ATP Citrate Lyase: Activation and Therapeutic Implications in Non–Small Cell Lung Cancer

One of the most common properties of cancer cells is a metabolic change into high glycolysis and lipogenesis to satisfy the increased demand for energy and macromolecules need for autonomous growth. It is assumed that phosphatidyl-inositol-3 kinase (PI3K)-Akt signaling pathway plays a part in this phenomenon, and constitutive activation of this pathway is frequently observed in many types of cancers. Proto-oncogene Akt, downstream from PI3K, exhibits oncogenic effects with influence on proliferation, survival, motility, and morphology. Akt also plays a pivotal role of cancer metabolism via the stimulation of nutrient uptake, glycolysis, and fatty acid biosynthesis through action on glucose transporters, glycolytic enzymes, and lipogenic enzymes (1–3).

Accumulating evidence indicates that dysregulated glycolysis and lipid synthesis contribute to cancer pathogenesis. Fluorodeoxyglucose- or ¹¹C-acetate–positron emission tomography imaging based on the metabolic difference between normal and neoplastic tissue is useful in cancer detection (4, 5). Thus, glucose and lipid metabolism–targeting therapeutics have attracted great attention (6–8). Inhibitors of hexokinase have been shown to reduce ATP concentrations leading to cytotoxicity in tumor cells (9). Also, inhibitors of lipid synthesis such as statins, farnesyl transferase inhibitors, and fatty acid synthase (FASN) inhibitors impair proliferation or induce apoptosis of tumor cells (7, 10).

ATP citrate lyase (ACLY) is a key enzyme of de novo fatty acid synthesis. ACLY is involved in the generation of cytosolic acetyl-CoA and oxaloacetate from citrate and, thus, contributes to the translocation of acetyl-CoA from mitochondria to cytosol. It has been found that Akt directly phosphorylates and activates ACLY (11). More recently, it has been reported that chemical inhibitors or ACLY RNAi suppress tumor cell growth (12, 13). Because ACLY functions depend on glucose utility and its product acetyl-CoA is a building block for cholesterol and fatty acid synthesis, ACLY must play a crucial role in cancer cell metabolism.

Lung cancer is the leading cause of cancer death in the worldwide and is a typical tumor displaying a high rate of glycolysis and fatty acid synthesis. Thus, targeting dysregulated metabolism may be an attractive strategy for lung cancer treatment.

In this study, we therefore examined the contribution of ACLY to lung cancer pathogenesis and its therapeutic potential. Overexpression and activation of ACLY in patients with lung adenocarcinoma could be shown, and it was found to be a statistically significant negative prognostic factor. Selective ACLY inhibition resulted in the inhibition of tumor cell growth in vitro and in vivo. Growth arrest induced by ACLY inhibition was associated with intracellular lipid accumulation, despite of the inhibition of de novo lipogenesis. Our findings suggest that ACLY plays a critical role in cancer progression and metabolic changes, and may be a promising therapeutic target in lung cancer.

Cell culture and clinical samples. The human lung adenocarcinoma cell lines A549, NCI-H460, NCI-H23, NCI-H522, HOP62, COR-L105, and PC14 were obtained from the American Type Culture Collection and grown in RPMI 1640 supplemented with 10% fetal bovine serum (both medium and serum from Life Technologies Bethesda Research Laboratories) and 1% penicillin/streptomycin in an atmosphere of 5% CO₂ at 37°C.

We also analyzed lung cancer tissue from operable patients that had developed lung adenocarcinoma. All tumors were pathologically diagnosed based on the WHO classification of lung tumors (14) and staged according to the classification of the Union Internationale Contre le Cancer. Data for clinicopathologic variables and patient prognosis were also obtained. All experiments were performed using a protocol approved by the Institutional Review Board of the Japanese Foundation for Cancer Research (#2007-1058).

Transient transfections. Constitutively active myristolated Akt and the dominant-negative Akt plasmids were generous gifts from Dr. Yukiko Gotoh (Institute of Molecular and Cellular Biosciences, University of Tokyo, Japan). Cells were plated at 7 × 10⁵ per well in 60-mm dishes and transfected in triplicate using the FuGENE 6 Transfection Reagent according to the manufacturers protocol (Roche Diagnostics, Inc.). Experiments were repeated at least thrice.

RNA preparation and reverse transcription-PCR. The cell lines in logarithmic phase of growth and the frozen tissue samples were collected for RNA extraction, using an RNeasy kit (Qiagen) and total RNA was applied for first-strand cDNA synthesis with a high-capacity cDNA Reverse Transcriptase kit (Applied Biosystems). Gene-specific probes and primer were obtained from Universal ProbeLibrary (Roche Applied Science) and primer sequences were as follows: 5′-gaagggagtgaccatcatcg-3′ (ACLY forward); 5′-ttaaagcacccaggcttgat-3′ (ACLY reverse); 5′-gctcctccatcaatgacaaaa-3′ (SREBP-1 forward); 5′-tgcgcaagacagcagattta-3′ (SREBP-1 reverse). PCR was carried out in 96-well plates using the LightCycler 480 System (Roche Applied Science). All reactions were performed at least in triplicate. The relative amounts of all mRNAs were calculated using the comparative computed tomography method after normalization to human β-2-microglobulin (B2M) for human samples. Semiquantitative reverse transcription-PCR (RT-PCR) was carried out using SYBR Green for the cell lines. β-actin was used as an internal control.

Cell lysis and immunoblotting. To obtain total protein lysates, frozen tissue and cells were homogenized and dissolved in radioimmunoprecipitation assay buffer [150 mmol/L NaCl, 1.0% Igepal CA-630, 0.5% sodium deoxycholate, 0.1% SDS, 50 mmol/L Tris (pH 8.0); Sigma] containing proteinase inhibitors and phosphatase inhibitors. The protein concentration of each lysate was determined using a protein assay reagent kit (Bio-Rad). The total cell lysate was applied on 10% SDS-PAGE. After electrophoresis, the proteins were transferred electrophoretically from the gel to polyvinylidene difluoride membranes (Millipore). The membranes were then blocked for 1 h in blocking buffer (5% low-fat dried milk in TBS) and probed with the primary antibodies overnight. After being washed, the protein content was made visible with horseradish peroxidase–conjugated secondary antibodies followed by enhanced chemiluminescence (Amersham). The primary antibodies used were raised against phosphorylated ACLY (Ser 454), total ACLY, phosphorylated Akt (Ser 473), total Akt, phosphorylated Erk1/2 (Thr 202/Tyr 204), PTEN (all obtained from Cell Signaling Technology), SREBP-1 (BD Bioscience), and β-actin (Sigma). LY294002 and wortmannin were purchased from Sigma.

ACLY activity. ACLY activity was measured via the malate dehydrogenase–coupled method (15). Total protein lysates extracted from cell lines or clinical samples were subjected to the assay. Change in absorbance in the absence of exogenous ATP was subtracted from change in the presence of ATP and was normalized to protein concentration to determine the specific ACLY activity.

RNA interference. siRNA oligonucleotides for ACLY and a negative control were purchased from Invitrogen. Three independent oligonucleotides were used for ACLY siRNA. The ACLY siRNA#1 sequence was 5′-UUCUUGAUCAGCUUUCUCGUGAGGG-3′. The others were designed to correspond to a previous report (13). Transfection was performed using Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's protocol. Briefly, 60 pmol of siRNA and 10 μL of Lipofectamine RNAiMAX were mixed in 1 mL of Opti-MEM medium (10 nmol/L final siRNA concentration). After 20 min of incubation, the mixture was added to the suspended cells and these were plated on dishes for each assay. Seventy-two hours after transfection, cells were analyzed for all experiments.

Flow cytometry. For determination of the cell cycle distribution after siRNA treatment, DNA content was measured by flow cytometry with propidium iodide (PI). Briefly, 72 h after siRNA treatment on 60-mm dishes, the cells were harvested and cell pellets were then fixed with 70% ethanol. Cells were incubated with 20 μg/mL RNase A at 37°C for 30 min and then stained with 10 μg/mL PI. The cell cycle distribution was acquired with a FACScalibur (Becton Dickinson), and the percentage of cells in each phase of the cell cycle was analyzed with ModFit software.

Assessment of cell proliferation and cell death. Cells (5 × 10⁴ cells per mL) were transfected with siRNA and placed in a 96-well culture plate on day 0. On day 2, vehicle or 20 μg/mL insulin (Life Technologies Bethesda Research Laboratories) or 100 μmol/L palmitate (Sigma) was added to the medium. On day 3, cell proliferation, as measured as the number of viable cells, was evaluated at 450 nm absorbance using Cell Count reagent SF (Nacalai Tesque). The cell death rate (% of total population) was also determined by trypan blue exclusion assay. Palmitate-bovine serum albumin (BSA) complex was prepared according to the previous report (16).

Lipid synthesis. A549 cells were transfected with siRNA and were plated onto 60-mm dishes. After culturing for 72 h, the cells were incubated in the medium containing 4 μCi/mL D-[6-¹⁴C]glucose (GE Healthcare) for 4 h. Cellular lipids were extracted by a Bligh Dyer method (17). The rate of conversion of glucose into lipid was evaluated as described (12). Assay was performed in triplicate, and the results were expressed as percentage change in cpm per mg protein.

Immunohistochemistry, immunocytochemistry, and lipid staining. Tissue microarrays (TMA) consisting of 162 cases of lung adenocarcinoma were constructed for immunohistochemistry (IHC). IHC was performed on 5-μm thick, formalin-fixed, paraffin-embedded sections using primary antibodies for p-ACLY (Ser 454) and p-AKT (Ser 473; purchased from Cell Signaling Technology). Antigen retrieval was performed for 30 min in citrate buffer for each antibody. The slides were developed using the labeled streptavidin biotinylated peroxidase method (Nichirei) according to the manufacturer's instructions. For p-Akt immunostaining, the Tyramide Signal Amplication System (CSA II; DakoCytomation) was used. 3,3′-Diaminobenzidine tetrahydro-chloride was used as the chromogen and hematoxylin as the counterstain. All immunohistochemical staining was accomplished with a Dako Autostainer (DakoCytomation) under the same conditions. The expression and staining intensity on TMAs were scored semiquantitatively without knowledge of patient identity or outcome: 0, no; 1, weak; 2, moderate; 3, moderate to strong; 4, strong and scores were averaged across three cores. Patients were classified into two groups: p-ACLY-low (score 0–2) and p-ACLY-high (score 3,4). p-Akt expression was also divided into two groups according to the ACLY classification.

For immunocytochemistry, the cells were fixed with 4% paraformaldehyde 72 h after transfection. After PBS washing and treatment with 0.2% Triton X-100, the cells were blocked using Protein Block (DakoCytomation) for 10 min at room temperature and then incubated with p-ACLY (Cell Signaling Technology) or total ACLY (C-20 and G-17; Santa Cruz) antibodies 4°C overnight. Cells were washed and then incubated with FITC-conjugated secondary antibodies (DakoCytomation) for 1 h at room temperature. Cells were washed in PBS, mounted, counterstained with 4′,6-diamidino-2-phenylindole dihydrochloride, and visualized with the fluorescence microscope (Olympus).

Intracellular lipid accumulation was detected by oil red O (Sigma) or HCS LipidTOX phospholipidosis and steatosis detection kit (Invitrogen) according to the manufacture's protocol.

In vivo siRNA treatment of xenograft. Nude mice were inoculated with 1 × 10⁶ A549 cells. The negative control or ACLY siRNA#1 (5 μmol/L) with 0.05% atelocollagen (Koken) in a 200-μL volume was injected s.c. around the tumor, according to the manufacture's instruction. Injection was performed at weekly intervals for 5 wk. Each experimental condition included 10 animals per group. Tumor dimensions and body weight were measured weekly and the tumor volumes calculated using the formula: 1/2 a b², where a and b represent the larger and smaller tumor diameters, respectively. At the end of the experiment, tumors were excised for histologic analysis.

Tumor samples were fixed in 20% formaldehyde-PBS(−) and embedded in paraffin for H&E staining and IHC. IHC was performed for expression of p-ACLY and Ki-67, a proliferation marker (DakoCytomation).

Statistical analysis. For in vitro experiments, statistical analysis was performed using Student's t tests. An exponential regression model was used for evaluation of in vivo experiment. Kaplan-Meier survival curves were constructed for patients with p-ACLY–low and p-ACLY–high expression, and differences between survival curves were tested by a multivariate Cox regression model. Fisher's exact tests for contingency tables were used in Table 1. For all analyses, P value of <0.05 was considered statistically significant. Statistical analyses were performed using statistical programming language of R (18) and STATISTIKA (Statsoft, Inc.).

Positive correlation between p-Akt and p-ACLY expression levels in lung adenocarcinoma cell lines. To investigate ACLY regulation, we assessed ACLY expression as well as a transcriptional factor SREBP-1 expression at both mRNA and protein levels in various human lung adenocarcinoma cell lines. ACLY was expressed in all cell lines, protein levels closely paralleling mRNA expression (Fig. 1A and B). SREBP-1 is a member of the basic helix-loop-helix-leucine zipper family of transcriptional factors that play a central role in the expression of lipogenic enzymes. However, our data showed the inverse relationship between ACLY and SREBP-1 mRNA levels (Fig. 1A). Moreover, total ACLY protein levels were independent of either the 125 kDa precursor form of SREBP-1 or the 60 to 70 kDa mature, cleaved form (Fig. 1B). These findings thus suggested that ACLY protein expression is regulated at the transcription level, however by an as yet unknown transcription factor other than SREBP-1.

Notably, phosphorylated ACLY levels were likely to follow total ACLY levels, suggesting ACLY phosphorylation contributes to its protein stabilization. Furthermore, the increase in ACLY phosphorylation status seemed to follow p-Akt levels in most cell lines (Fig. 1B). These findings suggest Akt contributes to stabilize ACLY protein by means of its phosphorylation. In addition, p-Erk1/2, a mitogen-activated protein kinase (MAPK), levels were not correlated with ACLY phosphorylation (data not shown).

ACLY phosphorylation is directly regulated by PI3K-Akt pathway. Next, we determined whether ACLY is a downstream target of the PI3K-Akt pathway in lung cancer. Treatment with LY294002 or wortmannin (data not shown), PI3K inhibitors, resulted in the decrease in ACLY phosphorylation in a dose- and time-dependent manner in A549 cells, in parallel with a dramatic decrease in Akt phosphorylation (Fig. 1C). In addition, U0126, a MAPK inhibitor, did not affect p-ACLY expression (data not shown). To determine whether Akt, a downstream effecter of PI3K, directly regulates ACLY phosphorylation, constitutively active Akt or dominant-negative Akt were introduced into NCI-H23 and COR-L105 cells. The introduction of constitutively active Akt into the cells resulted in the enhancement of ACLY phosphorylation, whereas the dominant negative Akt attenuated ACLY phosphorylation (Fig. 1D). Total ACLY levels were not changed substantially.

ACLY is activated and a prognostic factor in human lung adenocarcinomas. We then assessed ACLY expression and activity in human lung adenocarcinoma tissues. The average mRNA level of ACLY was up-regulated 2.1-fold compared with that in corresponding normal tissue samples (P < 0.05; Fig. 2A). Western blot also revealed p-ACLY levels to be much higher in lung adenocarcinomas than in background lung tissue, whereas total ACLY levels were slightly increased in lung adenocarcinoma (Fig. 2B). In spite of higher amounts of the precursor form of SREBP-1 in tumors, the levels of the mature form tended to be lower than in corresponding normal tissue (Fig. 2B). In line with expression analysis, ACLY activity was significantly higher in tumors compared with normal tissue (P < 0.01; Fig. 2C).

To identify associations with clinicopathologic variables, IHC for p-ACLY was performed on TMA. p-ACLY expression in tumor tissue was detected in both cytoplasm and nucleus of tumor cells, whereas was barely detectable in the corresponding normal tissue. Among the 162 tumor samples, immunostaining intensity was variable (Fig. 2D,, a–c). There was a strong p-ACLY expression in 63.0% (102 of 162) of patients with lung adenocarcinoma. Comparison between clinicopathologic variables and p-ACLY levels revealed that positivity for p-ACLY was significantly associated with advanced stage (P < 0.01), the presence of pleural invasion (P < 0.01), and poor differentiation (P < 0.001; Table 1). These findings suggest that high expression of p-ACLY could be generally related to aggressive biological behavior of lung cancer.

Cytosomal and/or nuclear staining of p-Akt was observed in 62.3% (101 of 162) of lung adenocarcinoma cases (Table 1), and we found that the p-Akt and p-ACLY immunostaining patterns well correlated (Table 1; Fig. 2D , d and e).

Next, we examined the relationship between expression of p-ACLY and survival in patients with lung adenocarcinoma (n = 162). The 5-year survival was 96.5% in patients with p-ACLY–low lung adenocarcinomas, whereas it was 63% in patients with p-ACLY–high lung adenocarcinomas (P < 0.0001; Fig. 2E). On univariate Cox regression analyses, stage, tumor size, pleural invasion, gender, tumor differentiation, and p-ACLY expression were significantly associated with survival (P < 0.05). Furthermore, in the multivariate model with these significant covariates, p-ACLY expression was a significant factor for prediction of a poor prognosis (P = 0.0025), together with stage and tumor size (P = 0.00029 and P = 0.027, respectively).

ACLY is a therapeutic target in non–small cell lung cancer. To determine whether ACLY is a therapeutic target in lung cancer, we analyzed the phenotype of A549 cells treated with ACLY inhibition. All three oligonucleotides of ACLY siRNA resulted in an efficient knockdown of total ACLY protein levels, as determined by Western blot and immunocytochemistry (Fig. 3A). Phoshophorylated and total ACLY proteins were localized in the cytoplasm as well as in the nucleus (Fig. 3A,, arrows). Morphologically, ACLY knockdown was associated with curved, thinner, and elongated cells, in contrast to negative control siRNA. DNA flow cytometric analysis indicated that ACLY knockdown induced a progressive reduction in the S phase fraction associated with a progressive parallel accumulation of G₁ phase (control siRNA: G₁, 51.5%; S, 32.6%; G₂-M, 15.9%; ACLY siRNA: G₁, 79.0%; S, 4.0%; G₂-M, 17.1%; Fig. 3B). No significant increase of sub-G₁ cell population was observed at least 72 hours after ACLY knockdown. Microscopic observations of trypan blue staining revealed negligible cell death after ACLY knockdown (data not shown).

Next, to examine the effects of ACLY inhibition on lipogenesis and cell proliferation, we assessed ACLY activity, de novo fatty acid synthesis, intracellular lipid levels, and cell numbers. The treatment of LY294002 or ACLY knockdown resulted in a significant decrease in ACLY activity (Fig. 4A,, a). As anticipated, the decreased conversion rate of glucose to lipid was observed after the treatment of LY294002 or ACLY knockdown (Fig. 4A,, b), consistent with previous report (13). However, despite the impairment of de novo lipogenesis, lipidTOX, or oil red O (data not shown) staining revealed the markedly increased accumulation of phospholipids and neutral lipids in the cytoplasm after the ACLY knockdown (Fig. 4A,, c). Cell proliferation was decreased significantly in the cells treated with LY294002 (data not shown) or ACLY siRNA compared with negative control siRNA (Fig. 4B). To examine whether the growth arrest induced by ACLY inhibition was related to the impairment of glucose or lipid metabolism, ACLY knockdown cells were exposed for 24 hours to exogenous insulin or palmitate. The supplementation of insulin improved cell proliferation when compared with ACLY siRNA alone (P < 0.005; Fig. 4B), whereas palmitate markedly suppressed cell proliferation accompanied by cell death (P < 0.001; Fig. 4B).

Finally, to confirm the efficacy of ACLY inhibition in vivo, siRNA oligo with atelocollagen was injected into A549 cell xenografts in nude mice. ACLY siRNA treatment significantly inhibited tumor growth (P < 0.0001; Fig. 5A) and enhanced gland formation of A549 cells (Fig. 5B,, arrows), whereas no significant weight loss or other toxicity was observed in both groups (data not shown). It should be noted that loss of p-ACLY and Ki-67 expression was confined to the glandular structures of A549 cells, suggesting ACLY is involved in tumor differentiation and proliferation (Fig. 5B , arrows).

Metabolic changes are commonly observed in malignant cells. In the 1920s, Otto Warburg (19) discovered that cancer cells displayed high aerobic glycolysis, in contrast to normal cells. It was further found that cancers derive most fatty acids from de novo synthesis (20). Although the ultimate reasons for high glycolysis and lipogenesis in cancer cells are not completely understood, it seems that cancer cells can coordinate the related metabolic enzymes and can use all products effectively.

ACLY is the first key enzyme of lipogenesis and a crosslink between glucose and lipid metabolism. ACLY overexpression or activation have been reported in urinary bladder, breast, liver, stomach, colon, and prostate tumors (21–26). However, to the best of our knowledge, this is the first observation of ACLY activation in lung cancer.

It has been reported that ACLY activity is regulated by the PI3K-Akt signaling pathway through the phosphorylation (11, 27). Akt also up-regulates ACLY mRNA levels via activation of SREBP-1, a transcription factor of cholesterol and fatty acid biosynthesis (28, 29). We observed a positive correlation among mRNA, total protein, and phosphorylated protein levels of ACLY in lung adenocarcinoma cell lines and clinical samples. However, ACLY protein levels were independent of SREBP-1, and there was an inverse relationship between mRNA of ACLY and SREBP-1, consistent with previous reports (30, 31). Thus, activation of the PI3K-Akt pathway may stimulate ACLY activity predominantly through its phosphorylation rather than transcriptional up-regulation and, may in part, contribute to protein stabilization.

ACLY can be phosphorylated at different site by other kinase such as nucleoside diphosphate kinase (32) and cyclic AMP-dependent protein kinase (33). Our study showed that despite the treatment of PI3K inhibitors resulted in the rapid and the sufficient dephosphorylation of Akt, it did the dephosphorylation of ACLY with less efficiency. Furthermore, ACLY activity was also decreased with the treatment of PI3K inhibitors, but its effect on activity was not as great as that of ACLY knockdown. Therefore, this suggested that ACLY activity is also regulated by other than PI3K-Akt pathway in lung cancer.

ACLY is considered to be localized in the cytoplasm and generates cytosolic acetyl-CoA from citrate. However, we observed the nuclear and cytoplasmic localization of ACLY protein in cell lines as well as in clinical samples. ACLY as a nuclear phosphoprotein has been shown in HeLa cells (34), but the function of nuclear ACLY remains to be elucidated. Further study is required to understand the molecular basis of differential ACLY subcellular localization.

Recent evidence suggests that dysregulated lipogenesis plays a causal role in cancer pathogenesis (35). FASN, which is a downstream de novo lipogenic enzyme of ACLY, is frequently overexpressed in many cancers. Interestingly, the genes of key lipogenic enzymes including acetyl-CoA carboxylase (ACC), FASN, and ACLY are all located at the 17q chromosome, and gene amplification has been reported in prostate cancer (36). Importantly, inhibition of FASN induces apoptosis in a variety of cancer cells concomitant with the decrease in phospholipids (37, 38). Supplementation of palmitate, which is the fatty acid generated by FASN, rescues apoptosis induced by ACC or FASN inhibition (39–41). To investigate potential mechanisms for the growth arrest induce by ACLY inhibition, we assessed de novo lipogenesis and cell proliferation in an ACLY knockdown model. First, the conversion rate from glucose to lipid was dramatically decreased by ACLY knockdown concomitant with the decrease in cell proliferation. Next, we found that the intracellular lipid levels were increased by ACLY knockdown, albeit de novo synthesis was impaired. Finally, the supplementation of insulin could improve cell proliferation; however, palmitate induced cell death in these cells. Taken all together, it is tempting to speculate that ACLY inhibition may affect cell proliferation due to the impairment of glucose metabolism rather than due to the depletion of lipid products generated by fatty acids or cholesterol synthesis. Because ACLY integrates glucose metabolism with lipid synthesis, ACLY inactivation could potentially affect not only lipid metabolism but also glucose metabolism. Our data showed that ACLY inhibition, unlike FASN or ACC inhibition, lead to proliferation arrest but not apoptosis. One possible mechanism is that the salvage of extracellular lipids can prevent cell death. The precise mechanism of lipid accumulation induced by ACLY inhibition will require further study.

Importantly, p-ACLY overexpression was significantly correlated with tumor differentiation and poorer prognosis in our cohort. Because all patients enrolled in this study were operable with a relatively early stage, ACLY can be a useful marker for the identification of early-stage tumors with poorer prognosis. Besides, given that in vivo treatment with ACLY siRNA inhibited tumor growth in xenograft model, ACLY can be an attractive target for RNA interference therapy in the clinical setting.

There is a growing realization that signaling abnormalities within nutrient signaling pathways, for example loss-of-function mutations in the LKB1, TSC, and PTEN, can lead to transformation and tumor progression (42). Cancer cells need to obtain high energy and macromolecules effectively. Because ACLY integrates between glycolytic and lipogenic process, it may be important for the coordination of cancer cell metabolism.

Recent cancer research has highlighted the importance of lipid metabolism in cancer cells (35, 43). The present study provided a solid link between a metabolic network and lung cancer pathogenesis. We believe research into ACLY will open up opportunities for identifying new drugs for treatment of this deadly disease as well as other types of malignant tumors.

No potential conflicts of interest were disclosed.

Grant support: Parts of this study was supported financially by Grants-in-Aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology; Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science; and by grants from the Ministry of Health, Labour and Welfare, the National Institute of Biomedical Innovation, the Smoking Research Foundation, and the Vehicle Racing Commemorative Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Tomoyo Kakita and Hironori Murayama for their excellent technical assistance, and Savyon Cohen, Jonathan Cohen, and Dr. Malcolm Moore for editing the manuscript.