Critical role of arachidonic acid-activated mTOR signaling in breast carcinogenesis and angiogenesis

Article metrics

Abstract

The mammalian target of rapamycin (mTOR) signaling pathway is upregulated in the pathogenesis of many cancers. Arachidonic acid (AA) and its metabolites play critical role in the development of breast cancer, but the mechanisms through which AA promotes mammary tumorigenesis and progression are poorly understood. We found that the levels of AA and cytosolic phospholipase A2 (cPLA2) strongly correlated with the signaling activity of mTORC1 and mTORC2 as well as the expression levels of vascular epithelial growth factor (VEGF) in human breast tumor tissues. In cultured breast cancer cells, AA effectively activated both mTOR complex 1 (mTORC1) and mTORC2. Interestingly, AA-stimulated mTORC1 activation was independent of amino acids, phosphatidylinositol 3-kinase (PI3-K) and tuberous sclerosis complex 2 (TSC2), which suggests a novel mechanism for mTORC1 activation. Further studies revealed that AA stimulated mTORC1 activity through destabilization of mTOR–raptor association in ras homolog enriched in brain (Rheb)-dependent mechanism. Moreover, we showed that AA-stimulated cell proliferation and angiogenesis required mTOR activity and that the effect of AA was mediated by lipoxygenase (LOX) but not cyclooxygenase-2 (COX-2). In animal models, AA-enhanced incidences of rat mammary tumorigenesis, tumor weights and angiogenesis were inhibited by rapamycin. Our findings suggest that AA is an effective intracellular stimulus of mTOR and that AA-activated mTOR plays critical roles in angiogenesis and tumorigenesis of breast cancer.

Introduction

Tumorigenesis is often driven by disturbances of cellular signal transduction and information processing at the genetic and epigenetic levels. These disturbances are initially caused by environmental stimuli, that is, genotoxic and non-genotoxic carcinogens, whereas endogenous agents derived from malfunctioned metabolic reactions may take over at later stages, thereby leading to a state of ‘genetic instability’ and ‘growth autonomy’. Among those deregulated metabolic reactions, eicosanoid biosynthesis from arachidonic acid (AA) plays a major role in carcinogenesis.1, 2

AA represents one of the major polyunsaturated fatty acid in the mammalian cell membranes. When tissues are exposed to diverse physiological and pathological stimuli, such as growth factors, hormones or cytokines, AA is produced acutely from membrane phospholipids by the action of phospholipase A2 (PLA2) enzymes and released into cytosol. AA can then be enzymatically metabolized by three main pathways: cytochrome P450 monooxygenase, cyclooxygenases (COXs) and lipoxygenases (LOXs), producing prostaglandins, hydroxyeicosatetraenoic acids (HETEs) and epoxyeicosatrienoic acids, respectively.3, 4 AA and its eicosanoid metabolites are involved in the regulation of many cellular processes, such as cell survival, angiogenesis, chemotaxis, mitogenesis, apoptosis and migration. However, their mechanisms of actions are poorly understood.5, 6

Recent studies have suggested that AA and its metabolite pathway play critical role in the development and progression of breast cancer. Both case–control and cohort studies indicate a moderate reduction in breast cancer risk among women taking nonsteroidal anti-inflammatory drugs, which inhibit COX and thereby reduce prostaglandin synthesis.7, 8, 9 Also, 5-LOX and 12-LOX are overexpressed in breast cancer,10 and LOX inhibitor prevents N-methyl-N-nitrosourea (NMU)-induced mammary tumorigenesis in rats.11 AA stimulates cell proliferation and migration in some breast cancer cell lines.12, 13 However, the mechanism through which AA promotes mammary tumorigenesis and progression remains largely unknown.

The mammalian target of rapamycin (mTOR) is a highly conserved Ser/Thr kinase in the phosphatidylinositol 3-kinase (PI3-K)/Akt signaling pathway. It integrates diverse signals to control cell growth, proliferation, survival and metabolism.14, 15 mTOR elicits its pleiotropic functions in the context of two functionally distinct signaling complexes termed mTOR complex 1 (mTORC1) and complex 2 (mTORC2). mTORC1, which contains mTOR, mLST8/GβL, raptor, and proline-rich Akt substrate of 40 kDa (PRAS40), plays a key role in translation initiation by directly phosphorylating p70 S6 kinase 1 (S6K1) and 4E-BP1. Its activity is sensitive to rapamycin. mTORC2 shares mTOR and mLST8/GβL with mTORC1 but possesses three unique components, rictor, mSin1 and PRR5/Protor. Despite the presence of mTOR, mTORC2 is not susceptible to rapamycin inhibition. The function of mTORC2 is less clear, but it has been shown to phosphorylate Akt at serine 473.16, 17 mTOR signaling is upregulated in many cancers as a result of genetic alterations or aberrant activation of the components of PI3-K/Akt pathway, contributing to dysregulation of cell proliferation, growth, differentiation and survival. But extra- and intra-cellular stimuli derived from deregulated metabolic reactions that cause upregulation of mTOR signaling in cancer cells remain to be identified.18, 19 In the current study, we report for the first time that AA is a strong activator of mTOR1/2 signaling in breast cancer cells and that AA-stimulated mTOR activity is critical for angiogenesis and carcinogenesis of breast cancer.

Results

AA and cPLA2 levels correlate with mTORC1/2 signaling in human breast tumor tissues

Upregulation of both cPLA2/AA metabolic pathway and mTOR activity has been found in breast cancer and is believed to play important roles in the development and progression of the disease.5, 20, 21, 22 However, a causal link of cPLA2 and AA with mTORC1/2 signaling in cancers has not been reported. To examined whether AA and cPLA2 levels correlate with mTORC1/2 signaling in human breast cancer, we detected the expression of cPLA2, P-S6 (S235/236; downstream effector of mTORC1 and S6K1) and vascular epithelial growth factor (VEGF); (a key angiogenic factor whose expression is controlled by mTORC1) in 82 human primary breast tumor specimens (tissue array) by immunochemical staining. As shown in Figures 1a and b, P-S6 (S235/236) levels were strongly correlated with the cPLA2 expression (P<0.001). Consistent with this finding, cPLA2 expression was also associated with VEGF level (P<0.001; Figures 1a and b). Consistent with the findings from tissue array, AA levels in the freshly prepared lysates of human breast tumor samples were correlated with the levels of VEGF, cPLA2, P-S6 (S235/236) and P-Akt (S473) (substrate of mTORC2; Figure 1c–e). In addition, the percentages of positive staining for cPLA2, P-S6 (S235/236) and VEGF were significantly lower in normal tissues (16.1, 25.8 and 22.6%) than those in tumor tissues (71.0, 71.0 and 80.6%; Supplementary Figure S1), suggesting that expression of cPLA2 and phosphorylation of S6 (S235/236) is pathologically relevant in breast cancer. Taken together, these results demonstrate that AA and cPLA2 levels correlate with mTORC1/2 signaling in human breast cancer.

Figure 1
figure1

AA levels and cPLA2 expression correlate with mTOR signaling activity in human breast tumor tissues. (a) Immunohistochemical staining of cPLA2, phospho-S6 (S235/236) and VEGF expression in primary human breast cancer specimens. Shown are two representative specimens. (b) Left panel shows percentages of specimens with low (−) or high (+) cPLA2 expression in which VEGF was or was not observed. Middle and right panels show the percentages of specimens with low (−) or high (+) phospho-S6 (S235/236) in which cPLA2 or VEGF expression was or was not observed. (c) The expression levels of cPLA2, phospho-S6 (S235/236), S6, phospho-Akt (S473) and Akt in freshly prepared human breast tumor lysates were determined by immunoblotting. (d) Levels of AA and VEGF in human breast tumor lysates were determined by ELISA. (e) Correlation analyses between cPLA2, phospho-S6 (S235/236) and Akt (S473), VEGF and AA levels were performed by Spearman's rank correlation.

AA activates both mTORC1 and mTORC2 in breast cancer cells

The correlation of AA level with both mTORC1 and mTORC2 activity suggested that AA and its metabolites may contribute to the upregulation of the two mTOR complexes in breast cancer. We first examined whether AA activates mTORC1 in breast cancer cells. AA stimulated mTORC1-directed phosphorylation of S6K1 (T389), 4EBP1 (T37/46) and S6 (S235/235) in a dose-dependent (Figure 2a) and time-dependent (Figure 2b) manner in breast cancer MCF-7 cell. It is interesting that AA stimulation of mTORC1 was persistent through 120 min (Figure 2b), longer than reported serum or amino-acid stimulation of mTORC1 that generally attenuates after 45 min.18, 19 Furthermore, the effect of AA was independent of serum and nutrients, as AA was able to activate mTORC1 in cells starved for serum, amino acids or serum and amino acids (Figures 2a–c). Similarly, AA treatment also induced mTORC1 activation in other breast cancer cell lines, including BcaP 37, BT549 and T47D (Figure 2d) and in some normal cells (Supplementary Figure S2).

Figure 2
figure2

AA activates mTORC1 in breast cancer cells. Upon serum deprivation for 16 h or amino-acid starvation for 1 h, MCF-7 cells were treated with 30 μM of AA for indicated times (a) or 30 min with indicated concentrations of AA (b). The levels of phospho-S6K1 (T389), S6 (S235/236) and 4E-BP1 (T37/46) in cell lysates were determined by western blotting. (c) MCF-7 cells were serum deprived for 16 h and amino-acid starved for 1 h, and stimulated with 30 μM of AA for indicated times (c) or 30 min with indicated concentrations of AA. The levels of phospho-S6K1 (T389) and S6 (S235/236) were determined by western blotting. (d) The levels of phospho-S6 (S235/236) were determined by western blotting in cell lysates from breast cancer cells BcaP 37, BT549 and T47D treated with 30 μM of AA for various times after serum starvation for 16 h.

We next examined whether AA regulates the activity of mTORC2 by monitoring the mTORC2-directed phosphorylation of Akt at position Ser 473.18, 23 AA acutely stimulated phosphorylation of Akt (S473), which correlated with an enhanced phosphorylation of two Akt substrates, glycogen synthase kinase-3β (S9) and PRAS40 (T246) (Figure 3a). Pan PI3-K inhibitor Ly294002, which inhibits PI3-K and mTOR, or mTOR kinase inhibitor Pp242, which inhibits both mTORC1 and mTORC2, abolished AA-induced Akt phosphorylation (Figure 3b). Although activation of mTORC1 by AA induced S6K1-dependent negative feedback on PI-3K/Akt signaling, as manifested by phosphorylation of insulin receptor substrate-1 (S636/639) (Figure 3c), Akt (S473) was still stimulated by AA (Figure 3a). This observation suggests that AA-activated Akt may relieve the effect of the negative feedback on Akt caused by mTORC1 activation. Furthermore, knockdown of Rictor, the unique component of mTORC2 by small interfering RNA, also reduced the AA-stimulated Akt phosphorylation (Figure 3d). Taken together, these results indicate that in addition to mTORC1, AA also activates mTORC2 in breast cancer cells.

Figure 3
figure3

AA activates mTORC2 in breast cancer cells. (a) MCF-7 cells were serum starved for 16 h followed by treatment with 30 μM AA for various times. The levels of phospho-Akt (S473), glycogen synthase kinase-3β (GSK-3β; S9) and PRAS40 (T246) in the lysates were determined by western blotting. (b) MCF-7 cells were incubated with or without 100 nM rapamycin (Rap), 50 μM Ly294002 (Ly) or 200 nM Pp242 followed by treatment with 30 μM AA for 30 min. The levels of phospho-Akt (S473) were determined by western blotting. (c) MCF-7 cells were serum starved for 16 h followed by treatment with 30 μM AA for various times. The levels of phospho-IRS-1 (S636/639) were determined by western blotting. (d) MCF-7 cells transfected with mTOR or rictor-specific small interfering RNA (siRNA) were serum deprived for 16 h followed by stimulation with 30 μM AA for 30 min. Phospho-Akt (S473) levels were determined by western blotting.

AA-stimulated MCF-7 cell proliferation requires activation of mTOR signaling pathway

Several previous studies showed that AA and its metabolites stimulated growth and proliferation of breast cancer cells.24, 25 To find the role of mTOR signaling in AA-induced cell growth, we examined the effect of rapamycin on MCF-7 cell proliferation. We confirmed that AA dose-dependently stimulated MCF-7 cell proliferation from 0.1 to 5 μM (Figure 4a). However, the AA-stimulated growth was significantly suppressed by rapamycin (Figure 4b). The phosphorylation of Akt (S473) (mTORC2) was not inhibited by the prolonged rapamycin treatment (Supplementary Figure S3a). These results suggest that mTORC1 is required for AA-enhanced MCF-7 cell proliferation.

Figure 4
figure4

mTOR is required for AA-stimulated MCF-7 cell proliferation. (a) Proliferation of MCF-7 cells was assessed by Cell Counting Kit-8 after being treated for 72 h with indicated concentrations of AA. (b) Proliferation of MCF-7 cells was analyzed after the cells were treated with 5 μM of AA alone or in combination with 100 nM of rapamycin. *P<0.01 compared with mock control, #P<0.01 compared with mock control and AA treatment.

AA-stimulated angiogenesis in breast cancer is dependent on activation of mTOR

It has been reported that HETEs, the LOX metabolites of AA, stimulate PI3K/Akt/mTOR signaling and enhance angiogenesis in human vascular endothelial cell (HVECs).26, 27 To determine the role of AA-activated mTOR in angiogenesis of breast cancer, we examined the effect of rapamycin on production of two critical angiogenic factors, hypoxia-induced factor-α (HIF-α) and VEGF in MCF-7 cells. We found that AA increased HIF-α expression (Figure 5a) and VEGF secretion (Figure 5b). However, the effect of AA in MCF-7 cells was significantly reduced by pretreating the cells with rapamycin (Figures 5a and c), suggesting that the effect is mTOR dependent. To further confirm the role of mTOR signaling in AA-induced angiogenesis, we examined the effect of AA on mTOR signaling in HVECs. As expected, AA induced mTORC1-dependent phosphorylation of S6K1 and S6 in HVECs (Figure 5d) and significantly increased HVEC tube formation (Figure 5e). Prolonged rapamycin (50 nM) treatment inhibited mTORC1 (P-S6) but not mTORC2 (P-Akt) (Supplementary Figure S3b) and blocked AA-induced tube formation in HVECs (Figure 5e). Furthermore, in a chick chorioallantoic membrane angiogenesis assay, AA drastically increased capillary density, whereas rapamycin treatment blocked the effect of AA (Figure 5f and Supplementary Figure S4). Taken together, these results suggest that mTORC1 signaling is required for AA-stimulated angiogenesis in breast cancer.

Figure 5
figure5

mTOR is required for AA-promoted angiogenesis. (a) MCF-7 cells maintained in conditioned medium were stimulated with 0–20 μM of AA. Hypoxia-induced factor-α (HIF-α) in the lysates was detected by western blotting. (b) Cells were stimulated with 5 μM of AA in the presence of 50 nM of rapamycin (Rap), and VEGF in culture medium (supernatant) was detected by ELISA. (c) Cells were stimulated with 5 μM of AA in the presence or absence of 50 nM rapamycin (Rap), and HIF-α in cell lysates was detected by western blotting. (d) MCF-7cells were treated for 30 min with various doses of AA and the levels of phospho-S6K1 (T389) and S6 (S235/236) were determined by western blotting. (e) mTOR activity is required for AA-stimulated tube formation. HVEC was stimulated with 5 μM of AA in the presence of 50 nM of rapamycin and then assessed for tube formation. (f) Effect of rapamycin (50 nM) on AA (10 μM)-stimulated chick chorioallantoic membrane angiogenesis was examined using an in vivo chick CAM assay. #P<0.01 compared with other groups.

LOXs but not COXs mediate the stimulatory effects of AA on mTOR, cell proliferation and angiogenesis

It has been well established that LOXs, COXs and their metabolites play important roles in the development and progression of breast cancer, but the mechanisms are poorly understood.9, 10 In an attempt to determine the pathway contributing to AA-induced activation of mTOR signaling, we found that nordihydroguaiaretic acid (NDGA), a LOX inhibitor, but not NS398, a COX-2 inhibitor, blocked AA-stimulated P-S6 (S235/236) and P-Akt (S473) in MCF-7 cells (Figure 6a), suggesting that LOX is required for AA-induced activation of mTOR. Furthermore, 5- and 12-HETE, the two LOX metabolites of AA, also effectively induced phosphorylation of S6 and Akt in MCF-7 cells (Figure 6b). MK886, inhibitor of 5-LOX, or Baicalein, inhibitor of 12-LOX, partially reduced the AA-induced phosphorylation (Figure 6c). In addition, AA-stimulated MCF-7 cell proliferation and HVEC tube formation were inhibited by LOX inhibitor NDGA, but not by COX-2 inhibitor NS398 (Figures 6d–f). These data indicate that LOX and its metabolites mediate the effects of AA on mTOR signaling, cell proliferation and angiogenesis.

Figure 6
figure6

The effects of AA on mTOR pathway, cell proliferation and angiogenesis are dependent on lipoxygenases. (a) MCF-7 cells were treated with 30 μM AA for 30 min in the presence of indicated doses of NDGA or NS398 and the levels of phospho-S6 and Akt in the lysates were determined by western blotting. (b) Cells were serum starved for 16 h followed by treatment for 30 min with 30 μM of 5- or 12-HETE. The levels of phospho-S6 were determined by western blotting. (c) MCF-7cells were serum starved for 16 h followed by 30 μM of AA treatment for 30 min in the presence or absence of MK886 or Baicalein. The levels of phospho-S6 and Akt in cell lysates were determined by western blotting. (d) MCF-7 cells were treated with 5 μM of AA in the presence or absence of NDGA (15 μM), rapamycin (50 nM) or NS398 (30 μM), and cell proliferation was analyzed by Cell Counting Kit-8. (e) HVEC was stimulated with 5 μM of AA in the presence or absence of NDGA (15 μM), rapamycin (50 nM) or NS398 (30 μM), and tube formation was measured by in vitro tube formation assay. Shown are typical images. (f) HVEC tube length in (e) was measured and quantified. #P<0.01 compared with mock control. *P<0.01 compared with NS398 groups.

AA activates mTORC1 by destabilization of mTOR–raptor association in a Rheb-dependent mechanism

We next asked the mechanism through which AA and its metabolites activate mTORC1. Repression of AA-induced P-S6K1 (T389) and P-S6 (S235/236) by mTOR, raptor or S6K1 small interfering RNA supported the notion that AA-induced P-S6K1 (T389) and P-S6 (S235/236) requires mTORC1 (Figures 7a and b). The kinase activity of mTORC1 from cells starved for amino acids was also strongly stimulated by AA, supporting the notion that mTORC1 kinase activity was activated by AA (Figure 7c).

Figure 7
figure7

AA destabilizes mTOR–raptor association and activates mTORC1 dependent of Rheb. (a) MCF-7 cells transfected with small interfering RNAs (siRNAs) specific for mTOR or S6K1 were serum starved for 16 h followed by stimulation with 30 μM of AA for 30 min. The levels of the phospho-S6K1 and S6 in the cell lysates were determined by western blotting. (b) MCF-7 cells were transfected with raptor siRNA and treated as in (a), and the levels of the phospho-S6 were determined by western blotting. (c) MCF-7 cells were starved for amino acids for 1 h followed by treatment with or without 30 μM of AA for 30 min or with 100 nM of rapamycin for 1 h. mTORC1 was immunopurified from the cell lysates and assayed for its kinase activity in an in vitro kinase assay. (d) MCF-7 cells transfected with p85 subunit of PI3-K siRNA were treated as in (a), and the levels of the phospho-S6 and Akt were determined by western blotting. (e) TSC+/+ and TSC−/− MEFs were serum starved for 16 h followed by stimulation with 30 μM of AA for 30 min. The levels of the phospho-S6 and Akt were determined by western blotting. (f) MCF-7 cells transfected with Rheb siRNA were treated as in (a), and the levels of phospho-S6 and Akt were determined by western blotting. (g) MCF-7 cells were starved for serum (16 h) and amino acids (1 h) followed by treatment with or without 30 μM of biotin-labeled AA, 5-HETE or 12-HETE for 30 min, mTORC1/2 was immunopurified from the cell lysates, or AA and HETEs were added in vitro instead of cell culture, and mTORC1/2 were then assayed for in vitro kinase activity.

It has been well established that mitogens such as insulin and growth factors activate mTORC1 via PI3-K/Akt/tuberous sclerosis complex 2 (TSC2)-dependent pathway.14, 28 But stimulation of mTORC1 through inactivation of TSC2, which integrates diverse signals to negatively control mTORC1, is dependent on the presence of amino acids.29, 30, 31 However, we found that AA was able to stimulate mTORC1 in the absence of amino acids (Figures 2a–c), which suggests a novel mechanism for mTORC1 activation. In addition, we observed that knockdown of p85α, a regulatory subunit of PI3-K by small interfering RNA, did not prevent AA-induced P-S6 (S235/236) and P-Akt (S473) (Figure 7d). Furthermore, AA induced phosphorylation of S6 in both TSC2 knockout (TSC2−/−) and wild-type (TSC2+/+) mouse embryo fibroblast cells (MEFs) (Figure 7e). It is suggested that a novel mechanism independent of PI3-K/Akt/TSC pathway is responsible for AA-stimulated mTORC1 activity.

It has been reported that ras homolog enriched in brain (Rheb), a small GTPase, is required for all major upstream signals to activate mTORC1.18, 19 Indeed, Rheb knockdown suppressed AA-stimulated P-S6 (S235/236; Figure 7f), suggesting the requirement of Rheb in AA stimulation of mTORC1. It has been shown that mTORC1 activity is regulated by mTOR–raptor interaction. Amino-acid starvation stabilizes mTOR–raptor association and inactivates mTORC1, whereas amino-acid stimulation destabilizes mTOR–raptor association and activates mTORC1.32 Interestingly, AA also destabilized mTOR–raptor association and stimulated mTORC1 kinase activity in the absence of amino acids (Figure 7b). To examine if AA or its metabolites directly bind to mTOR complex and stimulate mTORC1/2 activity in vitro, biotin-labeled AA or its LOX metabolites 5′-HETE and 12′-HETE were either added into cell cultures or added in vitro. It was found that AA, 5′-HETE or 12′-HETE in cells but not in vitro impaired mTOR–raptor association and stimulated mTORC1 and mTORC2 kinase activity (Figure 7g). The stability of mTORC2 (mTOR–rictor interaction) was not affected by AA and HETE, both in cells and in vitro (Figure 7g). Consistent with the results of kinase assay, AA, 5′-HETE or 12′-HETE were unable to directly bind to mTORC1/2 (Supplementary Figure S5). They were also unable to bind to Rheb (Supplementary Figure S5).

DEPTOR is a protein that binds to and negatively regulates both mTORC1 and mTORC2.33, 34 The possible role of DEPTOR in AA-activated mTOR was further examined. Although DEPTOR level was decreased by AA stimulation, the downregulation occurred 90 min after AA stimulation whereas mTORC1 could be activated by AA within 5 min. Furthermore, the degradation of DEPTOR was prevented by rapamycin (Supplementary Figure S6), suggesting that DEPTOR degradation is downstream of mTORC1 activation. This result is consistent with recent findings that serum (growth factors) stimulated RSK1 and mTORC1-S6K1 to phosphorylate DEPTOR, which is recognized and degraded by SKP1-cullins-F box protein (SCF) E3 ligase.33, 34 The association of DEPTOR with mTOR was also unaffected by AA (Supplementary Figure S6). These results indicate that DEPTOR is not a mediator of AA-stimulated mTOR.

AA promotes rat mammary tumorigenesis and angiogenesis via activation of mTOR pathway

We next asked whether AA-stimulated mTOR signaling contributes to mammary tumorigenesis and angiogenesis. NMU-induced rat mammary tumorigenesis assay revealed that AA application significantly enhanced the incidence of NMU-induced rat mammary tumorigenesis, whereas rapamycin treatment profoundly antagonized the effect of AA (Figure 8a). AA-enhanced tumor weight was also decreased by rapamycin treatment (Figure 8b). Furthermore, AA stimulated mTORC1 whereas rapamycin suppressed AA-activated mTORC1 in vivo, as manifested by inhibition of phosphorylation of S6 (Figure 8c). Moreover, rapamycin decreased basal level and AA-induced VEGF expression in mammary tumor tissues (Figure 8d). These data suggest that mTOR pathway is critical for AA-promoted mammary tumorigenesis and angiogenesis.

Figure 8
figure8

AA promotes rat mammary tumorigenesis and angiogenesis via activation of the mTOR pathway. (a) A total of 40 Sprague-Dawley (SD) rats were induced with NMU and randomized into four groups (n=10 for each group), followed by treatment with AA alone or in combination with rapamycin as described in the Materials and methods section. The incidence of tumorigenesis was examined and calculated. (b) Effect of AA and rapamycin on tumor weights described in (a) was measured. #P<0.01 compared with other groups. (c) The levels of phospho-S6 (S235/236) in lysates of the tumor tissues from (a) were examined by western blotting. S6 and actin were used as loading controls. (d) Effects of AA and rapamycin on VEGF expression in the tumor tissues from (a) were detected by ELISA. A full colour version of this figure is available at the Oncogene journal online.

Discussion

In the mammary gland, plasma-free fatty acids are used as an important source of energy and for milk lipid synthesis; however, epidemiological studies and other lines of evidence indicate an association of dietary fatty acids with the development of breast cancer.35, 36 However, the mechanisms of their actions are poorly understood. AA, an essential fatty acid from the diet or synthesized from linoleic acid, plays a critical role in mammary tumorigenesis.7, 10 Overexpression of PLA2, which produces AA from membranous phospholipids, has also been reported in a variety of human cancers including breast cancer.37 In the present study, we found that AA and cPLA2 levels strongly correlate with mTORC1 activity and VEGF levels in human breast cancer tissues, indicating a causal connection with clinical relevance between the AA metabolic pathway and mTOR signaling (Figure 1). Our in vitro studies revealed that AA is a strong activator of both mTORC1 and mTORC2 and that mTOR activity is required for AA-enhanced angiogenesis and breast cancer cell proliferation. Most importantly, administration of AA promoted NMU-induced rat mammary tumorigenesis and angiogenesis, whereas rapamycin prevented the action of AA. Together, these in vivo and in vitro observations provide evidence for the first time that AA-activated mTOR contributes to the development and progression of breast cancer.

Both nonsteroidal anti-inflammatory drugs and isozyme-specific COX-2 inhibitors have been shown to inhibit experimental and human cancer development.7, 8, 9 However, our results revealed that AA stimulates mTOR signaling and promotes angiogenesis and cell proliferation via LOX but not COX-2 in MCF-7 cells (Figure 7). These results are consistent with previous findings from the group of Zhang et al.,26, 27 demonstrating that LOX metabolites of AA promote angiogenesis via activation of PI3-K/Akt/mTOR pathway in human vascular smooth muscle cells. Elevated levels of both 5-LOX and 12-LOX have been found to be associated with higher TNM (tumor, node, metastasis) breast cancer staging.10 In a study involving cigarette smoking-promoted colon cancer formation, 5-LOX inhibitor is more effective than COX-2 inhibitor in the inhibition of colon cancer development.38 These observations indicate critical roles of LOX in tumorigenesis and implicate that inhibition LOX could represent therapeutic means to treat cancers associated with dietary fatty acids (AA).

Several signaling mechanisms have been found to arbitrate the biological functions of AA in breast cancer cells, such as reactive oxygen species, G-protein-coupled receptors, extracellular-signal-regulated kinase, src and focal activation kinase.12, 13, 39, 40 However, we find that AA-induced mTORC1 activation is not affected in cells treated with inhibitors of these mechanisms (Supplementary Figure S7). It is interesting that AA induced a rapid (5 min) and persistent (120 min) phosphorylation of S6K1/S6. It was found that LOX metabolites HETEs may mediate the action of AA. Previously, observations have shown that AA may be uptaken and metabolized to induce downstream effects in several minutes.41, 42 We also observed that 5- or 12-HETE stimulated P-S6 (S235/236) within 2 min (Supplementary Figure S8). Consistent with our results, the persistent activation of S6K1 by 15(S)-HETE (120 min) has been reported in vascular smooth muscle cells.26, 27 To maintain the persistent mTORC1 activation, a positive feedback mechanism may exist when mTORC1 is stimulated by these eicosanoids, or other eicosanoids besides 5- and 12-HETE produced from AA contribute to the activation. Future study will test these possibilities.

Regulation of mTORC1 is complex and involves several distinct pathways that impart various signals to mTORC1 function. One key factor for the regulation is the TSC1/TSC2 tumor-suppressor complex, which mediates signals from the PI3-K/Akt, Ras/extracellular-signal-regulated kinase and LKB1/AMPK pathways.14, 15 However, our findings that TSC2 is dispensable for AA-mediated mTORC1 activation suggests that AA does not act through these mechanisms. We found that AA-activated mTORC1 is dependent on Rheb, the upstream regulator of mTORC1, which appears to mediate all major signals including amino acid to activate mTORC1.18, 19 Amino acids reportedly activate mTORC1 through the Rag GTPases, which promote the translocation of mTORC1 to the lysosomal surface, the site of mTORC1 activation by Rheb.30, 31 However, our result showed that AA is able to stimulate mTORC1/2 activity under the serum and amino-acid starvation condition. It is possible that AA and its metabolites act like amino acids to activate Rag GTPase or other unknown molecules and promote the translocation of mTORC1 to lysosomes when cells are deprived of amino acids. AA stimulation renders the cells more resistant to amino-acid starvation. Association of raptor with mTOR, which negatively regulates mTORC1 activity and is destabilized by amino acids,32 was also impaired upon AA and its LOX metabolites stimulation, supporting the possibility that AA may act like amino acids when cells undergo amino-acid starvation. Taken together, our results indicate that AA and HETEs activate mTORC1 through Rheb and destabilization of mTOR–raptor association.

Upregulation of the mTOR signaling pathway is implicated in breast cancer development and progression.22 However, the environmental factors that induce sustained activation of mTOR in breast cancer are not clear. AA and its metabolites have recently generated a heightened interest because of growing evidence of their significant role in malignant cell proliferation, cell survival, tumor metastasis and neoangiogenesis,43 which are also regulated by mTOR signaling. Deletion of cPLA2 in mice produces less AA and prevents lung and glioblastoma tumorigenesis and angiogenesis.44, 45, 46 Some of AA metabolites such as 5- and 15-HETE and prostaglandin E2 have been shown to activate mTOR.26, 27, 47 Our results that levels of AA and cPLA2 correlated with the activity of mTORC1/2 in human breast tumor tissues suggest the link between AA, mTOR and breast carcinogenesis. High levels of cPLA2 may produce more AA, and is subsequently metabolized into HETEs and other eicosanoids, which activate mTORC1 through destabilization of mTOR–raptor association in a Rheb-dependent, PI3-K/TSC-independent mechanism. Hyperactivation of mTOR will stimulate cell proliferation and promote angiogenesis and breast tumorigenesis.

In summary, our study provide in vivo and in vitro evidences that AA is an effective activator of mTOR1/2 signaling and that AA-stimulated mTOR activity plays a critical role in angiogenesis and tumorigenesis of breast cancer.

Materials and methods

Materials and RNA interference

See Supplementary Information.

Human breast tumor tissue arrays, breast tumor samples and immunohistochemical staining

Human breast tumor and normal tissue arrays were purchased from Shanghai Outdo Biotech. Co. Ltd (Shanghai, China). The tissue specimens were incubated with antibodies to cPLA2, p-S6 (S235/236) and VEGF together with biotin-conjugated secondary antibody and then incubated with avidin-biotin-peroxidase complex, and visualization was performed with DAB (3, 3N-diaminobenzidine tertrahydrochloride). The fresh human breast tumor samples for western blotting and AA or VEGF enzyme-linked immunosorbent assay (ELISA) were collected from Breast Center, Nanfang Hospital of Southern Medical University in China. The immunoreactivities for cPLA2, P-S6 (S235/236) and VEGF were semiquantitatively scored using a well-established immunoreactivity score system in which immunoreactivity score was generated by incorporating both the percentage of positive tumor cells and the intensity of staining.48 The study was approved by the Ethical Committee of Southern Medical University and written informed consent was obtained from all patients.

Cell culture

Human breast cancer cell lines MCF-7 and T47D (obtained from American Type Culture Collection (ATCC), Manassas, VA, USA), primary cultured HVECs, TSC2+/+ and TSC2−/− mouse embryo fibroblast cells (kindly provided by Dr David J Kwiatkowski in Brigham and Women's Hospital, Boston, MA, USA) were cultured in high glucose Dulbecco's modified Eagle's medium. BcaP 37 and BT549 (obtained from ATCC) were maintained in RPMI-1640. The media were all supplemented with 10% fetal bovine serum, 50 units/ml penicillin and 50 μg perstreptomycin in a humidified atmosphere of 5% CO2.

Cell proliferation assays

Cells (104/well) were plated in 96-well plates and allowed to attach for 24 h, and then cultured under AA (0.1–5 μM) and/or inhibitors in culture medium for 48 h. Cell viability was assessed using Cell Counting Kit-8 (WST-8) (Dojindo Molecular Technologies Inc., Kumamoto, Japan) following the manufacture's instructions.

Immunoprecipitation, immunoblotting and in vitro kinase assays

Immunoprecipitation, immunoblotting and in vitro mTORC1/2 kinase assay were performed as previously described.32, 33, 49, 50, 51 Briefly, 80% confluent MCF-7 cells were starved for serum (16 h) and amino acids (1 h), followed by treatment with or without 30 μM of biotin-labeled AA, 5-HETE or 12-HETE for 30 min. mTORC1/2 was immunopurified from the cell lysates with anti-mTOR antibody, or biotin-labeled AA and HETEs were added into immunopurified mTOR complex in vitro instead of cell culture. mTORC1/2 were then assayed for in vitro kinase assay using the purified GST-4EBP1 (mTORC1) or inactive Akt (mTORC2) as a substrate. Immunopurified mTORC1/2 were also subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and blotted with horseradish peroxidase-conjugated streptavidin.

Tube formation assays

The 96-well culture plates were coated with 20 μl of growth factor-reduced Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) and allowed to solidify for 30 min at 37 °C. HVECs were trypsnized and resuspended at density of 5 × 104/ml, and 150 μl of this cell suspension was added into each well. Vehicle or AA and inhibitors at the indicated concentrations were added to the appropriate wells and the cells were incubated at 37 °C for 6 h. Tube formation was observed under an inverted microscope. Images were captured with a CCD color camera attached to the microscope and tube length was measured using the NIH Image J 1.31v Program (NIH, Bethesda, MD, USA).

In vivo chick chorioallantoic membrane (CAM) assays

Fertilized chick embryos were preincubated for 8 days at 37.5 °C in 85% humidity. A hole was drilled over the air sac at the end of the egg and the vascular zone was identified on the CAM. A 1 × 1 cm window in the shell was sectioned to expose the CAM. Sterilized filter-paper disks were loaded with 10 μM AA and/or 100 nM rapamycin and applied to the CAM surface. Upon sealing the openings with clear tape, the eggs were further incubated for 48 h. Blood vessels were viewed and photographed. Pro- or anti-angiogenic effects of AA and rapamycin on CAMs were quantified by counting the number of blood vessel branch points.

VEGF secretion

The concentration of VEGF in the medium of control and treated cells or breast tumor tissue lysates were measured using commercially available human and rat VEGF sandwich ELISA kit according to the manufacturer's instructions (Dakewe Biotech Company Ltd., Shenzhen, China). Briefly, cells were plated in 24-well plates and treated with different concentration of AA or inhibitors in serum-free conditioned medium. After 48 h, the media were collected and tested by immunoassay kit. Results were normalized to the number of cells counted after exposure and reported as pg of VEGF protein per 106 cells or ng of VEGF protein per g tumor tissue lysates.

Rat mammary tumorigenesis assay

Female Sprague-Dawley rats (43 days old), purchased from Experimental Animal Care Center of Southern Medical University (Guangzhou, China), were housed at 22 °C, 50% humidity with a 12-h light/dark cycle. Tap water was provided ad libitum throughout the experiment. At the age of 50 days, they were given a single intraperitoneal injection of 50 mg/kg NMU (Sigma Chemical, St Louis, MO, USA), dissolved in 0.05% acetic acid in normal saline and used within 30 min of preparation. A week later, the animals were randomized into control and experimental groups (10/group). AA (6 g/kg/day), rapamycin (1 mg/kg/day), or both AA and rapamycin were given p.o. once every 24 h. Animals were weighed and palpated for mammary lesions weekly. At the end of the experiment, the animals were anesthetized with a 1.5% isofluorane/air mixture and killed by cervical dislocation. Tumors were excised, weighed and sectioned. Tumor sections were fixed in 10% buffered formalin for hematoxylin and eosin staining or immunohistochemical assay. Mammary tumors were also homogenized on ice in a glass tissue grinder with phosphate-buffered saline containing Phosphatase and Protease Inhibitor Cocktail (Calbiochem Chemicals, San Diego, CA, USA). The homogenates were centrifuged and the supernatants were used for ELISA or western analysis. All animal protocols were approved by the Southern Medical University Animal Care and Use Committee in accordance with the guidelines for the ethical treatment of animals.

AA analysis

AA levels in human breast tumor lysates were measured using human AA ELISA kit from Cusabio Biotech Co., Ltd (Wuhan, China) following the manufacture's protocol.

Statistical analysis

All the experiments were repeated thrice with similar results. Data on proliferation assay, tube formation, ELISA and angiogenesis assay are presented as mean±s.d. The treatment effects were analyzed by one-way analysis of variance and P<0.05 was considered statistically significant. Correlation analyses were performed by Spearman's rank correlation. In the case of western blot analysis, one representative set of data is shown.

References

  1. 1

    Chen X, Sood S, Yang CS, Li N, Sun Z . Five-lipoxygenase pathway of arachidonic acid metabolism in carcino-genesis and cancer chemoprevention. Curr Cancer Drug Targets 2006; 6: 613–622.

  2. 2

    Horn L, Backlund M, Johnson DH . Targeting the eicosanoid pathway in non-small-cell lung cancer. Expert Opin Ther Targets 2009; 13: 675–688.

  3. 3

    Harizi H, Corcuff JB, Gualde N . Arachidonic-acid-derived eicosanoids: roles in biology and immunopathology. Trends Mol Med 2008; 14: 461–469.

  4. 4

    Kim C, Kim JY, Kim JH . Cytosolic phospholipase A(2), lipoxygenase metabolites, and reactive oxygen species. BMB Rep 2008; 41: 555–559.

  5. 5

    Nakanishi M, Rosenberg DW . Roles of cPLA2alpha and arachidonic acid in cancer. Biochim Biophys Acta 2006; 1761: 1335–1343.

  6. 6

    Menna C, Olivieri F, Catalano A, Procopio A . Lipoxygenase inhibitors for cancer prevention: promises and risks. Curr Pharm Des 2010; 16: 725–733.

  7. 7

    Agrawal A, Fentiman IS . NSAIDs and breast cancer: a possible prevention and treatment strategy. Int J Clin Pract 2008; 62: 444–449.

  8. 8

    Singh-Ranger G, Salhab M, Mokbel K . The role of cyclooxygenase-2 in breast cancer: review. Breast Cancer Res Treat 2008; 109: 189–198.

  9. 9

    Ulrich CM, Bigler J, Potter JD . Non-steroidal anti-inflammatory drugs for cancer prevention: promise, perils and pharmacogenetics. Nat Rev Cancer 2006; 6: 130–140.

  10. 10

    Jiang WG, Douglas-Jones AG, Mansel RE . Aberrant expression of 5-lipoxygenase-activating protein (5-LOXAP) has prognostic and survival significance in patients with breast cancer. Prostaglandins Leukot Essent Fatty Acids 2006; 74: 125–134.

  11. 11

    McCormick DL, Spicer AM . Nordihydroguaiaretic acid suppression of rat mammary carcinogenesis induced by N-methyl-N-nitrosourea. Cancer Lett 1987; 37: 139–146.

  12. 12

    Navarro-Tito N, Robledo T, Salazar EP . Arachidonic acid promotes FAK activation and migration in MDA-MB-231 breast cancer cells. Exp Cell Res 2008; 314: 3340–3355.

  13. 13

    Kennett SB, Roberts JD, Olden K . Requirement of protein kinase C micro activation and calpain-mediated proteolysis for arachidonic acid-stimulated adhesion of MDA-MB-435 human mammary carcinoma cells to collagen type IV. J Biol Chem 2004; 279: 3300–3307.

  14. 14

    Sengupta S, Peterson TR, Sabatini DM . Regulation of the mTOR complex 1 pathway by nutrients, growth factors, and stress. Mol Cell 2010; 40: 310–322.

  15. 15

    Bai X, Jiang Y . Key factors in mTOR regulation. Cell Mol Life Sci 2010; 67: 239–253.

  16. 16

    Zoncu R, Efeyan A, Sabatini DM . mTOR: from growth signal integration to cancer, diabetes and ageing. Nat Rev Mol Cell Biol 2011; 12: 21–35.

  17. 17

    Foster KG, Fingar DC . Mammalian target of rapamycin (mTOR): conducting the cellular signaling symphony. J Biol Chem 2010; 285: 14071–14077.

  18. 18

    Efeyan A, Sabatini DM . mTOR and cancer: many loops in one pathway. Curr Opin Cell Biol 2010; 22: 169–176.

  19. 19

    Guertin DA, Sabatini DM . Defining the role of mTOR in cancer. Cancer Cell 2007; 12: 9–22.

  20. 20

    Martinez-Orozco R, Navarro-Tito N, Soto-Guzman A, Castro-Sanchez L, Perez Salazar E . Arachidonic acid promotes epithelial-to-mesenchymal-like transition in mammary epithelial cells MCF10A. Eur J Cell Biol 2010; 89: 476–488.

  21. 21

    Pidgeon GP, Lysaght J, Krishnamoorthy S, Reynolds JV, O’Byrne K, Nie D et al. Lipoxygenase metabolism: roles in tumor progression and survival. Cancer Metastasis Rev 2007; 26: 503–524.

  22. 22

    Hynes NE, Boulay A . The mTOR pathway in breast cancer. J Mammary Gland Biol Neoplasia 2006; 11: 53–61.

  23. 23

    Jacinto E, Loewith R, Schmidt A, Lin S, Ruegg MA, Hall A et al. Mammalian TOR complex 2 controls the actin cytoskeleton and is rapamycin insensitive. Nat Cell Biol 2004; 6: 1122–1128.

  24. 24

    Razanamahefa L, Prouff S, Bardon S . Stimulatory effect of arachidonic acid on T-47D human breast cancer cell growth is associated with enhancement of cyclin D1 mRNA expression. Nutr Cancer 2000; 38: 274–280.

  25. 25

    Cakir Y, Plummer III HK, Tithof PK, Schuller HM . Beta-adrenergic and arachidonic acid-mediated growth regulation of human breast cancer cell lines. Int J Oncol 2002; 21: 153–157.

  26. 26

    Zhang B, Cao H, Rao GN . Fibroblast growth factor-2 is a downstream mediator of phosphatidylinositol 3-kinase-Akt signaling in 14,15-epoxyeicosatrienoic acid-induced angiogenesis. J Biol Chem 2006; 281: 905–914.

  27. 27

    Zhang B, Cao H, Rao GN . 15(S)-hydroxyeicosatetraenoic acid induces angiogenesis via activation of PI3K-Akt-mTOR-S6K1 signaling. Cancer Res 2005; 65: 7283–7291.

  28. 28

    Yecies JL, Manning BD . mTOR links oncogenic signaling to tumor cell metabolism. J Mol Med 2011; 89: 221–228.

  29. 29

    Mieulet V, Lamb RF . Tuberous sclerosis complex: linking cancer to metabolism. Trends Mol Med 2010; 16: 329–335.

  30. 30

    Sancak Y, Peterson TR, Shaul YD, Lindquist RA, Thoreen CC, Bar-Peled L et al. The Rag GTPases bind raptor and mediate amino acid signaling to mTORC1. Science 2008; 320: 1496–1501.

  31. 31

    Goberdhan DC . Intracellular amino acid sensing and mTORC1-regulated growth: new ways to block an old target? Curr Opin Investig Drugs 2010; 11: 1360–1367.

  32. 32

    Kim DH, Sarbassov DD, Ali SM, King JE, Latek RR, Erdjument-Bromage H et al. mTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machinery. Cell 2002; 110: 163–175.

  33. 33

    Duan S, Skaar JR, Kuchay S, Toschi A, Kanarek N, Ben-Neriah Y et al. mTOR generates an auto-amplification loop by triggering the betaTrCP- and CK1alpha-dependent degradation of DEPTOR. Mol Cell 2011; 44: 317–324.

  34. 34

    Peterson TR, Laplante M, Thoreen CC, Sancak Y, Kang SA, Kuehl WM et al. DEPTOR is an mTOR inhibitor frequently overexpressed in multiple myeloma cells and required for their survival. Cell 2009; 137: 873–886.

  35. 35

    Holmes MD, Hunter DJ, Colditz GA, Stampfer MJ, Hankinson SE, Speizer FE et al. Association of dietary intake of fat and fatty acids with risk of breast cancer. JAMA 1999; 281: 914–920.

  36. 36

    Thiebaut AC, Chajes V, Gerber M, Boutron-Ruault MC, Joulin V, Lenoir G et al. Dietary intakes of omega-6 and omega-3 polyunsaturated fatty acids and the risk of breast cancer. Int J Cancer 2009; 124: 924–931.

  37. 37

    Yamashita S, Yamashita J, Ogawa M . Overexpression of group II phospholipase A2 in human breast cancer tissues is closely associated with their malignant potency. Br J Cancer 1994; 69: 1166–1170.

  38. 38

    Ye YN, Wu WK, Shin VY, Bruce IC, Wong BC, Cho CH . Dual inhibition of 5-LOX and COX-2 suppresses colon cancer formation promoted by cigarette smoke. Carcinogenesis 2005; 26: 827–834.

  39. 39

    Paine E, Palmantier R, Akiyama SK, Olden K, Roberts JD . Arachidonic acid activates mitogen-activated protein (MAP) kinase-activated protein kinase 2 and mediates adhesion of a human breast carcinoma cell line to collagen type IV through a p38 MAP kinase-dependent pathway. J Biol Chem 2000; 275: 11284–11290.

  40. 40

    Navarro-Tito N, Soto-Guzman A, Castro-Sanchez L, Martinez-Orozco R, Salazar EP . Oleic acid promotes migration on MDA-MB-231 breast cancer cells through an arachidonic acid-dependent pathway. Int J Biochem Cell Biol 2010; 42: 306–317.

  41. 41

    Buzzard CJ, Pfister SL, Campbell WB . Endothelium-dependent contractions in rabbit pulmonary artery are mediated by thromboxane A2. Circ Res 1993; 72: 1023–1034.

  42. 42

    Zhou L, Vessby B, Nilsson A . Quantitative role of plasma free fatty acids in the supply of arachidonic acid to extrahepatic tissues in rats. J Nutr 2002; 132: 2626–2631.

  43. 43

    Hyde CA, Missailidis S . Inhibition of arachidonic acid metabolism and its implication on cell proliferation and tumour-angiogenesis. Int Immunopharmacol 2009; 9: 701–715.

  44. 44

    Linkous AG, Yazlovitskaya EM, Hallahan DE . Cytosolic phospholipase A2 and lysophospholipids in tumor angiogenesis. J Natl Cancer Inst 2010; 102: 1398–1412.

  45. 45

    Meyer AM, Dwyer-Nield LD, Hurteau GJ, Keith RL, O’Leary E, You M et al. Decreased lung tumorigenesis in mice genetically deficient in cytosolic phospholipase A2. Carcinogenesis 2004; 25: 1517–1524.

  46. 46

    Weiser-Evans MC, Wang XQ, Amin J, Van Putten V, Choudhary R, Winn RA et al. Depletion of cytosolic phospholipase A2 in bone marrow-derived macrophages protects against lung cancer progression and metastasis. Cancer Res 2009; 69: 1733–1738.

  47. 47

    Kuehn HS, Jung MY, Beaven MA, Metcalfe DD, Gilfillan AM . Prostaglandin E2 activates and utilizes mTORC2 as a central signaling locus for the regulation of mast cell chemotaxis and mediator release. J Biol Chem 2011; 286: 391–402.

  48. 48

    Camp RL, Rimm EB, Rimm DL . Met expression is associated with poor outcome in patients with axillary lymph node negative breast carcinoma. Cancer 1999; 86: 2259–2265.

  49. 49

    Li M, Zhao L, Liu J, Liu A, Jia C, Ma D et al. Multi-mechanisms are involved in reactive oxygen species regulation of mTORC1 signaling. Cell Signal 2010; 22: 1469–1476.

  50. 50

    Bai X, Ma D, Liu A, Shen X, Wang QJ, Liu Y et al. Rheb activates mTOR by antagonizing its endogenous inhibitor, FKBP38. Science 2007; 318: 977–980.

  51. 51

    Sarbassov DD, Ali SM, Sengupta S, Sheen JH, Hsu PP, Bagley AF et al. Prolonged rapamycin treatment inhibits mTORC2 assembly and Akt/PKB. Mol Cell 2006; 22: 159–168.

Download references

Acknowledgements

We greatly appreciate the gift of TSC2+/+ and TSC2−/− MEFs from Dr David J Kwiatkowski (Brigham and Women's Hospital, Boston, MA, USA). This work was supported by The State Key Development Program for Basic Research of China (No. 2009CB 918904), National Natural Sciences Foundation of China (30870955, 91029727, 30900555) and Program for New Century Excellent Talents in University (NCET-08-0646).

Author information

Correspondence to Y-F Dai or X-C Bai.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • mammalian target of rapamycin
  • arachidonic acid
  • breast cancer
  • tumorigenesis
  • angiogenesis

Further reading