Up-regulation of Acetyl-CoA Carboxylase α and Fatty Acid Synthase by Human Epidermal Growth Factor Receptor 2 at the Translational Level in Breast Cancer Cells*

Expression of the HER2 oncogene is increased in ∼30% of human breast carcinomas and is closely correlated with the expression of fatty acid synthase (FASN). In the present study, we determined the mechanism by which FASN and acetyl-CoA carboxylase α (ACCα) could be induced by HER2 overexpression. SK-BR-3 and BT-474 cells, breast cancer cells that overexpress HER2, expressed higher levels of FASN and ACCα compared with MCF-7 and MDA-MB-231 breast cancer cells in which HER2 expression is low. The induction of FASN and ACCα in BT474 cells were not mediated by the activation of SREBP-1. Exogenous HER2 expression in MDA-MB-231 cells induced the expression of FASN and ACCα, and the HER2-mediated increase in ACCα and FASN was inhibited by both LY294002, a phosphatidylinositol 3-kinase inhibitor, and rapamycin, a mammalian target of rapamycin (mTOR) inhibitor. In addition, the activation of mTOR by the overexpression of RHEB in MDA-MB-231 cells increased the synthetic rates of both FASN and ACCα. On the other hand, FASN and ACCα were reduced in BT-474 cells by a blockade of the mTOR signaling pathway. These changes observed in their protein levels were not accompanied by changes in their mRNA levels. The 5′- and 3′-untranslated regions of both FASN and ACCα mRNAs were involved in selective translational induction that was mediated by mTOR signal transduction. These results strongly suggest that the major mechanism of HER2-mediated induction of FASN and ACCα in the breast cancer cells used in this study is translational regulation primarily through the mTOR signaling pathway.

a number of studies have demonstrated abnormally high levels of FASN in many human epithelial cancers and preneoplastic lesions (2,3). FASN, a lipogenic enzyme, catalyzes the biosynthesis of palmitic acid that is used for the synthesis of triacylglycerol as a storage fuel molecule as well as membrane lipids including phospholipids and sphingolipids (4). In breast cancer cells, the expression of FASN is closely related to the aggressiveness of cancers as well as to the development, maintenance, and cell cycle progression of human cancers (5)(6)(7). Breast cancer cells that overexpress FASN undergo apoptosis when treated with small interfering RNAs (siRNAs) against FASN or FASN inhibitors, such as C75 and cerulenin (8 -11). Under physiological conditions, the activities of lipogenic enzymes, including FASN, are tightly regulated by nutritional and hormonal parameters at the transcription level. Sterol regulatory element-binding proteins (SREBP-1a, SREBP-1c, and SREBP-2) are the major transcription factors that mediate this regulation (12). SREBPs reside in endoplasmic reticulum membranes as inactive precursors. To become active, the NH 2 -terminal segments of SREBPs are released from the endoplasmic reticulum by proteolytic cleavage and enter the nucleus where they activate their target genes. SREBP-1 preferentially activates the genes involved in fatty acid synthesis, whereas SREBP-2 preferentially activates the genes involved in cholesterol synthesis. Activation of SREBPs has been suggested as one of the major mechanisms by which the induction of FASN in breast cancer cells occurs (13). SREBP-1-mediated induction of the expression of FASN was shown in MCF10a epithelial cells transformed by H-ras; SREBP-1 was activated by mitogen-activated protein kinase and phosphatidylinositol 3-kinase (PI3K)/Akt pathways (14,15) in these cells. Progestin also can induce FASN expression in T47D breast cancer cells through SREBP-1 activation (16).
For the overexpression of FASN in breast cancer cell lines, human epithelial growth factor receptor-2 (HER2) was reported to play an important role (8,17). HER2 belongs to the type I family of growth factor receptors in which four structurally related proteins are included: HER1 (ErbB1), HER2 (ErbB2), HER3 (ErbB3), and HER4 (ErbB4). HER2 forms a heterodimer with other HER family members to activate its intrinsic protein-tyrosine kinase activity, triggering a cascade of signal transduction events that result in cellular proliferation (18 -20). It was reported that HER2 gene amplification and/or protein overexpression occurs in 14 -30% of all breast cancer cases (21,22), and its overexpression is associated with rapid tumor growth, poor response to conventional Taxol chemotherapy, and poor prognosis (23,24). Accordingly, Herceptin, a monoclonal antibody that blocks signals transduced by HER2, has been shown to augment the effects of chemotherapeutic agents in metastatic breast cancers (25).
One of the suggested mechanisms that correlates FASN overexpression with HER2 is that aberrant PI3K transduction cascades by HER2 drive the constitutive activation of SREBP-1c, resulting in transcriptional induction of the FASN gene (13,14). A number of growth factors, hormones, and cytokines activate PI3K, which in turn activates Akt. Activated Akt phosphorylates tuberous sclerosis 2 (TSC2), then represses the GTPaseactivating protein activity of TSC1/TSC2 on RHEB, the major regulator of the mammalian target of rapamycin (mTOR). The mTOR protein, one of the downstream effectors of PI3K, is a Ser/Thr kinase that controls cell growth and proliferation by the regulation of protein synthesis (26,27). In the absence of either TSC1 or TSC2, RHEB-GTP levels are high, which leads to the constitutive activation of the mTOR-raptor signaling pathway. The activation of mTOR leads to the phosphorylation and inactivation of eIF4E-binding protein, resulting in the derepression of translation. Another target of mTOR is the 70-kDa subunit of S6 kinase 1 that controls protein synthesis in response to various hormones, mitogens, and nutrients (28 -30). It was reported that mTOR, which is activated by HER2 overexpression, plays a role in the proliferation and survival of breast cancer cells both in vitro and in vivo (31). Several studies have demonstrated that mTOR inhibition sensitizes certain resistant cancer cells to endocrine therapy with tamoxifen (32,33) and certain chemotherapeutic drugs, such as paclitaxel, and carboplatin, thus suggesting that the mTOR signaling pathway is a candidate molecular target for breast cancer treatment (34).
Although lipogenesis has proven to be an important process for the survival and proliferation of breast cancer cells, little is known about the mechanisms by which HER2 mediates the regulation of lipogenic enzymes. In the present study, we show that the expression of two lipogenic enzymes, FASN and ACC␣, is increased by HER2 overexpression, and that the activation of mTOR by HER2 plays an important role in their overexpression through the selective translational activation of their mRNAs. It is suggested that the 5Ј-and 3Ј-untranslated regions (UTRs) of both FASN and ACC␣ mRNAs contribute to the increase in the translational efficiency of their mRNAs by mTOR.

EXPERIMENTAL PROCEDURES
Cell Culture-Human breast cancer cell lines, MDA-MB-231, MCF-7, BT-474, and SK-BR-3, were obtained from the American Type Culture Collection (ATCC, Manassas, VA). The cells were maintained in the respective media (Invitrogen) as recommended by the ATCC at 37°C in a humidified atmosphere of 5% CO 2 in air.
Preparation of Recombinant Adenovirus and Viral Transfection-Recombinant adenoviruses expressing HER2 (AdHER2) and RHEB (AdRHEB) were constructed using the AdEasy adenoviral vector system vector (Stratagene, La Jolla, CA) according to the manufacturer's protocol. Briefly, cDNAs encoding HER2 and RHEB were amplified by PCR and cloned into the pShuttle-CMV plasmid. The recombinant shuttle vector constructs and pAdEasy-1 vector were co-transformed into BJ5183 Escherichia coli by electroporation. The positive recombinant adenovirus plasmids were identified by restriction digestion. Five micrograms of the recombinant adenovirus plasmids were linearized by digestion with PacI, then transfected into HEK-293A cells using Lipofectamine and the PLUS reagent (Invitrogen). The adenoviruses were amplified in 293A cells, and infective viruses were isolated using CsCl 2 ultracentrifugation. The infective recombinant adenoviruses were desalted and the viral titers were determined by plaque-forming assay using an agarose overlay in HEK-293A cells. MDA-MB-231 and BT-474 cells were transduced with recombinant adenovirus at a multiplicity of infection of 100 in serum-free media at 37°C. Culture medium was replaced with fresh medium containing 10% fetal bovine serum and antibiotics 2 h after transduction. After an overnight incubation, cells were treated with LY294002 (Alexis, San Diego, CA), Herceptin (a generous gift from Genentech, South San Francisco, CA), or rapamycin (Cell Signaling Technology, Danvers, MA) for 24 h. The cells were harvested at 48 h after viral transduction and processed for either immunoblot analysis or RNA isolation. A replicon-defective adenoviral vector expressing green fluorescence protein (AdGFP) was used as a control.
Immunoblot Analyses-The cells treated as indicated in the figure legends were harvested and lysed in 2ϫ SDS loading buffer, then briefly sonicated. Lysates were cleared by centrifugation at 12,000 ϫ g for 10 min at 4°C. Supernatants were collected and protein concentrations were determined by Bradford assay (Bio-Rad). The proteins were subjected to electrophoresis on SDS-polyacrylamide gels and then transferred to Protran nitrocellulose membranes (Schleicher & Schuell). The membranes were blocked in phosphate-buffered saline containing 0.05% (v/v) Tween 20 (PBST) and 5% (w/v) nonfat dried milk, and probed with primary antibodies as indicated in the figure legends. The following antibodies were used: polyclonal antibodies against rat ACC␣ from Genepia (Seoul, Korea); polyclonal antibodies against S6 protein, phospho-S6 protein, mTOR, phospho-mTOR, and RHEB from Cell Signaling Technology; polyclonal antibodies against HER2 from Oncogene (San Diego, CA); and monoclonal antibodies against ␣-tubulin from Calbiochem. Polyclonal antibodies against FASN and SREBP-1 were described previously (35,36). Antibodies against human SREBP-2, and bovine low density lipoprotein receptor were kindly provided by Jay Horton (University of Texas Southwestern Medical Center). Immunoreactive bands were visualized by horseradish peroxidase-conjugated secondary antibodies (Pierce) using the SuperSignal West Pico Chemiluminescent Substrate System (Pierce).
siRNA Experiment-siRNA oligonucleotides for mTOR were purchased from Ambion (Austin, TX). A scrambled siRNA (Dharmacon, Chicago, IL) was used as a control. Transfection was performed using FuGENE 6 (Roche Applied Sciences) according to the manufacturer's protocol. Briefly, 100 pmol of siRNA and 6 l of FuGENE 6 were mixed in 100 l of serumfree media. After 30 min of incubation, the siRNA-FuGENE 6 mixture was added to the cultured cells. Forty-eight hours after transfection, lysates were prepared and immunoblot analyses were performed.
Cell Viability Assay-Cells (3 ϫ 10 3 cells/well) were placed in a 96-well culture plate on day 0. On day 1, the medium was replaced with fresh medium containing 10 g/ml cerulenin (Sigma) solubilized in dimethyl sulfoxide. On day 2, cell viability was determined using the CellTiter 96 Aqueous Non-radioactive Cell Proliferation Assay kit (Promega) according to the manufacturer's protocol.
Measurement of RHEB-induced Translational Activation of FASN and ACC␣-MDA-MB-231 cells (8 ϫ 10 5 cells/10-cm dish) were plated on day 0. On day 1, the cells were transduced with either AdGFP or AdRHEB at a multiplicity of infection of 100 as described above. Twenty-four hours after transduction, the medium was replaced with methionine/cysteine-free Dulbecco's modified Eagle's medium (Invitrogen) for 1 h. Rates of protein synthesis were measured by a pulse labeling experiment; [ 35 S]Met-Cys (PerkinElmer Life Sciences) was added to the medium at a final concentration of 0.3 mCi/ml. Cells were incubated for 0, 1, 2, and 4 h. After incubation, the cells were washed with ice-cold phosphate-buffered saline and harvested with 500 l of lysis buffer (100 mM Tris-HCl, pH 7.5, 300 mM NaCl, 2% (v/v) Nonidet P-40) on ice. Protein degradation was measured by pulse-chase assay; the cells were rinsed after a 4-h pulse labeling, incubated in medium containing 2 mM L-methionine and 20 g/ml cycloheximide for 0, 1, 2, and 4 h, and then harvested in lysis buffer.
The cell lysates were collected by centrifugation at 12,000 ϫ g for 10 min at 4°C and precleared by mixing with 50 l of protein A-agarose (50% slurry), followed by incubation for 2 h on a rotating device at 4°C. After removal of protein A-agarose by centrifugation, the indicated antibodies were added to the cleared cell lysates and incubated overnight on a rotating device at 4°C. After the addition of 50 l of protein A-agarose, the mixture was incubated for 2 h. The protein A-agarose beads were collected by centrifugation, washed twice with 50 mM Tris-HCl, pH 7.5, 0.5 M NaCl, 0.1% (v/v) Nonidet P-40, and 0.05% (w/v) deoxycholate, and washed once with 10 mM Tris-HCl, pH 7.5, 0.1% (v/v) Nonidet P-40, and 0.05% (w/v) deoxycholate. The immunoprecipitated proteins were eluted from the protein A-agarose by boiling for 5 min with 1ϫ SDS loading buffer, and subjected to electrophoresis on a 6% SDSpolyacrylamide gel. After electrophoresis, the gels were soaked in 30% methanol and 10% acetic acid for 10 min, dried, and exposed to x-ray film.
Generation of Promoter-Reporter Constructs and Transient Transfection Assay-To evaluate the regulatory roles of the UTRs of ACC␣, FASN, and ACLY mRNAs on gene expression, two ACC␣ 5Ј-UTRs, designated Ex1 (599 bp) and Ex2 (161 bp), the 5Ј-UTR of FASN (100 bp), and the 5Ј-UTR of ACLY (101 bp) were amplified from the cDNA prepared using total RNA isolated from BT-474 cells. The amplified PCR fragments were inserted into the HindIII/NcoI sites of the pGL3 promoter vector (Promega), resulting in constructs harboring the amplified 5Ј-UTR fragments immediately upstream of the open reading frame of the firefly luciferase gene in the pGL3 promoter. The 3Ј-UTR sequences of ACC␣, FASN, and ACLY were obtained by digesting human ACC␣ cDNA image clone 3874305 with XhoI/BamHI, human FASN cDNA image clone 6172538 with NcoI/XhoI, and the human ACLY cDNA image clone 2959339 with NdeI/XhoI, respectively. The digested fragments with sizes of 2.4, 1, and 1 kb for ACC␣, FASN, and ACLY, respectively, were subcloned into the XbaI site of the pGL3 promoter vector, resulting in constructs harboring the 3Ј-UTR fragments downstream of the firefly luciferase gene. To construct the pGL3 promoter containing both the 5Ј-and 3Ј-UTRs, the amplified 5Ј-UTR fragments were inserted upstream of the firefly luciferase gene in the pGL3 promoter constructs containing the 3Ј-UTRs of ACC␣, FASN, and ACLY downstream of the luciferase gene. After transient transfection of these constructs into MDA-MB-231 cells, both the luciferase assay and PCR were performed to determine the role of the UTR sequences in gene expression under conditions of RHEB overexpression. The cells (5 ϫ 10 5 cells/well) were plated in 6-well plates on day 0. On day 1, the cells were transfected with the indicated plasmids using Lipofectamine and the PLUS reagent (Invitrogen) according to the manufacturer's protocol with a minor modifi-cation for the simultaneous transduction with AdRHEB. Plasmid DNA (0.6 g) was mixed with Lipofectamine and the PLUS reagent in serum-free media. While the complexes were forming, the cells were washed with phosphate-buffered saline, and then incubated with 1 ml of serum-free media containing a multiplicity of infection of 100 of either AdGFP or AdRHEB. After 15 min, the Lipofectamine-DNA complexes were added to the medium. After a 3-h transfection, the medium was replaced with fresh medium containing 10% fetal bovine serum. For BT-474 cells, the same transfection procedures were performed except for transduction with adenoviruses. On day 2, the cells were treated with rapamycin for 24 h. On day 3, the cells were harvested and lysed with 200 l of reporter lysis buffer (Promega). The luciferase activity assay was performed using 5 l of the cell extract and 45 l of luciferase assay reagent (Promega). Each transfection experiment was performed in triplicate, and mean Ϯ S.E. of three independent experiments were calculated. For quantitative analysis of the luciferase mRNA in the transfected cells, poly(A)-rich RNAs were isolated using the PolyATtract mRNA isolation system IV (Promega) from total RNA, and treated with RNase-free DNase (Ambion) to remove the transfected DNA. PCR for the luciferase mRNA was performed using the luciferase primer set: 5Ј primer, 5Ј-AGCGACCAACGCCTTGATT-3Ј, and 3Ј primer, 5Ј-ACTTCAGGCGGTCAACGATG-3Ј. To verify the complete removal of transfected DNA, amplification of the SV40 promoter region of the reporter constructs was performed as a negative control with the following primers: 5Ј primer, 5Ј-CTAGCAAAATAGGCTGTCCCC-3Ј, and 3Ј primer, 5Ј-CTTTATGTTTTTGGCGTCTTC C-3Ј.

FASN Expression Is Closely Linked to HER2 Expression in
Breast Cancer Cell Lines-Increased expression of FASN has previously been demonstrated in breast and ovarian cancer cells (5,6,37) and was recently reported to be related to HER2 overexpression (8,11,38). FASN and HER2 expression levels were evaluated by immunoblot analysis in four human breast cancer cell lines. HER2 was not detected in MDA-MB-231 and MCF-7 cells, whereas overexpression of HER2 was shown in BT-474 and SK-BR-3 cells (Fig. 1A). Consistent with previous reports, cells that overexpress HER2 showed higher expression levels of FASN (Fig. 1A, lanes 3 and 4) than cells lacking detectable levels of HER2. To determine whether expression of FASN is important for cell growth and survival in these cells, cells were treated with cerulenin (10 g/ml) for 24 h and viable cells were assayed. Cerulenin, a natural product derived from Cephalosporium caerulens, is a suicide inhibitor that binds irreversibly to the catalytic site of the ␤-ketoacyl synthase domain of multifunctional FASN (39,40). BT-474 and SK-BR-3 cells that overexpress FASN and HER2 showed higher susceptibility to cerulenin than did MDA-MB-231 or MCF-7 cells, resulting in lower cell viability (Fig. 1B). These results suggest a strong association between FASN expression and HER2; the increased expression of FASN might be an important factor for cell growth and survival in breast cancer cells that overexpress HER2.

SREBP-1 Is Not the Major Regulator for the Expression of Lipogenic Enzymes in Breast Cancer
Cells-To study the role of SREBP-1 on HER2-related increases in FASN, we compared the relative amounts of FASN, ACC␣, ACLY, and SREBP-1 in BT-474 and MDA-MB-231 cells ( Fig. 2A). Whereas certain studies have reported that SREBP-1c is the mediator that  induces the transcription of FASN in cancer cells (3,41), the amounts of the precursor form of SREBP-1 were very similar in both cell lines. The amount of the nuclear form of SREBP-1, which has transcriptional activity, was significantly lower in BT-474 compared with MDA-MB-231 cells. However, the expression levels of ACC␣ and FASN were significantly increased in BT-474 cells compared with MDA-MB-231 cells. ACLY, which is one of lipogenic enzymes that is regulated by SREBP-1, shows no significant differences between these two cell lines. This result suggests that the overexpression of FASN and ACC␣, shown in BT-474 cells, is not mediated by the activation of SREBP-1.
To further study whether the expression of FASN, ACC␣, and ACLY is affected by the sterol-dependent control of SREBP processing, immunoblot analyses were performed in cells grown in either the presence or absence of cholesterol/25-hydroxycholesterol supplemented with 10% delipidated serum (Fig. 2B). Deprivation of sterol significantly increased the levels of the nuclear forms of SREBP-1 and SREBP-2 in both cell lines (compare Fig. 2B, lanes 1 and 3, with A, lanes 1 and 2), whereas the addition of sterol almost completely blocked the processing of SREBP-1 and SREBP-2, resulting in the depletion of the nuclear forms of SREBPs (Fig. 2B, lanes 2 and 4). As expected, sterols significantly down-regulated the amount of low density lipoprotein receptor, one of the target genes of SREBP-2 in both cell lines, as the nuclear form of SREBP-2 disappeared (Fig. 2B,  lanes 2 and 4). However, the expression of FASN, ACC␣, and ACLY was not significantly changed by the presence of sterols. These results strongly suggest that neither SREBP-1 nor -2 is the major effector that increases the expression of FASN and ACC␣ in BT-474 cells that overexpress HER2.
The PI3K/Akt/mTOR Signaling Pathway Is Involved in the Induction of FASN and ACC␣ by HER2-To examine whether HER2 overexpression changes the rate of transcription and/or translation of FASN, ACC␣, and ACLY, their protein and mRNA levels were determined in MDA-MB-231 cells that were transduced with adenoviruses expressing HER2 (AdHER2). Overexpression of HER2 markedly increased the protein levels of ACC␣ and FASN, whereas that of ACLY remained unchanged (Fig. 3A, lanes 1 and 2). However, the mRNA levels for FASN, ACC␣, and ACLY were not significantly changed by HER2 overexpression (Fig. 3B). When HER2 signaling was blocked by Herceptin, a humanized monoclonal antibody against HER2, the HER2-mediated induction of FASN and ACC␣ was completely abolished (Fig. 3A, lane 4) as expected. These results suggest that the induction of FASN and ACC␣ by HER2 overexpression take place at a post-transcriptional level.
To determine whether HER2-mediated induction of ACC␣ and FASN proteins is mediated through the PI3K/Akt pathway, MDA-MB-231 cells infected with AdHER2 were treated with LY294002, a PI3K inhibitor. The induction of FASN and ACC␣ transduced by AdHER2 was completely blocked by LY294002 treatment (Fig. 3A, lane 3), implying that PI3K signaling is involved in the HER2-mediated induction of FASN and ACC␣.
The PKI/Akt signals that are triggered by HER2 are able to activate the mTOR signaling pathway, which can regulate the translation of several genes (42)(43)(44). To determine whether the induction of FASN and ACC␣ by HER2 results from the acti-vation of the PI3K/Akt/mTOR pathway, cells were treated with rapamycin. The induction of ACC␣ and FASN by HER2 was blocked by treatment with rapamycin as well as Herceptin (Fig.  4A). Treatment of BT-474 cells with LY294002, rapamycin, or Herceptin also resulted in the decreased expression of FASN and ACC␣ without a concomitant change in the levels of ACLY (Fig. 4B), suggesting that PI3K/Akt and mTOR mediate increased FASN and ACC␣ levels by endogenous HER2 overexpression. Phosphorylation of the S6 protein, one of the targets of mTOR (Fig. 4B), was blocked by these treatments as well. However, the mRNA levels of FASN and ACC␣ were not altered by either Ly294002 or rapamycin (Fig. 4C). Silencing of mTOR by siRNA also led to a marked suppression of FASN and ACC␣, in contrast to a slight reduction in ACLY expression (Fig. 4D).
To determine whether the expression of FASN and ACC␣ can be induced by mTOR activation in the absence of HER2/ PI3K/Akt signals, changes in both FASN and ACC␣ were analyzed in MDA-MB-231 cells that were transduced with adenovirus that overexpressed RHEB. RHEB, a small GTP-binding protein, is directly upstream of mTOR and its induced expression activates mTOR (45,46). Overexpression of RHEB in MDA-MB-231 cells increased total mTOR levels as well as the phosphorylated form of mTOR (Fig. 5A). RHEB-mediated mTOR activation is accompanied by increases in the levels of both FASN and ACC␣, whereas no significant changes in ACLY expression were observed (Fig. 5A). The induction of FASN and ACC␣ is thought to be regulated at a post-transcriptional level because their mRNA levels were unchanged (Fig. 5B).
To determine whether the mTOR-mediated increase in FASN and ACC␣ is due to an increase in translation and/or a decrease in the degradation of FASN and ACC␣ proteins, metabolic labeling assays were performed in MDA-MB-231 cells transduced with AdRHEB. The overexpression of RHEB significantly increased the rate of protein synthesis of FASN and ACC␣, but no significant differences in the levels of ACLY synthesized were found between the cells expressing either GFP or RHEB (Fig. 6A). The degradation of ACC␣ and ACLY were not significantly changed by RHEB overexpression, and the degradation of FASN was increased 4 h after radioisotope labeling (Fig. 6B). These results suggest that translational activation via mTOR is the primary mechanism by which the expression of FASN and ACC␣ is increased in breast cancer cells that overexpress HER2.
The 5Ј-and 3Ј-UTRs of ACC␣ and FASN mRNAs Are Involved in Their Translational Activation by mTOR-To elucidate the mechanism by which FASN and ACC␣ are activated by mTOR signaling, we hypothesized that either the 5Ј-or 3Ј-UTRs of ACC␣ and FASN mRNAs might play putative roles in the regulation of their expression. The 5Јand/or 3Ј-UTR sequences were inserted either upstream or downstream of the luciferase gene in a SV40 promoterdriven reporter construct (pGL3 promoter) as shown in Fig.   7A. Insertion of the 5Ј-UTRs of either FASN (100 bp) or ACLY (101 bp) markedly increased luciferase activity (Fig. 7, B and D,  white bars). The transcription of ACC␣ can be driven by two promoters, P-I and P-II, which generate two different transcripts containing either exon 1 or exon 2 sequences (47,48). Because most breast cancer cell lines generate both transcripts (data not shown), both 5Ј-UTRs, designated Ex1 and Ex2 in the present study, were tested. The constructs for Ex1 and Ex2 contained exon 1 and 2 sequences, respectively, followed by exon 4 and a part of exon 5 of ACC␣ mRNA. Exon 1 in Ex1 is relatively long (ϳ550 bp) and has a high GC content and multiple ATG codons, whereas exon 2 in Ex2 is short (ϳ120 bp) and lacks ATG codons. The 5Ј-UTR in Ex1 strongly inhibited luciferase activity, whereas the 5Ј-UTR in Ex2 showed an approximate 6.4-fold stimulation of luciferase activity compared with pGL3p (Fig. 7C, white bars). Overexpression of RHEB further increased the luciferase activities of the constructs with the 5Ј-UTR of FASN and Ex2 of ACC␣ by 2.3-and 1.7-fold, respectively. However, the 5Ј-UTR of ACLY or Ex1 of ACC␣ did not respond to RHEB overexpression (Fig. 7, B-D, black bars). The 3Ј-UTRs of FASN and ACC␣ slightly reduced basal luciferase activities (Fig. 7, B-D, white bars) compared with the luciferase activity of pGL3p; however, they significantly increased luciferase activities by 1.7-and 2.0-fold, respectively, in response to RHEB overexpression (Fig. 7, B and C,  black bars). The 3Ј-UTR of ACLY also reduced basal luciferase activity slightly; however, RHEB overexpression did not affect luciferase activity (Fig. 7D). The combination of the 5Јand 3Ј-UTRs of FASN attenuated the increase in the basal luciferase activity driven by the 5Ј-UTR; however, RHEB overexpression increased luciferase activity 3.7-fold (Fig. 7B,  5Ј and 3Ј columns). The presence of Ex2 and the 3Ј-UTR of ACC␣ showed additive increases in luciferase activity by 3.9fold in response to RHEB overexpression (Fig. 7C, Ex2 and  -3Ј columns, black bar). The mRNA levels of the luciferase gene remained unchanged between the constructs containing different UTRs or when RHEB was overexpressed (Fig.  7E). The possibility of amplification from transfected plasmid DNA during PCR was excluded by confirming that the The cell lysates were prepared and immunoblot analyses were carried out with antibodies against ACLY, ACC␣, FASN, HER2, S6 protein, phospho-S6 protein, or tubulin. C, total RNA was prepared from BT-474 cells treated with 25 M LY294002 or 10 nM rapamycin and RT-PCR was carried out as described under "Experimental Procedures." The numbers below the figure denote the relative amounts of mRNAs calculated using the comparative C t method. GAPDH was used as an invariant control. D, BT-474 cells were transfected with 100 pmol of scrambled siRNA or mTOR siRNA as described under "Experimental Procedures." Forty-eight hours after transfection, cell lysates were prepared and immunoblot analyses were carried out with antibodies against mTOR, ACC␣, FASN, ACLY, or tubulin. Similar results were obtained from two independent experiments. SV40 promoter sequence was not amplified. These results suggest that the UTRs of FASN and ACC␣ play important roles in the regulation of gene expression at a translational level, and that this translational activation renders cells responsive to RHEB overexpression.
To further study whether the UTRs of FASN and ACC␣ could induce gene expression in breast cancer cells that overexpress HER2, transient transfection in BT474 cells was performed with the reporter constructs bearing both the 5Ј-and 3Ј-UTR sequences of FASN, ACC␣, and ACLY. The luciferase activities of pGL3p and the construct containing the UTRs of ACLY were not significantly affected by rapamycin. However, the increase in the luciferase activities driven by the UTRs of FASN and ACC␣ Ex2 were significantly repressed by rapamycin (Fig. 8). These results suggest that the UTRs of FASN and ACC␣ mRNA contribute to the mTOR-dependent translational activation of both ACC␣ and FASN mRNAs.

DISCUSSION
Mammalian FASN is a multifunctional enzyme that catalyzes de novo biosynthesis of palmitate (49). The substrates for this anabolic enzyme, such as acetyl-CoA and malonyl-CoA, are supplied by the serial reactions catalyzed by ACLY and ACC␣. Many studies have shown that the levels of these lipogenic enzymes are dramatically but tightly regulated in liver and adipose tissues at the transcriptional level, and that SREBP-1 is reported to be indispensable for the nutritional regulation of these lipogenic enzymes (12). Although the transduction pathways that regulate the expression of FASN in normal and cancer cells appear to share identical downstream elements, the  upstream mechanisms that control the expression of FASN in cancer cells, however, must be different from those in normal tissues because cancer-associated FASN expression appears to be insensitive to nutritional signals (3). In prostate cancer cells, the expression of FASN was increased by both androgens and growth factors, and the mechanism for this induction was demonstrated to be the transcriptional regulation mediated by SREBP-1 activation (50,51). In breast cancer cells, the overexpression of FASN is associated with HER2 amplification and/or overexpression (8,17), which is considered to be a poor prognostic marker. Kumar-Sinha et al. (17) reported that HER2 overexpression increased the transcription of FASN via the PI3K signal pathway. Taken together, transcriptional control has been hypothesized to be the primary mechanism by which FASN is induced in breast cancer cells that overexpress HER2 and SREBP-1 is thought to be a key regulator in this process (3). In the present study, we demonstrated that the HER2-associated induction of FASN and ACC␣ was achieved through the translational activation mediated by mTOR via the PI3K/ Akt pathway in breast cancer cells and not by the transcriptional activation through SREBPs. This translational activation was confined to FASN and ACC␣ because the other lipogenic enzymes, such as ACLY, malic enzyme, or glucose-6-phosphate dehydrogenase, showed no association with HER2 overexpression (data not shown).
The activation of the Akt/mTOR pathway has been found in breast cancers with high expression levels of HER2, suggesting possible roles of the Akt/mTOR pathway in HER2-mediated tumor progression (31). PKB/Akt was previously demonstrated to protect cancer cells from tamoxifen-induced apoptosis (52). Recently, the activation of mTOR, a downstream effector of the PKB/Akt signal, was reported to play a critical role in the development of hormone-refractory phenotypes in certain breast cancer cells. Accordingly, the inhibition of mTOR by rapamycin resulted in the restoration of sensitivity to tamoxifen (32). The induction of FASN and ACC␣ by the mTOR signaling pathway shown in the present study suggests that FASN and ACC␣ are the important target molecules in the HER2/PI3K signaling pathway in breast cancer. It was also reported that the activation of mTOR increased overall translation rates by inactivating eIF4E-binding protein, which in turn activated eIF-4E, In C, Ex1 and Ex2 represent the two 5Ј-UTRs of ACC␣, which contain either exon 1 or exon 2 sequences, respectively. Aliquots of the cell lysates were analyzed 48 h after transfection for luciferase activity. Each result represents the mean Ϯ S.D. of triplicate incubations. The numbers above the black bars express the -fold increase induced by RHEB overexpression. E, the mRNA levels of luciferase transcripts were measured by RT-PCR using poly(A)-rich RNA isolated from the cells transfected with 5Ј and 3Ј-UTR constructs as indicated in the figure. GAPDH cDNA was also amplified as a control to demonstrate that equal amounts of total RNA were used for RT-PCR. To show that the luciferase cDNA amplified by RT-PCR was not derived from the transfected plasmid, the primers for the SV40 promoter region were used for PCR. and by activating the 70-kDa subunit of S6 kinase 1, resulting in increasing the synthesis of ribosomal proteins. In the present study, we demonstrated that the activation of mTOR induced by HER2-overexpression led to a selective increase in the translation of FASN and ACC␣ and not an overall increase in the rate of protein synthesis.
The selective translational controls for CXCR4 by mTOR were previously reported in breast cancer cells (53), which are implicated in tumor metastasis (54). Src and c-myc could be increased by translational controls via the activated Akt/mTOR pathway (44). The 5Ј-UTR derived from the major promoter of the c-myc gene, P2, mediated translational activation by HER2 signal (42). However, the precise mechanisms by which these selective translational controls are achieved by the mTOR signal require elucidation (42,55). We, in this study, have found that the 5Ј-UTR of FASN significantly activated gene expression at the translational level, and mTOR activation by RHEB overexpression further augmented this effect. The ACC␣ gene has multiple promoters, including P-I, P-II, and P-III (47,48). The P-I promoter is unique to humans and produces relatively long 5Ј-UTRs with multiple ATG codons in its transcript (48). The P-I promoter was shown to be constitutively active in breast cancer cell lines, including MDA-MB-231 and BT-474 (data not shown). The P-II promoter is reported to be regulated by various hormones, including insulin and thyroid hormone, and is highly conserved between species (42). In this study, it was shown that the 5Ј-UTR (Ex2) derived from ACC␣ P-II strongly activated reporter expression, whereas the 5Ј-UTR (Ex1) derived from the P-I promoter showed significant suppression of reporter expression as well as non-responsiveness to RHEB overexpression. The 5Ј-UTR derived from P-III was not evaluated for its transcriptional activation in this study because the anti-ACC␣ antibody used in this study was raised against 74 amino acids at the NH 2 terminus of ACC␣, which is not encoded by transcripts derived from the P-III promoter. The 5Ј-UTRs derived from P-II also increased reporter expression when mTOR was activated by RHEB overexpression. All 3Ј-UTRs of FASN, ACC␣, and ACLY decreased luciferase activity to a slight degree. However, the 3Ј-UTRs of FASN and ACC␣ may play roles in mTOR-mediated translational activation, but not the 3Ј-UTR of ACLY. These UTR-mediated changes in luciferase activity were shown to result from translational control rather than transcriptional control because the mRNA levels of luciferase were almost identical between the constructs harboring different UTRs.
At present, most of the studies on the regulation of lipogenic enzymes by hormones in lipogenic organs have been focused primarily on the mechanism at the level of transcription. In this study, we demonstrated that ACC␣ and FASN could be selectively up-regulated at the translational level by the HER2/PI3K/ Akt pathway via mTOR activation in breast cancer cells, and that the 5Ј-and 3Ј-UTRs of their transcripts play important roles in this regulation. These results suggest that the expression of ACC␣ and FASN might be regulated at translational levels as well as at transcriptional or post-translational levels in normal lipogenic organs. It also suggests that aberrant translational regulation might contribute to pathological conditions, that is, the constitutive overexpression of lipogenic enzymes as in insulin-resistant diabetes.