survival were severely blunted indicating a vital need for lipogenesis in controlling cell survival in rapidly replicating cells. The findings that either cholesterol or oleate could blunt the inhibitory effects of salicylate on clonogenic survival suggest substantial overlap between de novo fatty acid and cholesterol synthesis pathways. One explanation for these potential findings may be related to the presence of a trans-methylglutaconate shunt in which mevalonate is diverted to HMG-CoA and free acetoacetate is then utilized preferentially for fatty acid synthesis [5153]. Future studies investigating whether this occurs in prostate and lung cancer cells are warranted.

It is known that metformin activates AMPK by reducing the adenylate charge of the cell resulting in subsequent activation of the kinase via the AMPK γ-subunit [27,28]. In contrast, salicylate increases AMPK activity through an allosteric mechanism requiring the AMPK β1 subunit [8,41]. Given their distinct mechanism of AMPK activation, we tested whether combining these two well-tolerated therapies could lead to the potentiation of the anti-tumour activity of salicylate in cancer cells. We found that the effects of metformin were either additive or synergistic with salicylate for inhibiting prostate and lung cancer cell survival. These effects were not observed in the prostate and lung epithelial cell lines. Consistent, with changes in ACC phosphorylation the rate of cellular lipogenesis and cell survival was significantly reduced with combination therapy. Importantly, the combined effects of salicylate and metformin on lipogenesis were dependent on the expression of the β1 subunit of AMPK and the inhibition of clonogenic survival was eliminated when the media were supplemented with oleate and/or mevalonate. These data indicate that when metformin is combined with an allosteric activator of AMPK such as salicylate there is a potentiation of the allosteric response. These findings are consistent with our observations in cell-free assays which demonstrate that AMP (generated in hepatocytes by metformin [54]) and salicylate together increase the activity of purified, non-phosphorylated AMPK α1β1γ1 heterotrimers by greater than 60-fold, but only have minor effects on allosteric activation when provided alone [55]. This is also consistent with the findings that the drug A769662, which binds to the same site as salicylate, dramatically potentiates allosteric activation of AMPK when applied in combination with AMP or the AMP mimetic AICAR [56,57].

In conclusion, the present study is the first to investigate the effects of therapeutic concentrations of salicylate on AMPK, ACC and lipogenesis in prostate and lung cancer cells. The inhibitory effects of salicylate on prostate and lung cancer survival are largely dependent on de novo lipogenesis suggesting that inhibition of this pathway may be vital for its therapeutic effects. Furthermore, we underscore the pharmacological potential of combining clinical concentrations of metformin with salicylate to reduce the survival of prostate and lung cancers. However, studies in prostate and lung cancer cells with AMPK β1 deletion or ACC Ser79Ala knock-in mutations will be required to definitively establish the importance of this pathway for inhibiting prostate and lung cancer growth and survival in response to salicylate or salicylate and metformin combination therapy. Future clinical studies evaluating whether well-tolerated salicylate based drugs such as salsalate (which does not promote gastrointestinal bleeding like aspirin) also inhibit prostate and lung cancer development in pre-clinical animal models are also warranted. In addition careful retrospective analysis of interactions between aspirin and metformin in clinical trials investigating cancer development may be informative to guide population-based chemopreventive strategies.

AUTHOR CONTRIBUTION

Andrew O’Brien, Theodoros Tsakirdis, Paolo Muti and Gregory Steinberg conceived and designed the study. Andrew O’Brien, Linda Villani, Lindsay Broadfield, Vanessa Houde and Sandra Galic preformed experiments. Andrew O’Brien analysed data. Andrew O’Brien, Theodoros Tsakirdis, Vanessa Houde, Paola Muti, Sandra Galic and Gregory Steinberg interpreted results of experiments. Andrew O’Brien, Vanessa Houde and Gregory Steinberg prepared figures. Andrew O’Brien and Gregory Steinberg drafted the manuscript. Andrew O’Brien, Vanessa Houde, Paolo Muti, Giovanni Blandino, Theodoros Tsakirdis, Bruce Kemp, Sandra Galic and Gregory Steinberg edited and revised the manuscript.

We thank Dr Toran Sanli, Dr Rebecca Ford and Dr Sabrina Strano for helpful discussions and Andrew Collins for technical assistance. Gregory R. Steinberg is a Canada Research Chair in Metabolism and Obesity and the J. Bruce Duncan Chair in Metabolic Diseases.

FUNDING

This work was supported by the Canadian Institutes of Health Research [grant number MOP-11480 (to G.R.S. and T.T.)]; the Canadian Cancer Society [grant number P#20001191 (to G.R.S. and P.M.)]; the Australian Research Council [grant number DP130104548 (to G.R.S., B.E.K. and S.G.)]; the National Health and Medical Research Council [grant number APP1085460 (to G.R.S., B.E.K. and S.G.)]; and the Victorian Government's Operational Infrastructure Support Program.

Abbreviations

     
  • ACC

    acetyl-CoA carboxylase

  •  
  • AICAR

    5-Aminoimidazole-4-carboxamide ribonucleotide

  •  
  • AMPK

    AMP-activated protein kinase

  •  
  • HMG-CoA

    3-hydroxy-3-methyl-glutaryl-CoA reductase

  •  
  • KO

    knockout

  •  
  • MEF

    mouse embryonic fibroblast

  •  
  • mTOR

    mammalian target of rapamycin

  •  
  • WT

    wild-type

References

References
1
Rothwell
 
P.M.
Fowkes
 
F.G.R.
Belch
 
J.F.
Ogawa
 
H.
Warlow
 
C.P.
Meade
 
T.W.
 
Effect of daily aspirin on long-term risk of death due to cancer: analysis of individual patient data from randomised trials
Lancet.
2011
, vol. 
377
 (pg. 
31
-
41
)
[PubMed]
2
Liu
 
Y.
Chen
 
J.Q.
Xie
 
L.
Wang
 
J.
Li
 
T.
He
 
Y.
Gao
 
Y.
Qin
 
X.
Li
 
S.
 
Effect of aspirin and other non-steroidal anti-inflammatory drugs on prostate cancer incidence and mortality: a systematic review and meta-analysis
BMC Med.
2014
, vol. 
12
 pg. 
55
 
[PubMed]
3
Shebl
 
F.M.
Sakoda
 
L.C.
Black
 
A.
Koshiol
 
J.
Andriole
 
G.L.
Grubb
 
R.
Church
 
T.R.
Chia
 
D.
Zhou
 
C.
Chu
 
L.W.
, et al 
Aspirin but not ibuprofen use is associated with reduced risk of prostate cancer: a PLCO study
Br. J. Cancer
2012
, vol. 
107
 (pg. 
207
-
214
)
[PubMed]
4
Thun
 
M.J.
Namboodiri
 
M.M.
Calle
 
E.E.
Flanders
 
W.D.
Heath
 
C.W.
 
Aspirin use and risk of fatal cancer
Cancer Res.
1993
, vol. 
53
 (pg. 
1322
-
1327
)
[PubMed]
5
Din
 
F.V.
Theodoratou
 
E.
Farrington
 
S.M.
Tenesa
 
A.
Barnetson
 
R.A.
Cetnarskyj
 
R.
Stark
 
L.
Porteous
 
M.E.
Campbell
 
H.
Dunlop
 
M.G.
 
Effect of aspirin and NSAIDs on risk and survival from colorectal cancer
Gut
2010
, vol. 
59
 (pg. 
1670
-
1679
)
[PubMed]
6
Higgs
 
G.A.
Salmon
 
J.A.
Henderson
 
B.
Vane
 
J.R.
 
Pharmacokinetics of aspirin and salicylate in relation to inhibition of arachidonate cyclooxygenase and antiinflammatory activity
Proc. Natl. Acad. Sci. U.S.A.
1987
, vol. 
84
 (pg. 
1417
-
1420
)
[PubMed]
7
Steinberg
 
G.R.
Dandapani
 
M.
Hardie
 
D.G.
 
AMPK: mediating the metabolic effects of salicylate-based drugs?
Trends Endocrinol. Metab.
2013
, vol. 
24
 (pg. 
481
-
487
)
[PubMed]
8
Hawley
 
S.A.
Fullerton
 
M.D.
Ross
 
F.A.
Schertzer
 
J.D.
Chevtzoff
 
C.
Walker
 
K.J.
Peggie
 
M.W.
Zibrova
 
D.
Green
 
K.A.
Mustard
 
K.J.
, et al 
The ancient drug salicylate directly activates AMP-activated protein kinase
Science
2012
, vol. 
336
 (pg. 
918
-
922
)
[PubMed]
9
Hay
 
N.
 
The Akt-mTOR tango and its relevance to cancer
Cancer Cell
2005
, vol. 
8
 (pg. 
179
-
183
)
[PubMed]
10
Gwinn
 
D.M.
Shackelford
 
D.B.
Egan
 
D.F.
Mihaylova
 
M.M.
Mery
 
A.
Vasquez
 
D.S.
Turk
 
B.E.
Shaw
 
R.J.
 
AMPK phosphorylation of raptor mediates a metabolic checkpoint
Mol. Cell
2008
, vol. 
30
 (pg. 
214
-
226
)
[PubMed]
11
Inoki
 
K.
Zhu
 
T.
Guan
 
K.L.
 
TSC2 mediates cellular energy response to control cell growth and survival
Cell
2003
, vol. 
115
 (pg. 
577
-
590
)
[PubMed]
12
Manning
 
B.D.
Cantley
 
L.C.
 
United at last: the tuberous sclerosis complex gene products connect the phosphoinositide 3-kinase/Akt pathway to mammalian target of rapamycin (mTOR) signalling
Biochem. Soc. Trans.
2003
, vol. 
31
 (pg. 
573
-
578
)
[PubMed]
13
Law
 
B.K.
Waltner-Law
 
M.E.
Entingh
 
A.J.
Chytil
 
A.
Aakre
 
M.E.
Norgaard
 
P.
Moses
 
H.L.
 
Salicylate-induced growth arrest is associated with inhibition of p70s6k and down-regulation of c-myc, cyclin D1, cyclin A, and proliferating cell nuclear antigen
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
38261
-
38267
)
[PubMed]
14
Lissa
 
D.
Senovilla
 
L.
Rello-Varona
 
S.
Vitale
 
I.
Michaud
 
M.
Pietrocola
 
F.
Boileve
 
A.
Obrist
 
F.
Bordenave
 
C.
Garcia
 
P.
, et al 
Resveratrol and aspirin eliminate tetraploid cells for anticancer chemoprevention
Proc. Natl. Acad. Sci. U.S.A.
2014
, vol. 
111
 (pg. 
3020
-
3025
)
[PubMed]
15
Din
 
F.V.
Valanciute
 
A.
Houde
 
V.P.
Zibrova
 
D.
Green
 
K.A.
Sakamoto
 
K.
Alessi
 
D.R.
Dunlop
 
M.G.
 
Aspirin inhibits mTOR signaling, activates AMP-activated protein kinase, and induces autophagy in colorectal cancer cells
Gastroenterology
2012
, vol. 
142
 pg. 
1504–1515 e1503
 
16
Vincent
 
E.E.
Coelho
 
P.P.
Blagih
 
J.
Griss
 
T.
Viollet
 
B.
Jones
 
R.G.
 
Differential effects of AMPK agonists on cell growth and metabolism
Oncogene
2014
 
doi: 10.1038/onc.2014.301
17
Blandino
 
G.
Valerio
 
M.
Cioce
 
M.
Mori
 
F.
Casadei
 
L.
Pulito
 
C.
Sacconi
 
A.
Biagioni
 
F.
Cortese
 
G.
Galanti
 
S.
, et al 
Metformin elicits anticancer effects through the sequential modulation of DICER and c-MYC
Nat. Commun.
2012
, vol. 
3
 pg. 
865
 
[PubMed]
18
Barba
 
M.
Muti
 
P.
 
Is it time to test metformin in breast cancer prevention trials? A reply to the authors
Cancer Epidemiol. Biomarkers Prev.
2009
, vol. 
18
 pg. 
2565
 
[PubMed]
19
Muti
 
P.
Berrino
 
F.
Krogh
 
V.
Villarini
 
A.
Barba
 
M.
Strano
 
S.
Blandino
 
G.
 
Metformin, diet and breast cancer: an avenue for chemoprevention
Cell Cycle
2009
, vol. 
8
 pg. 
2661
 
[PubMed]
20
Barba
 
M.
Schunemann
 
H.J.
Sperati
 
F.
Akl
 
E.A.
Musicco
 
F.
Guyatt
 
G.
Muti
 
P.
 
The effects of metformin on endogenous androgens and SHBG in women: a systematic review and meta-analysis
Clin. Endocrinol.
2009
, vol. 
70
 (pg. 
661
-
670
)
21
Anisimov
 
V.N.
Berstein
 
L.M.
Egormin
 
P.A.
Piskunova
 
T.S.
Popovich
 
I.G.
Zabezhinski
 
M.A.
Kovalenko
 
I.G.
Poroshina
 
T.E.
Semenchenko
 
A.V.
Provinciali
 
M.
, et al 
Effect of metformin on life span and on the development of spontaneous mammary tumors in HER-2/neu transgenic mice
Exp. Gerontol.
2005
, vol. 
40
 (pg. 
685
-
693
)
[PubMed]
22
Dowling
 
R.J.
Goodwin
 
P.J.
Stambolic
 
V.
 
Understanding the benefit of metformin use in cancer treatment
BMC Med.
2011
, vol. 
9
 pg. 
33
 
[PubMed]
23
Evans
 
J.M.
Donnelly
 
L.A.
Emslie-Smith
 
A.M.
Alessi
 
D.R.
Morris
 
A.D.
 
Metformin and reduced risk of cancer in diabetic patients
BMJ
2005
, vol. 
330
 (pg. 
1304
-
1305
)
[PubMed]
24
Pollak
 
M.
 
Metformin and other biguanides in oncology: advancing the research agenda
Cancer Prev. Res.
2010
, vol. 
3
 (pg. 
1060
-
1065
)
25
Pollak
 
M.
 
Potential applications for biguanides in oncology
J. Clin. Invest.
2013
, vol. 
123
 (pg. 
3693
-
3700
)
[PubMed]
26
Wheaton
 
W.W.
Weinberg
 
S.E.
Hamanaka
 
R.B.
Soberanes
 
S.
Sullivan
 
L.B.
Anso
 
E.
Glasauer
 
A.
Dufour
 
E.
Mutlu
 
G.M.
Budigner
 
G.S.
Chandel
 
N.S.
 
Metformin inhibits mitochondrial complex I of cancer cells to reduce tumorigenesis
eLife
2014
, vol. 
3
 pg. 
e02242
 
[PubMed]
27
Hawley
 
S.A.
Gadalla
 
A.E.
Olsen
 
G.S.
Hardie
 
D.G.
 
The antidiabetic drug metformin activates the AMP-activated protein kinase cascade via an adenine nucleotide-independent mechanism
Diabetes
2002
, vol. 
51
 (pg. 
2420
-
2425
)
[PubMed]
28
Zhou
 
G.
Myers
 
R.
Li
 
Y.
Chen
 
Y.
Shen
 
X.
Fenyk-Melody
 
J.
Wu
 
M.
Ventre
 
J.
Doebber
 
T.
Fujii
 
N.
, et al 
Role of AMP-activated protein kinase in mechanism of metformin action
J. Clin. Invest.
2001
, vol. 
108
 (pg. 
1167
-
1174
)
[PubMed]
29
Menendez
 
J.A.
Lupu
 
R.
 
Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis
Nat. Rev. Cancer
2007
, vol. 
7
 (pg. 
763
-
777
)
[PubMed]
30
Hardie
 
D.G.
 
AMPK: a key regulator of energy balance in the single cell and the whole organism
Int. J. Obesity
2008
, vol. 
32
 
Suppl 4
(pg. 
S7
-
S12
)
31
O’Neill
 
H.M.
Holloway
 
G.P.
Steinberg
 
G.R.
 
AMPK regulation of fatty acid metabolism and mitochondrial biogenesis: implications for obesity
Mol. Cell. Endocrinol.
2013
, vol. 
366
 (pg. 
135
-
151
)
[PubMed]
32
Fullerton
 
M.D.
Galic
 
S.
Marcinko
 
K.
Sikkema
 
S.
Pulinilkunnil
 
T.
Chen
 
Z.P.
O’Neill
 
H.M.
Ford
 
R.J.
Palanivel
 
R.
O’Brien
 
M.
, et al 
Single phosphorylation sites in Acc1 and Acc2 regulate lipid homeostasis and the insulin-sensitizing effects of metformin
Nat. Med.
2013
, vol. 
19
 (pg. 
1649
-
1654
)
[PubMed]
33
Smith
 
M.J.
 
The effects of salicylate on the metabolism of acetate in the rat
J. Biol. Chem.
1959
, vol. 
234
 (pg. 
144
-
147
)
[PubMed]
34
Zhou
 
J.
Huang
 
W.
Tao
 
R.
Ibaragi
 
S.
Lan
 
F.
Ido
 
Y.
Wu
 
X.
Alekseyev
 
Y.O.
Lenburg
 
M.E.
Hu
 
G.F.
Luo
 
Z.
 
Inactivation of AMPK alters gene expression and promotes growth of prostate cancer cells
Oncogene
2009
, vol. 
28
 (pg. 
1993
-
2002
)
[PubMed]
35
Zadra
 
G.
Photopoulos
 
C.
Tyekucheva
 
S.
Heidari
 
P.
Weng
 
Q.P.
Fedele
 
G.
Liu
 
H.
Scaglia
 
N.
Priolo
 
C.
Sicinska
 
E.
 
A novel direct activator of AMPK inhibits prostate cancer growth by blocking lipogenesis
EMBO Mol. Med.
2014
, vol. 
6
 (pg. 
519
-
538
)
[PubMed]
36
Xiang
 
X.
Saha
 
A.K.
Wen
 
R.
Ruderman
 
N.B.
Luo
 
Z.
 
AMP-activated protein kinase activators can inhibit the growth of prostate cancer cells by multiple mechanisms
Biochem. Biophys. Res. Commun.
2004
, vol. 
321
 (pg. 
161
-
167
)
[PubMed]
37
Sanli
 
T.
Rashid
 
A.
Liu
 
C.
Harding
 
S.
Bristow
 
R.G.
Cutz
 
J.C.
Singh
 
G.
Wright
 
J.
Tsakiridis
 
T.
 
Ionizing radiation activates AMP-activated kinase (AMPK): a target for radiosensitization of human cancer cells
Int. J. Radiat. Oncol. Biol. Phys.
2010
, vol. 
78
 (pg. 
221
-
229
)
[PubMed]
38
Chou
 
T.C.
Motzer
 
R.J.
Tong
 
Y.
Bosl
 
G.J.
 
Computerized quantitation of synergism and antagonism of taxol, topotecan, and cisplatin against human teratocarcinoma cell growth: a rational approach to clinical protocol design
J. Natl. Cancer Inst.
1994
, vol. 
86
 (pg. 
1517
-
1524
)
[PubMed]
39
Yin
 
M.-J.
Yamamoto
 
Y.
Gaynor
 
R.B.
 
The anti-inflammatory agents aspirin and salicylate inhibit the activity of IκB kinase-β
Nature
1998
, vol. 
396
 (pg. 
77
-
80
)
[PubMed]
40
Blacklock
 
C.
Lawrence
 
J.
Wiles
 
D.
Malcolm
 
E.
Gibson
 
I.
Kelly
 
C.
Paterson
 
J.
 
Salicylic acid in the serum of subjects not taking aspirin. Comparison of salicylic acid concentrations in the serum of vegetarians, non-vegetarians, and patients taking low dose aspirin
J. Clin. Pathol.
2001
, vol. 
54
 (pg. 
553
-
555
)
[PubMed]
41
Gowans
 
G.J.
Hawley
 
S.A.
Ross
 
F.A.
Hardie
 
D.G.
 
AMP is a true physiological regulator of AMP-activated protein kinase by both allosteric activation and enhancing net phosphorylation
Cell Metab.
2013
, vol. 
18
 (pg. 
556
-
566
)
[PubMed]
42
Xiang
 
X.
Saha
 
A.K.
Wen
 
R.
Ruderman
 
N.B.
Luo
 
Z.
 
AMP-activated protein kinase activators can inhibit the growth of prostate cancer cells by multiple mechanisms
Biochem. Biophys. Res. Commun.
2004
, vol. 
321
 (pg. 
161
-
167
)
[PubMed]
43
Choudhury
 
Y.
Yang
 
Z.
Ahmad
 
I.
Nixon
 
C.
Salt
 
I.P.
Leung
 
H.Y.
 
AMP-activated protein kinase (AMPK) as a potential therapeutic target independent of PI3K/Akt signaling in prostate cancer
Oncoscience
2014
, vol. 
1
 (pg. 
446
-
456
)
[PubMed]
44
Beynen
 
A.C.
Geelen
 
M.J.
 
Short-term inhibition of fatty acid biosynthesis in isolated hepatocytes by mono-aromatic compounds
Toxicology
1982
, vol. 
24
 (pg. 
183
-
197
)
[PubMed]
45
Hundal
 
R.S.
Petersen
 
K.F.
Mayerson
 
A.B.
Randhawa
 
P.S.
Inzucchi
 
S.
Shoelson
 
S.E.
Shulman
 
G.I.
 
Mechanism by which high-dose aspirin improves glucose metabolism in type 2 diabetes
J. Clin. Invest.
2002
, vol. 
109
 (pg. 
1321
-
1326
)
[PubMed]
46
Algra
 
A.M.
Rothwell
 
P.M.
 
Effects of regular aspirin on long-term cancer incidence and metastasis: a systematic comparison of evidence from observational studies versus randomised trials
Lancet Oncol.
2012
, vol. 
13
 (pg. 
518
-
527
)
[PubMed]
47
Pizer
 
E.S.
Pflug
 
B.R.
Bova
 
G.S.
Han
 
W.F.
Udan
 
M.S.
Nelson
 
J.B.
 
Increased fatty acid synthase as a therapeutic target in androgen-independent prostate cancer progression
Prostate
2001
, vol. 
47
 (pg. 
102
-
110
)
[PubMed]
48
Swinnen
 
J.V.
Esquenet
 
M.
Goossens
 
K.
Heyns
 
W.
Verhoeven
 
G.
 
Androgens stimulate fatty acid synthase in the human prostate cancer cell line LNCaP
Cancer Res.
1997
, vol. 
57
 (pg. 
1086
-
1090
)
[PubMed]
49
Piyathilake
 
C.J.
Frost
 
A.R.
Manne
 
U.
Bell
 
W.C.
Weiss
 
H.
Heimburger
 
D.C.
Grizzle
 
W.E.
 
The expression of fatty acid synthase (FASE) is an early event in the development and progression of squamous cell carcinoma of the lung
Hum. Pathol.
2000
, vol. 
31
 (pg. 
1068
-
1073
)
[PubMed]
50
Visca
 
P.
Sebastiani
 
V.
Botti
 
C.
Diodoro
 
M.G.
Lasagni
 
R.P.
Romagnoli
 
F.
Brenna
 
A.
De Joannon
 
B.C.
Donnorso
 
R.P.
Lombardi
 
G.
Alo
 
P.L.
 
Fatty acid synthase (FAS) is a marker of increased risk of recurrence in lung carcinoma
Anticancer Res.
2004
, vol. 
24
 (pg. 
4169
-
4173
)
[PubMed]
51
Edmond
 
J.
Popjak
 
G.
 
Transfer of carbon atoms from mevalonate to n-fatty acids
J. Biol. Chem.
1974
, vol. 
249
 (pg. 
66
-
71
)
[PubMed]
52
Fogelman
 
A.
Edmond
 
J.
Popjåk
 
G.
 
Metabolism of mevalonate in rats and man not leading to sterols
J. Biol. Chem.
1975
, vol. 
250
 (pg. 
1771
-
1775
)
[PubMed]
53
Schwabauer
 
R.A.
Li
 
C.
Adams
 
G.
Gamble
 
W.
 
Utilization of mevalonate by aorta for the synthesis of medium-chain n-fatty acids (C8, C10) and acylglycerols
Biochim. Biophys. Acta
1988
, vol. 
960
 (pg. 
139
-
147
)
[PubMed]
54
El-Mir
 
M.Y.
Nogueira
 
V.
Fontaine
 
E.
Averet
 
N.
Rigoulet
 
M.
Leverve
 
X.
 
Dimethylbiguanide inhibits cell respiration via an indirect effect targeted on the respiratory chain complex I
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
223
-
228
)
[PubMed]
55
Ford
 
R.J.
Fullerton
 
M.D.
Pinkosky
 
S.L.
Day
 
E.A.
Scott
 
J.W.
Oakhill
 
J.S.
Bujak
 
A.L.
Smith
 
B.K.
Crane
 
J.D.
Blupsilonmer
 
R.M.
, et al 
Metformin and salicylate synergistically activate liver AMPK, inhibit lipogenesis and improve insulin sensitivity
Biochem. J.
2015
, vol. 
468
 (pg. 
125
-
132
)
[PubMed]
56
Scott
 
J.W.
Ling
 
N.
Issa
 
S.M.
Dite
 
T.A.
O’Brien
 
M.T.
Chen
 
Z.P.
Galic
 
S.
Langendorf
 
C.G.
Steinberg
 
G.R.
Kemp
 
B.E.
Oakhill
 
J.S.
 
Small molecule drug A-769662 and AMP synergistically activate naive AMPK independent of upstream kinase signaling
Chem. Biol.
2014
, vol. 
21
 (pg. 
619
-
627
)
[PubMed]
57
Ducommun
 
S.
Ford
 
R.J.
Bultot
 
L.
Deak
 
M.
Bertrand
 
L.
Kemp
 
B.E.
Steinberg
 
G.R.
Sakamoto
 
K.
 
Enhanced activation of cellular AMPK by dual-small molecule treatment: AICAR and A769662
Am. J. Physiol. Endocrinol. Metab.
2014
, vol. 
306
 (pg. 
E688
-
E696
)
[PubMed]