Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Expression of Fatty Acid Synthase Depends on NAC1 and Is Associated with Recurrent Ovarian Serous Carcinomas

Expression of Fatty Acid Synthase Depends on NAC1 and Is Associated with Recurrent Ovarian Serous... Hindawi Publishing Corporation Journal of Oncology Volume 2010, Article ID 285191, 12 pages doi:10.1155/2010/285191 Research Article Expression of Fatty Acid Synthase Depends on NAC1 and Is Associated with Recurrent Ovarian Serous Carcinomas 1, 2 3 4, 5 3 3 Stefanie M. Ueda, Kai Lee Yap, Ben Davidson, Yuan Tian, Vivek Murthy, 1, 2 2 3 1, 2 Tian-Li Wang, Kala Visvanathan, Francis P. Kuhajda, Robert E. Bristow, 3 1, 2, 3 Hui Zhang, and Ie-Ming Shih Department of Gynecology and Obstetrics, Johns Hopkins Medical Institutions, Baltimore, MD 21231, USA Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, MD 21231, USA Department of Pathology, Johns Hopkins Medical Institutions, Baltimore, MD 21231, USA Division of Pathology, Norwegian Radium Hospital, Rikshospitalet University Hospital, N-0310 Oslo, Norway Faculty Division Radiumhospitalet, The Medical Faculty, University of Oslo, N-0310 Oslo, Norway Correspondence should be addressed to Ie-Ming Shih, [email protected] Received 2 January 2010; Accepted 21 February 2010 Academic Editor: Ben Davidson Copyright © 2010 Stefanie M. Ueda et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Our previous reports demonstrated that NAC1, a BTB/POZ domain-containing nuclear protein, upregulates in recurrent ovarian serous carcinoma and participates in developing drug resistance in cancer cells. The current study applies quantitative proteomics to identify the proteins controlled by NAC1 by comparing the proteomes of SKOV3 cells with and without expression of a dominant negative NAC1 construct, N130. From the proteins that are downregulated by N130 (upregulated by NAC1), we chose to further characterize fatty acid synthase (FASN). Similar to change in protein level, the FASN transcript level in SKOV3 cells was significantly reduced by N130 induction or by NAC1 knockdown. Immunohistochemistry showed that NAC1 and FASN immunointensities in ovarian serous carcinoma tissues had a highly significant correlation (P< .0001). Moreover, we found that recurrent serous carcinomas exhibited higher FASN immunointensities than their matched primary tumors (P< .001). Multivariate analysis showed that an FASN staining score of >1 in serous carcinomas was associated with a worse overall survival time (P< .01). Finally, C93, a new FASN inhibitor, induced massive apoptosis in carboplatin/paclitaxel resistant ovarian cancer cells. In conclusion, we show that NAC1 is essential for FASN expression in ovarian serous carcinomas and the expression of FASN significantly correlates with tumor recurrence and disease aggressiveness. The dependence of drug resistant tumor cells on FASN suggests a potential application of FASN-based therapeutics for recurrent ovarian cancer patients. 1. Introduction studied ovarian cancer genome and transcriptome and have identified several genes and pathways that are potentially Ovarian cancer is a neoplastic disease that exemplifies involved in this phenotype. One of these genes, NACC1 many of the major issues underlying current chemotherapy encoding NAC1 (or NAC-1) protein, shows significantly regimens in clinical oncology [1, 2]. Although most ovarian higher expression in recurrent chemoresistant ovarian serous carcinomas at advanced stages are responsive to initial carcinomas than in primary untreated tumors [3]. NACC1 carboplatin and paclitaxel treatment, tumor clones resistant belongs to the BTB/POZ domain gene family and contains to these drugs eventually evolve as recurrent diseases. As the BEN domain that potentially mediates protein-DNA and a consequence, the main contributors to the mortality protein-protein interactions during chromatin organization and morbidity of advanced stage ovarian cancer patients and transcription [4]. Biologically, NAC1 has been demon- are chemoresistant tumors. In an effort to elucidate the strated to be an embryonic stem cell marker that controls molecular mechanisms underlying chemoresistance, we have proliferation and pluriopotency in embryonic stem cells 2 Journal of Oncology [5–7]. NAC1 homodimerizes through the highly conserved the same period were also analyzed. For Kaplan-Meier BTB/POZ domain (a.a. from 1–129) [8] and its complex survival analysis, we included 184 primary high-grade serous formation is essential for a variety of its biological functions. carcinoma patients who underwent optimal primary cytore- In our previous study, we found that recurrent ovarian ductive surgery followed by platinum-based chemotherapy serous carcinomas showed a higher NAC1 expression level at the Johns Hopkins Hospital. The paraffintissues were than their matched primary untreated tumors [3, 9]. Ectopic obtained from the surgical pathology repository at the Johns expression of NAC1 increased paclitaxel resistance while Hopkins Hospital and clinical data was obtained from medi- knockdown of NAC1 or disruption of NAC1 homodimer- cal records, including age, race, stage, histologic subtype, and ization sensitized cancer cells to chemotherapeutic drugs date of patient’s current status (alive or deceased). Median [3, 9]. In order to understand how NAC1 contributes to follow-up in those patients was 44.7 months. A subset of 56 drug resistance, we previously compared the gene expression high-grade serous carcinomas including 28 pairs of matched profiles of SKOV3 ovarian cancer cells to those of NAC1 primary and recurrent tumors was analyzed to correlate inactivated SKOV3 cells where the inactivation was induced FASN expression levels and to evaluate primary/recurrent by expression of N130, a mutant protein containing only tumor status. Another subset of 162 high-grade serous the BTB/POZ domain of NAC1 that competitively inhibits carcinomas was analyzed for correlation of NAC1 and FASN NAC1 homodimerization. We found that NAC1 negatively expression levels because the NAC1 immunostaining slides regulated the components of the Gadd45 tumor suppres- from these cases were available from our previous study sor pathway including Gadd45α and its binding protein, [3]. Collection of tissue samples was in accordance with the Gadd45gip1 [9, 10]. However, suppressing the Gadd45 guidelines of the institutional review board. tumor suppressor pathway did not completely rescue ovarian cancer cells after NAC1 inactivation, thus suggesting that 2.2. Quantitative Proteomics. Cell lysates were collected 48 other mechanisms also play a role. hours after induction (removal of doxycyclin) or mock In this study, we identified proteins that are potentially induction [3]. Protein concentration was measured by BCA regulated by NAC1 by applying a quantitative proteomic assay. The same amounts of proteins (1 mg) from each method using tandem mass spectrometry (MS/MS) and condition were reduced, alkylated, and digested with 1 to 50 spectral count [11]. A total of 2914 proteins were identified. Trypsin/protein ratio at 37 C overnight. The peptides were To reduce the sampling error and increase the quantification purified with C18 columns and resuspended in water with a accuracy, 208 proteins identified by at least 20 spectra and final concentration of 10 μg/μL. two unique peptides were quantified to identify candidate For protein identification and quantification, each pep- proteins controlled by N130 expression. By comparing the tide mixture was analyzed twice by the LTQ ion trap protein quantity in the proteomes between N130-induced mass spectrometer (Thermo Finnigan, San Jose, CA). For and noninduced SKOV3 cells, we identified new NAC1 each analysis, 2 μg of peptides were injected into a peptide regulated proteins that could be responsible for the NAC1- cartridge packed with C18 resin, and then passed through mediated drug resistance and for other biological functions. a10cm × 75 μm i.d. microcapillary HPLC (μLC) column Among the proteins found to be downregulated in N130- packed with C18 resin. A linear gradient of acetonitrile from induced cells, we selected fatty acid synthase (FASN) for 5%–32% over 100 minutes at flow rate of ∼300 nL/min was further characterization because it has been shown to be applied. During the LC-MS mode, data was acquired in the associated with tumor progression in a variety of human can- m/z range of 400 and 2000. The MS/MS was also turned on cers and its inhibitors are available for potential translation to collect CID using data dependent mode. studies. MS/MS spectra were searched with SEQUEST against the human IPI protein database (version 2.28). The peptide mass tolerance is 3.0 Da. Other parameters of database searching 2. Methods are as follows: cysteine modification (add cysteine 57) and oxidized methionine (add methionine with 16 Da). The 2.1. Cell Lines and Clinical Tissue Samples. High-grade output files were evaluated by INTERACT [13] and Peptide ovarian carcinoma cell lines, SKOV3, A2780, and OVCAR3, Prophet [14]. The identified peptides with a probability were obtained from the American Type Culture Collection score ≥ 0.9 were used for the spectral count. To determine (Rockville, MD), and a low-grade serous cell line, MPSC1, the number of MS/MS spectra used for identification of was previously established by us [12]. OSE10 was as SV- each proteinindifferent conditions using our in-house 40-immortalized ovarian surface epithelial cell line. All developed software tool and to reduce the error for protein cell lines were maintained in RPMI media with 5% heat- identification and quantification using MS/MS spectra, we inactivated fetal bovine serum (HyClone, Logan, UT) and only quantified proteins identified by at least 20 spectra 2% penicillin/streptomycin (Gibco, Rockville, MD). and 2 independent peptides from the N130-induced and- A total of 427 ovarian carcinoma tumors collected noninduced cells. between January 1990 and December 2006 were arranged in tissue microarrays. These included 269 high-grade serous, 45 low-grade serous, 45 endometrioid, and 68 clear cell 2.3. Western Blot and Real Time PCR. Western blot analysis carcinomas. Fifteen histologically normal ovarian tissues was performed on the protein lysates prepared from ovarian and 20 benign ovarian serous cystadenomas collected over cancer cell lines and OSE10. Similar amounts of total Journal of Oncology 3 proteinfromeachlysatewereseparated on 10% Tris-Glycine- 2.6. Lentivirus Production and Transduction. Two NAC1 SDS polyacrylamide gels (Novex, San Diego, CA) and then targeting shRNA plasmids previously cloned into the pLKO.1 electroblotted to Millipore Immobilon-P polyvinylidene (Sigma) and the packaging lentivirus plasmids were cotrans- difluoride membranes. The membranes were probed with an fected into 293FT cells using Lipofectamine 2000 (Invit- anti-FASN mouse monoclonal antibody (1 : 100) (FASgen, rogen. The sequences of NAC1 targeting shRNAs were Baltimore, MD) followed by a peroxidase-conjugated goat 5 -CCGGCCCAAGTGAGATTG CACATTTCTCGAGAAA- antimouse immunoglobulin (1 : 6,000). Western blots were TGTG CAATCTCACTTGGGTTTTTTG-3 (shRNA-A) and developed by chemiluminescence (Pierce, Rockford, IL) 5 -CCGG CAAGTACTACTGCC AGAACTTCTCGAG AA- utilizing glyceraldehyde-3-phosphate dehydrogenase as the GTTCTGGCAGTAGTACTTGTTTTTTG-3 (shRNA-C). loading control. To determine the mRNA levels of FASN, Five hours after incubation, the transfection reagents were we performed quantitative real-time PCR using a BioRad replaced with 10% FBS RPMI culture media supplemented iCycler. Total RNA was isolated with the TRIzol method with bovine serum albumin. The virus-containing super- (Invitrogen) and cDNA was synthesized from 2 to 5 μgtotal natant was collected, the mixture was centrifuged, and RNA. FASN primer sequences were 5 -CATCCAGATAGG- polybrene was added to a final concentration of 8 μg/mL. CCTCATAGAC-3 (forward) and 5 -CTCCATGAAGTA- Transduction was carried out by adding 0.25 mL of virus GGAGTGGAAG-3 (reverse). The expression of FASN was supernatant to the SKOV3 cells. A second transduction was normalized to that of human amyloid beta precursor protein performed 24 hours later using the same protocol. shRNA or based with threshold cycle numbers calculated from dupli- control (vector only) virus transduced cells were enriched cate measurements. Mean fold expression differences were by adding 2 μg/mL of puromycin. Following this second further normalized to those of ovarian surface epithelium, round of transduction, cells were collected at 48 and 72 hours OSE10. for quantitative real-time PCR analysis of NAC1 and FASN mRNA expression levels. 2.4. Immunohistochemistry. For immunohistochemistry, paraffinsectionsafter deparaffinization were incubated with 3. Results a primary antimouse FASN antibody at a dilution of 1 : 50 in a4 C moist chamber overnight. Negative controls included 3.1. High-Throughput Quantitative Proteomics Identifies benign serous cystadenomas and normal ovaries. Two FASN as an NAC1-Regulated Protein. To determine the pos- independent observers scored the FASN immunoreactivity sible mechanism by which NAC1 promotes drug resistance, using a categorical scoring system from 0 (not detectable) to we used the quantitative proteomics to compare the pro- 3 (intense) with the mean score recorded from triplicates. teomes of N130-induced and-noninduced SKOV3 ovarian Among the FASN stained cases, there were 162 that had been cancer cells using tandem mass spectrometry (MS/MS) and previously stained with an anti-NAC1 antibody [3, 15]. spectral count [11]. We were able to identify proteins that corresponded to a total of 2914 proteins. To reduce the sam- 2.5. Cell Number and Apoptosis Detection. Cancer cells were pling error and increase the quantification accuracy, 208 pro- seeded in 96-well plates at 2.5 × 10 cells/well for 24 hours teins identified by at least 20 spectra and two unique peptides and then were incubated with 100 μL of the FASN inhibitor, were quantified to identify candidate proteins controlled by C93, at concentrations ranging from 2.5 μg/mL to 50 μg/mL N130 expression. Based on the protein expression ratio of for 48 hours. After incubation, the number of viable N130 noninduced to N130 induced cells (N130 OFF versus cells was measured by the CellTiter-Blue assay (Promega, N130 ON), we listed the 19 proteins whose abundances Madison, WI). These numbers were plotted against FASN were at least increased 50% (OFF/ON ≥ 1.5, Table 1)and inhibitor concentrations, and the value of IC (i.e., the identified 18 proteins whose levels were at least increased C93 concentration at which cell growth dropped to 50% of 50% (ON/OFF ≥ 1.5) by N130 induction (Table 1). the control level) was estimated. To detect early apoptotic Cells with induced expression of N130 showed significantly cells, we grew ovariancancercelllinesin6wellplates(5 × higher levels NAC1 expression than N130-noninduced cells 10 cells/well) and treated them with C93 at their IC and (ON/OFF = 76.5), and all the identified peptides from with DMSO of equal concentration. The early apoptotic cells NAC1 matched the N130 (BTB/POZ) domain, indicating the were quantified utilizing the Annexin V-FITC detection kit robustness of N130 induction. Since our primary interest (Biovision, Mountain View, CA) and annexin V-stained cells was to identify the proteins that are upregulated by NAC1 were determined by flow cytometry. (i.e., downregulated by N130), we selected fatty acid synthase After treating cells with C93 or DMSO, cells were (FASN) for validation and characterization since its level trypsinized, washed, and resuspended in a solution contain- in noninduced cells is 1.57 times that in induced cells. We ing 0.6% NP-40, 3.7% formaldehyde, and 11 mg/mL Hoechst made this choice because, more than other proteins in the 33258 in a phosphate buffered solution. The stained cells list, FASN has been associated with tumor progression and were then quantified by the BD LSR cytometer (Becton because specific inhibitors of FASN are available for further Dickinson, Franklin Lakes, NJ). Cell cycle analysis was studies. The levels of FASN are less abundant in N130 performed using the CellQuest software (Becton Dickinson) induced cells than in N130-noninduced cells, suggesting and cells in the subG phase that represent the late phase of that NAC1 inactivation by N130 suppresses FASN protein apoptosis were also measured. expression in SKOV3 cells. 4 Journal of Oncology Table 1: Proteins that are differentially expressed in N130-induced and-noninduced SKOV3 cells. (a) Proteins downregulated by N130 induction ON- ON- OFF- OFF- Total IPI Protein name Off/on spectra peptides spectra peptides spectra Heterogeneous nuclear IPI00013881 4 5 20 7 24 4.62 ribonucleoprotein H Transcription intermediary IPI00012388 8 7 23 9 31 3.07 factor 1-beta IPI00220985 Keratin, type I cytoskeletal 18 8 4 18 5 26 2.35 RNA-binding protein regulatory IPI00298547 10 7 23 7 33 2.30 subunit IPI00299571 Protein disulfide isomerase A6 11 3 21 8 32 1.91 aldolase C, IPI00216976 10 3 19 4 29 1.90 fructose-bisphosphate Heterogeneous nuclear IPI00012074 9 5 17 7 26 1.89 ribonucleoprotein R IPI00013917 40S ribosomal protein S12 10 3 19 3 29 1.85 T-complex protein 1, beta IPI00297779 10 5 18 8 28 1.80 subunit IPI00027434 Transforming protein RhoC 8 5 14 5 22 1.75 IPI00141318 P63 protein 8 6 13 5 21 1.73 IPI00219217 lactate dehydrogenase B 11 6 18 8 28 1.67 IPI00219096 high-mobility group box 1 15 4 25 5 40 1.67 IPI00009236 Caveolin-1 9 2 15 2 24 1.67 Probable ATP-dependent RNA IPI00328343 11 6 18 6 29 1.64 helicase p47 T-complex protein 1, delta IPI00302927 8 3 13 6 21 1.63 subunit IPI00328188 fatty acid synthase 21 7 33 11 54 1.57 IPI00015027 AHNAK-related protein 11 7 17 11 27 1.55 Probable RNA-dependent IPI00017617 11 8 17 7 28 1.50 helicase p68 (b) Proteins upregulated by N130 induction ON- ON- OFF- OFF- Total IPI Protein name On/off spectra peptides spectra peptides spectra IPI00017454 Hypothetical protein FLJ13940 23 4 0 0 23 ? IPI00045207 NAC1 protein 153 9 2 1 155 76.50 IPI00395440 Unknown 26 5 5 4 31 5.13 IPI00107117 Peptidylprolyl isomerase B 14 5 6 3 20 2.33 T-complex protein 1, alpha IPI00290566 14 66320 2.33 subunit similar to Heterogeneous nuclear IPI00176692 27 5 12 6 39 2.25 ribonucleoprotein A1 IPI00221088 ribosomal protein S9 17 6 8 6 25 2.13 IPI00215918 ADP-ribosylation factor 4 17 7 9 7 26 1.89 IPI00015786 Spectrin alpha chain, brain 53 27 30 18 83 1.77 IPI00152412 Hypothetical protein 14 4 8 5 22 1.75 IPI00020984 Calnexin 29 10 17 7 46 1.71 Chloride intracellular channel IPI00010896 17 5 10 4 27 1.70 protein 1 T-complex protein 1, zeta IPI00027626 18 9 11 9 29 1.64 subunit IPI00216587 40S ribosomal protein S8 13 6 8 6 21 1.63 Journal of Oncology 5 (b) Continued. ON- ON- OFF- OFF- Total IPI Protein name On/off spectra peptides spectra peptides spectra IPI00216318 tyrosine 3-monooxygenase 40 10 25 9 65 1.58 60 kDa heat shock protein, IPI00329351 66 21 44 17 110 1.52 mitochondrial IPI00217468 H1 histone family, member 5 21 5 14 4 35 1.50 Malate dehydrogenase, IPI00291006 15 6 10 4 25 1.50 mitochondrial Table 2: Immunointensities of NAC1 and FASN in high-grade obtained from ovarian cancer ascites and tissues for NAC1 serous carcinoma. and FASN and correlated their immunointensities. The specificity of the anti-NAC1 has been demonstrated previ- FASN 2+/3+ FASN 0/1+ ously [3] and the specificity of the anti-FASN monoclonal NAC1 intense 51 34 antibody is shown in Figure 2(a). Western blot analysis NAC1 weak 19 58 shows a single protein band at the molecular mass of FASN Total (n)70 92 protein in all three ovarian cancer cell lines (OVCAR3, A2780, and SKOV3) but not in a low-grade serous carci- noma cell line, MPSC1, nor in ovarian surface epithelial Table 3: FASN immunointensity in primary and recurrent ovarian cells (OSE10). Analyzing 162 high-grade serous carcinomas serous carcinomas. showed a significant positive correlation in NAC1 and Score FASN immunointensities (P< .0001, Fisher Exact test) Tumor 0/1+ 2+/3+ (Table 2). These immunointensities for four representative Primary 18 10 high-grade serous carcinomas are shown in Figure 2(b).Our observations support the view that FASN expression is at Recurrent 4 24 least in part regulated by NAC1. Total (n)22 34 χ P< .0001. 3.3. FASN Expression in Ovarian Serous Tumors. To extend the above immunostaining findings, we investigated FASN 3.2. Validation of NAC1-Dependent FASN Expression. To immunoreactivity in various types of ovarian tumors. All validate the proteomic result above, we used quantitative benign cystadenomas and normal ovarian surface epithe- real-time PCR to determine FASN transcript levels in two lium (n = 35) display undetectable to very low FASN independent experimental systems. In our first system, staining (mean score = 0) whereas ovarian carcinomas of we used the same NAC1 dominant negative (N130) cell various histologic subtypes show FASN immunoreactivity model that was used in the proteomic analysis to assess (mean score ≥ 1) in most cases (Figures 2(c) and 2(d)). whether disruption of NAC1 homodimerization by N130 The FASN staining score of high-grade serous carcino- leads to a decrease in FASN mRNA level. We observed that mas is significantly higher than that of low-grade serous N130 mRNA, as detected by a PCR primer pair that amplifies carcinomas (P< .0001, t-test) and normal ovaries and the N130 region, increases 48 and 72 hours after induction benign cystadenomas (P< .0001). The FASN score in high- (Figure 1(a)). At these same time points, SKOV3 cells with grade serous carcinomas is marginally higher than that in N130 induction exhibit a downregulation of FASN mRNA. clear cell carcinoma (P = .024) and there is no statistical Of note, at 48 hours after N130 induction, the decrease significance of FASN score between high-grade serous and in FASN expression is similar between mRNA (46%) and endometrioid carcinoma (P = .099). Since NAC1 has protein (36%) levels. In our second system, we knocked been shown to be highly expressed in recurrent rather than down NAC1 in SKOV3 cells to determine the FASN levels. primary high-grade ovarian serous carcinomas, we expect Two NAC1 shRNAs (shRNA-A and -C) that target different that FASN expression levels will follow the same pattern. To coding regions of NACC1 were designed and packaged test this possibility, we assessed the FASN immunoreactivity into lentivirus. We found that both shRNAs effectively of matched primary and first recurrent tumors from the reduce NAC1 transcript levels (to less than 10% of control) same individuals. As expected, the FASN staining score is after transduction with shRNA lentivirus (Figure 1(b)). significantly higher in recurrent than in primary tumors Correspondingly, FASN expression levels also decrease as (P< .0001, paired t-test) (Figure 3(a)). Based on a compared to control. The above findings from cell culture 2 × 2 contingency table and chi-square analysis, we also systems suggest that NAC1 expression and its dimerization found that recurrent tumors exhibit a higher percentage of domain are essential for maintaining FASN expression in cases with intense FASN immunoreactivity than primary tumor cells. tumors do (Table 3). Figure 3(b) illustrates representa- In order to assess the biological significance of NAC1- tive pairs of primary and recurrent tumors with FASN dependent FASN expression, we stained tumor samples stain. 6 Journal of Oncology NAC1 (N130) FASN NAC1 FASN (a) (b) Figure 1: NAC1-dependent FASN expression in ovarian cancer cells. (a) N130 mRNA, as detected by a PCR primer pair that amplifies the N130 region, increases after N130 induction. In contrast, FASN mRNA decreases 48 and 72 hours after induction. (b) After shRNA lentivirus transduction,bothshRNAs(-A and-C) effectively reduce NAC1 transcript levels to less than 10% of control and reduce FASN expression. 3.4. Clinical Significance of FASN Expression in High-Grade for parental, paclitaxel, and carboplatin resistant SKOV3 Ovarian Serous Carcinomas. To evaluate the clinical signif- cells, respectively; 7.4 μg/mL, 7.5 μg/mL, and 8.4 μg/mL for icance of FASN expression, we correlated the expression parental, paclitaxel, and carboplatin resistant A2780 cells, levels in primary high-grade serous carcinoma tumors respectively; and 8.6 μg/mL, 6.5 μg/mL, and 8.8 μg/mL for with overall survival of the patient. We found that higher parental, paclitaxel, and carboplatin resistant OVCAR3 cells, FASN staining scores (scores > 1) correlated with worse respectively. When applied to each cell line at its IC overall survival (P< .002) (Figure 3(c)). Patients with concentration, C93 significantly increases the percentage of primary tumors that showed minimal or undetectable FASN annexin V stained cells (representing the early phase of immunoreactivity (score ≤ 1) had a median survival time apoptosis) and the sub-G fraction in cell cycle analyses of 60.4 months (range 1–193 months), whereas those with (representing the late phase) (Figure 4). Thus, C93 induces tumors that showed positive immunoreactivity (score > 1) apoptosis in all three ovarian cancer cell lines. The number of had a median survival time of 36.9 months (range 1–140 annexin V-stained cells increases in a time dependent fashion months, P< .01). After adjusting for age, stage, and race in all cancer cell lines including those resistant to carboplatin in a multivariate analysis, we found that, in patients with and paclitaxel (Figure 4(b)). Moreover, paclitaxel resistant primary serous carcinomas, FASN expression remains an cells are more sensitive to C93 than carboplatin resistant independent marker for prognosis with a hazard ratio of 1.87 cells, especially in the A2780 and SKOV3 cell lines. In all (95% CI: 1.12–3.11, P = .02). three cell lines, the C93-treated group shows a significantly higher percentage of sub-G cells than the DMSO-treated group (P< .01). These sub-G cells can be detected as 3.5. C93 Suppresses Growth of Paclitaxel-Resistant and Car- early as 12 hours after C93 treatment and become more boplatin-Resistant Ovarian Cancer Cells. The above findings pronounced by 48 hours (Figure 4(c)). C93 fails to affect cell suggest that FASN preferentially expresses in recurrent and cycle progression in SKOV3 and OVCAR3 cells by 96 hours in most aggressive types of ovarian serous carcinomas, after treatment but it arrests A2780 cells in the G phase as raising the possibility that FASN expression contributes to early as 24 hours after treatment (data not shown). this phenotype. Thus, to determine if FASN expression is essential for cell growth and survival of high-grade serous carcinoma cells (including those that are resistant 4. Discussion to paclitaxel or carboplatin), we applied C93, a second generation FASN inhibitor, to ovarian cancer cell lines One of the major challenges facing ovarian cancer patients is (including SKOV3, A2780, and OVCAR3). We first used the the development of chemoresistant tumors after cytoreduc- cell number counted by CellTiter Blue assay to determine tion surgery and chemotherapy. In this study, we provide new the IC of C93 for each ovarian cancer cell line. The evidence that homodimerization of NAC1, a drug resistance- IC of C93 was 7.4 μg/mL, 7.4 μg/mL, and 8.7 μg/mL associated nuclear protein, is essential for maintaining FASN mRNA level (% of non-induced control) 48h-cont 48h-N130 72h-cont 72h-N130 mRNA levels (% of pLKO control) 48h-pLKO 48h-shNAC1-A 48h-shNAC1-C 72h-pLKO 72h-shNAC1-A 72h-shNAC1-C Journal of Oncology 7 FASN Tumor A Tumor B Tumor C Tumor D 3+ 0 3+ 1+ NAC1 3+ 0 3+ 1+ GAPDH FASN (a) (b) Normal ovary serous carcinoma (solid tumor) 3.5 2.5 Serous cystadenoma Serous carcinoma 1.5 (ascites) 0.5 (c) (d) Figure 2: Expression of NAC1 and FASN in ovarian tumor tissues. (a) Western blot analysis shows a single protein band that corresponds to FASN protein in all three ovarian cancer cell lines (OVCAR3, A2780, and SKOV3) but not in a low-grade serous carcinoma cell line (MPSC1) nor in ovarian surface epithelial cells (OSE10). (b) Immunoreactivity of NAC1 (upper panels) and FASN (bottom panels) in four representative high-grade serous carcinomas. For each staining, the immunostaining scores are shown in the upper left corner (c)-(d). FASN expression in different types of ovarian carcinomas. (c) FASN immunointensity scores for different types of ovarian carcinoma including high-grade serous, low-grade serous, clear cell, and endometrioid carcinomas. The mean and standard deviation of the immunostaining score are shown. (d) Representative tumor sections of FASN immunostaining are illustrated. FASN immunoreactivity is detectable in high- grade serous carcinoma cells but not in normal ovarian surface epithelium nor in cystadenomas. expression at both protein and mRNA levels in ovarian Mammalian FASN is a ∼260 kD cytoplasmic enzyme that cancer cells. We also demonstrate that FASN expression is responsible for all the steps of de novo fatty acid synthesis. It is highly correlated with the status of recurrent chemore- catalyzes the NADPH-dependent condensation of malonyl- sistant ovarian serous carcinomas and is independently CoA and acetyl-CoA to palmitate [16]. Importantly, normal correlated with poor overall survival. Suppressing FASN adult tissues express minimal amounts of FASN due to enzyme activity with its inhibitor induces apoptosis in cancer the presence of abundant dietary lipids. In contrast, a cells that are resistant to paclitaxel and carboplatin. Thus, variety of tumors including breast, colon, ovary, and prostate molecular studies that illuminate the fundamental properties cancer express elevated FASN levels [17–23], since tumor of chemoresistance should provide new therapeutic targets cells become less sensitive to regulatory nutritional signals for treating recurrent ovarian cancers. and prefer the de novo lipogenesis pathway. Furthermore, Mean FAS IHC score OSE10 High-grade serous MPSC1 OVCAR3 Low-grade serous A2780 Clear cell SKOV3 Endometrioid Cyst and OSE 8 Journal of Oncology Primary Recurrent 4 P< .0001 Patient Patient Primary Recurrent (a) (b) P< .01 0 50 100 150 200 Survival (months) FASH IHC score ≤ 1 FASH IHC score > 1 (c) Figure 3: Clinical significance of FASN expression in high-grade ovarian serous carcinoma. (a) Comparison of FASN immunointensity scores for 28 pairs of matched primary and recurrent tumors from the same patients. Recurrent tumors exhibit elevated expression when compared to primary specimens from the same patient (P< .0001). (b) FASN staining of the primary and recurrent tumors from two representative patients. (c) Kaplan-Meier curve analysis shows that patients whose tumors exhibit higher FASN immunostaining scores have significantly shorter survival times than those whose tumors show undetectable or very low FASN immunoreactivity (60.4 versus 36.9 months, P< .01). FASN expression level highly correlates with the clinical KLF5/SREBP-1 function upstream of FASN [29, 30]. Thus, aggressiveness of tumors [17, 18, 24–27]. at least in ovarian cancer cells, the NAC1 pathway represents Although previous reports have shown the role of FASN another mechanism for controlling FASN expression. Unlike expression in ovarian cancer [18, 28], the current study other members of the BTB/POZ family, NAC1 lacks the provides new findings that should have several biological zinc finger DNA-binding domain. Rather, it has been and clinical implications. First, we use in vitro cell culture reported to act as a transcription corepressor with other studies using NAC1 targeting shRNAs and the dominant BTB/POZ proteins [31]. Moreover, it has also been shown negative NAC1 (N130) approach to demonstrate that FASN to interact with nuclear proteins potentially involved in is one of the proteins regulated at least in part by NAC1. tumorigenesis, including Nanog [7], CoREST [32], and This NAC1-dependent FASN expression in protein level HDAC3 and HDAC4 [33]. Thus, it is possible that FASN is further supported by its mRNA level and the positive expression is indirectly controlled by NAC1 through binding correlation between NAC1 and FASN immunointensities with its specific partner(s). Identification of the NAC1-FASN in ovarian serous carcinoma samples. The mechanism pathway sheds new light on the molecular mechanism by underlying upregulation of FASN in human cancer is not which NAC1 promotes tumor progression. Further studies clear and it likely involves multiple pathways. Previous are required to elucidate the transcriptional regulation of studies have shown that, in prostate cancer, caveolin-1 and FASN by NAC1. FASN IHC score Patients (%) Journal of Oncology 9 SKOV3 80 A2780 OVCAR3 50 40 20 20 0 0 0 0 8 18 24 36 48 0 8 18 24 36 48 0 8 18 24 36 48 Hours of incubation Hours of incubation Hours of incubation DMSO DMSO DMSO C93 C93 C93 (a) 50 60 10 10 0 0 0 8 18 24 36 48 0 8 18 24 36 48 0 8 18 24 36 48 Hours of incubation Hours of incubation Hours of incubation TR TR TR TR CR TR SKOV3 C93 SKOV3 DMSO A2780 C93 A2780 DMSO OVCAR3 C93 OVCAR3 DMSO CR CR CR CR TR CR SKOV3 C93 SKOV3 DMSO A2780 C93 A2780 DMSO OVCAR3 C93 OVCAR3 DMSO (b) SKOV3 OVCAR3 A2780 DMSO C93 (c) Figure 4: C93 treatment promotes apoptosis in ovarian cancer cells including cells that are resistant to carboplatin and paclitaxel. (a) In three ovarian cancer cell lines, C93 at its IC concentration increases the percentage of annexin V stained cells in a time dependent fashion. DMSO was used as the vehicle control. (b) C93 also increases annexin V stained cells in carboplatin resistant (CR) and paclitaxel resistant (TR) cell lines. (c) In all three cell lines, C93 treatment leads to an increase in sub-G cells when compared to the DMSO-treated group (P< .001). The percentage of sub-G cells is measured 48 hours after C93 treatment. The second finding in the current study involves the palmitic acid as a result of FASN overexpression [34]. Despite positive correlation between FASN expression and recur- recent advances in discovering drug-resistance biomarkers rence status in ovarian serous carcinoma tissue. It has been that are associated with recurrent and chemoresistant ovar- shown that ectopic overexpression of FASN results in drug ian cancer [1, 3, 15, 36–41], reversing drug resistance by resistance and that reducing the FASN expression increased targeting these markers remains a tantalizing objective. The the drug sensitivity in breast cancer cell lines [34]. At low problem lies in the lack of reagents that can be used to inhibit concentrations, FASN inhibitor also sensitized tumor cells these genes in clinical studies. Thus, therapeutic targeting with FASN overexpression to chemotherapeutic agents [34, of FASN provides an attractive option since selective small 35]. The FASN-mediated drug resistance appears to be due compound inhibitors for this protein are available. The first to a decrease in drug-induced apoptosis due to abundant generation of FASN inhibitors, including C75 and Orlistat, Annexin V stained cells (%) Annexin V stained cells (%) Annexin V stained cells (%) Annexin V stained cells (%) Sub-G cells (%) Annexin V stained cells (%) Annexin V stained cells (%) 10 Journal of Oncology potently inhibited tumor growth in a mouse xenograft model [3, 15], hence suggesting that the NAC1-FASN pathway con- but the adverse effects associated with these drugs prevented stitutes one of the mechanisms that propels ovarian cancer their further consideration for clinical applications [23]. On progression. How FASN contributes to disease aggressiveness the other hand, C93, the second generation FASN inhibitor in ovarian cancer remains speculative. Besides endowing used in this study, pharmacologically eliminates concomitant drug resistance, FASN may enhance oncogenesis via cellular CPT-1 stimulation and does not induce the anorexia and mechanisms such as enhancing the Wnt [53], c-Met [54], feeding behavior changes in mice that were caused by earlier and proteosome pathways [55]. Moreover, upregulation of generation FASN inhibitors [28, 42]. These features are FASN gives cancer cells a growth and survival advantage by important for any FASN inhibitor before it can be considered blocking apoptosis under hypoxia, a common condition in for clinical testing. Our in vitro studies demonstrate that solid tumors and tumor effusions [56, 57]. C93 affects both carboplatin and paclitaxel resistant ovarian In summary, this study identified candidate proteins cancer cells. Thus, ovarian cancer cells that overexpress controlled by NAC1 and provides new evidence that FASN FASN are molecularly dependent on it for cell survival. This expression is at least in part regulated by NAC1. Thus the observation is significant because FASN inhibitors provide NAC1-FASN pathway may represent a new mechanism for an alternative treatment for ovarian cancer patients who tumor progression that creates ovarian tumor cells that have developed recurrent tumors after initial paclitaxel and are resistant to chemotherapy. Our findings also indicate carboplatin treatment. that FASN is a novel biomarker for recurrent ovarian The antitumor effects of FASN inhibitors, like C93, serous carcinoma and its enzyme activity is essential for are thought to result from depletion of end product fatty the survival of chemoresistant tumor cells. New generation acids with accumulation of toxic intracellular malonyl-CoA FASN inhibitors, like C93, deserve consideration in future and altered production of phospholipids with diminished clinical trials involving advanced ovarian serous carcinomas, membrane synthesis [23, 43]. Alternatively, FASN inhibitors particularly those that are refractory to paclitaxel and plat- may suppress tumors through metabolism-independent inum drugs. Further studies will be required to delineate the mechanisms. For example, FASN inhibition has been shown biological and translational roles of FASN in drug resistance to selectively activate AMP-activated protein kinase (AMPK) in ovarian and perhaps other types of cancers. in ovarian cancer cells causing cytotoxicity while sparing most normal human tissues from these pleiotropic effects of Abbreviations AMPK activation [28]. Moreover, a positive feedback regu- lation has been reported in ovarian carcinoma cells between FASN: Fatty acid synthase. AKT activation and FASN expression [44]. Phosphorylated NAC1: Nucleus accumbens associated 1. AKT significantly correlates with FASN expression and FASN inhibition by either C75 or cerulenin downregulates phosphorylated AKT [44–46]. Therefore, FASN inhibitors Acknowledgment may contribute to antioncogenesis by suppressing tumor- promoting signaling pathways such as AKT, a pathway that This study is in memory of Ms. Sean Patrick, founder is frequently activated in ovarian serous carcinoma [47]. of the HERA Women’s Cancer Foundation, who coura- geously fought against recurrent ovarian cancer. The work Thus, the C93-induced apoptosis in ovarian cancer cells may be related to FASN inactivation and/or suppression of is supported by NIH/NCI RO1CA103937 (IMS), the HERA AKT activity. It has been demonstrated that FASN inhibition Women’s Cancer Foundation, and the Entertainment Indus- initiates apoptosis more effectively in neoplastic cells with try Foundation (SMU). mutant TP53 than in those with wild-type TP53 [27, 48]. Our current finding supports this view as the A2780 cell line References that harbors wild-type TP53 [49] responds to C93 with cell cycle arrest in the G phase in addition to apoptosis, while [1] K. R. Cho and I. M. Shih, “Ovarian cancer,” Annual Review of both the SKOV3 and OVCAR3 cell lines with their deleted Pathology: Mechanisms of Disease, vol. 4, pp. 287–313, 2009. and mutated TP53, respectively, respond to C93 with massive [2] S. B. Kaye, “Reversal of drug resistance in ovarian cancer: apoptosis [50]. Since mutated TP53 appears in the majority wheredowegofromhere?” Journal of Clinical Oncology, vol. of ovarian high-grade serous carcinomas [1, 51], it is likely 26, no. 16, pp. 2616–2618, 2008. that apoptosis is the predominant antitumor mechanism of [3] K. Nakayama, N. Nakayama, B. Davidson, et al., “A BTB/POZ FASN inhibition. protein, NAC-1, is related to tumor recurrence and is essential As with other epithelial cancers of the breast, colon, for tumor growth and survival,” Proceedings of the National and prostate [20, 21, 52], FASN overexpression in ovar- Academy of Sciences of the United States of America, vol. 103, ian cancer, as this study shows, appears to be associated no. 49, pp. 18739–18744, 2006. with the most malignant type, that is, high-grade serous [4] S. Abhiman, L. M. Iyer, and L. Aravind, “BEN: a novel domain carcinoma. More importantly, in patients with this cancer, in chromatin factors and DNA viral proteins,” Bioinformatics, the level of FASN expression significantly correlates with vol. 24, no. 4, pp. 458–461, 2008. worse clinical outcome, supporting the view that FASN [5] J. Kim, J. Chu, X. Shen, J. Wang, and S. H. Orkin, “An extended expression contributes to disease aggressiveness in cancer transcriptional network for pluripotency of embryonic stem cells. A similar observation has been reported for NAC1 cells,” Cell, vol. 132, no. 6, pp. 1049–1061, 2008. Journal of Oncology 11 [6] J. Wang, S. Rao, J. Chu, et al., “A protein interaction network [22] L. Z. Milgraum,L.A.Witters,G.R.Pasternack, andF.P. for pluripotency of embryonic stem cells,” Nature, vol. 444, no. Kuhajda, “Enzymes of the fatty acid synthesis pathway are 7117, pp. 364–368, 2006. highly expressed in in situ breast carcinoma,” Clinical Cancer [7] T. Ma, Z. Wang, Y. Guo, and D. Pei, “The C-terminal Research, vol. 3, no. 11, pp. 2115–2120, 1997. pentapeptide of Nanog tryptophan repeat domain interacts [23] E. S. Pizer, F. D. Wood,H.S.Heine,F.E.Romantsev,G. with Nac1 and regulates stem cell proliferation but not R. Pasternack, and F. P. Kuhajda, “Inhibition of fatty acid pluripotency,” Journal of Biological Chemistry, vol. 284, no. 24, synthesis delays disease progression in a xenograft model of pp. 16071–16081, 2009. ovarian cancer,” Cancer Research, vol. 56, no. 6, pp. 1189–1193, [8] M.A.Stead,S.B.Carr, andS.C.Wright, “Structure of the 1996. human Nac1 POZ domain,” Acta Crystallographica Section F, [24] H. Orita, J. Coulter, E. Tully, F. P. Kuhajda, and E. Gabrielson, vol. 65, no. 5, pp. 445–449, 2009. “Inhibiting fatty acid synthase for chemoprevention of chem- [9] N. Jinawath, C. Vasoontara, K.-L. Yap, et al., “NAC-1, a ically induced lung tumors,” Clinical Cancer Research, vol. 14, potential stem cell pluripotency factor, contributes to pacli- no. 8, pp. 2458–2464, 2008. taxel resistance in ovarian cancer through inactivating Gadd45 [25] V. Jensen, M. Ladekarl, P. Holm-Nielsen, F. Melsen, and F. pathway,” Oncogene, vol. 28, no. 18, pp. 1941–1948, 2009. B. Soerensen, “The prognostic value of oncogenic antigen 519 (OA-519) expression and proliferative activity detected [10] K. Nakayama, N. Nakayama, T.-L. Wang, and I.-M. Shih, “NAC-1 controls cell growth and survival by repressing by antibody MIB-1 in node-negative breast cancer,” Journal of transcription of Gadd45GIP1, a candidate tumor suppressor,” Pathology, vol. 176, no. 4, pp. 343–352, 1995. Cancer Research, vol. 67, no. 17, pp. 8058–8064, 2007. [26] M. S. Shurbaji, J. H. Kalbfleisch, and T. S. Thurmond, [11] H. Liu, R. G. Sadygov, and J. R. Yates III, “A model for random “Immunohistochemical detection of a fatty acid synthase (OA- sampling and estimation of relative protein abundance in 519) as a predictor of progression of prostate cancer,” Human shotgun proteomics,” Analytical Chemistry, vol. 76, no. 14, pp. Pathology, vol. 27, no. 9, pp. 917–921, 1996. 4193–4201, 2004. [27] J. A. Menendez and R. Lupu, “Fatty acid synthase and the [12] G. Pohl, C.-L. Ho, R. J. Kurman, R. Bristow, T.-L. Wang, and I.- lipogenic phenotype in cancer pathogenesis,” Nature Reviews M. Shih, “Inactivation of the mitogen-activated protein kinase Cancer, vol. 7, no. 10, pp. 763–777, 2007. pathway as a potential target-based therapy in ovarian serous [28] W. Zhou, F. H. Wan, L. E. Landree, et al., “Fatty acid synthase tumors with KRAS or BRAF mutations,” Cancer Research, vol. inhibition activates AMP-activated protein kinase in SKOV3 65, no. 5, pp. 1994–2000, 2005. human ovarian cancer cells,” Cancer Research, vol. 67, no. 7, [13] D. K. Han, J. Eng, H. Zhou, and R. Aebersold, “Quantita- pp. 2964–2971, 2007. tive profiling of differentiation-induced microsomal proteins [29] D. Di Vizio, F. Sotgia, T. M. Williams, et al., “Caveolin-1 is using isotope-coded affinity tags and mass spectrometry,” required for the upregulation of fatty acid synthase (FASN), a Nature Biotechnology, vol. 19, no. 10, pp. 946–951, 2001. tumor promoter, during prostate cancer progression,” Cancer [14] A. Keller, A. I. Nesvizhskii, E. Kolker, and R. Aebersold, Biology and Therapy, vol. 6, no. 8, pp. 1263–1268, 2007. “Empirical statistical model to estimate the accuracy of [30] M.-Y.Lee,J.-S. Moon,S.W.Park, Y.-K. Koh, Y.-H.Ahn, peptide identifications made by MS/MS and database search,” and K.-S. Kim, “KLF5 enhances SREBP-1 action in androgen- Analytical Chemistry, vol. 74, no. 20, pp. 5383–5392, 2002. dependent induction of fatty acid synthase in prostate cancer [15] B. Davidson, A. Berner, C. G. Trope’, T.-L. Wang, and I.- cells,” Biochemical Journal, vol. 417, no. 1, pp. 313–322, 2009. M. Shih, “Expression and clinical role of the bric-a-brac [31] L. Korutla, P. Wang, T. G. Jackson, and S. A. Mackler, tramtrack broad complex/poxvirus and zinc protein NAC-1 in “NAC1, a POZ/BTB protein that functions as a corepressor,” ovarian carcinoma effusions,” Human Pathology, vol. 38, no. 7, Neurochemistry International, vol. 54, no. 3-4, pp. 245–252, pp. 1030–1036, 2007. 2009. [16] F. P. Kuhajda, “Fatty acid synthase and cancer: new application [32] L. Korutla, R. Degnan, P. Wang, and S. A. Mackler, “NAC1, a of an old pathway,” Cancer Research, vol. 66, no. 12, pp. 5977– cocaine-regulated POZ/BTB protein interacts with CoREST,” 5980, 2006. Journal of Neurochemistry, vol. 101, no. 3, pp. 611–618, 2007. [17] F. P. Kuhajda, K. Jenner, F. D. Wood, et al., “Fatty acid syn- [33] L. Korutla, P. J. Wang, and S. A. Mackler, “The POZ/BTB thesis: a potential selective target for antineoplastic therapy,” protein NAC1 interacts with two different histone deacetylases Proceedings of the National Academy of Sciences of the United in neuronal-like cultures,” Journal of Neurochemistry, vol. 94, States of America, vol. 91, no. 14, pp. 6379–6383, 1994. no. 3, pp. 786–793, 2005. [18] T. S. Gansler,W.Hardman III, D. A. Hunt,S.Schaffel, and R. [34] H. Liu, Y. Liu, and J.-T. Zhang, “A new mechanism of A. Hennigar, “Increased expression of fatty acid synthase (OA- drug resistance in breast cancer cells: fatty acid synthase 519) in ovarian neoplasms predicts shorter survival,” Human overexpression-mediated palmitate overproduction,” Molecu- Pathology, vol. 28, no. 6, pp. 686–692, 1997. lar Cancer Therapeutics, vol. 7, no. 2, pp. 263–270, 2008. [19] J. I. Epstein, M. Carmichael, and A. W. Partin, “OA-519 (fatty [35] A. Vazquez-Martin, S. Ropero, J. Brunet, R. Colomer, and acid synthase) as an independent predictor of pathologic stage J. A. Menendez, “Inhibition of fatty acid synthase (FASN) in adenocarcinoma of the prostate,” Urology,vol. 45, no.1,pp. synergistically enhances the efficacy of 5-fluorouracil in breast 81–86, 1995. carcinoma cells,” Oncology Reports, vol. 18, no. 4, pp. 973–980, [20] P. L. Alo’, P. Visca, A. Marci, A. Mangoni, C. Botti, and U. Di 2007. Tondo, “Expression of fatty acid synthase (FAS) as a predictor [36] J. Song, I.-M. Shih, R. Salani, D. W. Chan, and Z. Zhang, of recurrence in stage I breast carcinoma patients,” Cancer, vol. “Annexin XI is associated with cisplatin resistance and related 77, no. 3, pp. 474–482, 1996. to tumor recurrence in ovarian cancer patients,” Clinical [21] A. Rashid, E. S. Pizer, M. Moga, et al., “Elevated expression of Cancer Research, vol. 13, no. 22, pp. 6842–6849, 2007. fatty acid synthase and fatty acid synthetic activity in colorectal [37] I.-M. Shih, J. J.-C. Sheu, A. Santillan, et al., “Amplification neoplasia,” American Journal of Pathology, vol. 150, no. 1, pp. of a chromatin remodeling gene, Rsf-1/HBXAP, in ovarian 201–208, 1997. carcinoma,” Proceedings of the National Academy of Sciences of 12 Journal of Oncology the United States of America, vol. 102, no. 39, pp. 14004–14009, cancer,” Molecular Cancer Research, vol. 1, no. 10, pp. 707–715, 2005. 2003. [38] A. Sorrentino, C.-G. Liu, A. Addario, C. Peschle, G. Scambia, [53] M. Fiorentino,G.Zadra,E.Palescandolo,etal., “Overexpres- and C. Ferlini, “Role of microRNAs in drug-resistant ovarian sion of fatty acid synthase is associated with palmitoylation of cancer cells,” Gynecologic Oncology, vol. 111, no. 3, pp. 478– Wnt1 and cytoplasmic stabilization of β-catenin in prostate 486, 2008. cancer,” Laboratory Investigation, vol. 88, no. 12, pp. 1340– [39] D. Gupta, M. Venkatesh, H. Wang, et al., “Expanding the 1348, 2008. rolesfor pregnane Xreceptorincancer: proliferation anddrug [54] D. T. Coleman, R. Bigelow, and J. A. Cardelli, “Inhibi- resistance in ovarian cancer,” Clinical Cancer Research, vol. 14, tion of fatty acid synthase by luteolin post-transcriptionally no. 17, pp. 5332–5340, 2008. down-regulates c-Met expression independent of proteoso- [40] M. Ishibashi, K. Nakayama, S. Yeasmin, et al., “A BTB/POZ mal/lysosomal degradation,” Molecular Cancer Therapeutics, gene, NAC-1, a tumor recurrence-associated gene, as a vol. 8, no. 1, pp. 214–224, 2009. potential target fortaxol resistance in ovarian cancer,” Clinical [55] J. L. Little, F. B. Wheeler, C. Koumenis, and S. J. Kridel, Cancer Research, vol. 14, no. 10, pp. 3149–3155, 2008. “Disruption of crosstalk between the fatty acid synthesis and [41] J. Wang, J.-Y. Zhou, and G. S. Wu, “ERK-dependent MKP-1- proteasome pathways enhances unfolded protein response signaling and cell death,” Molecular Cancer Therapeutics, vol. mediated cisplatin resistance in human ovarian cancer cells,” Cancer Research, vol. 67, no. 24, pp. 11933–11941, 2007. 7, no. 12, pp. 3816–3824, 2008. [42] F. P. Kuhajda, L. E. Landree, and G. V. Ronnett, “The [56] E. Furuta, S. K. Pai, R. Zhan, et al., “Fatty acid synthase gene connections between C75 and obesity drug-target pathways,” is up-regulated by hypoxia via activation of Akt and sterol Trends in Pharmacological Sciences, vol. 26, no. 11, pp. 541– regulatory element binding protein-1,” Cancer Research, vol. 544, 2005. 68, no. 4, pp. 1003–1011, 2008. [43] E. S. Pizer, F. J. Chrest,J.A.DiGiuseppe, andW.F.Han, [57] A. L. Harris, “Hypoxia—a key regulatory factor in tumour “Pharmacological inhibitors of mammalian fatty acid synthase growth,” Nature Reviews Cancer, vol. 2, no. 1, pp. 38–47, 2002. suppress DNA replication and induce apoptosis in tumor cell lines,” Cancer Research, vol. 58, no. 20, pp. 4611–4615, 1998. [44] H. Q. Wang, D. A. Altomare, K. L. Skele, et al., “Positive feedback regulation between AKT activation and fatty acid synthase expression in ovarian carcinoma cells,” Oncogene, vol. 24, no. 22, pp. 3574–3582, 2005. [45] T. W. Grunt, R. Wagner, M. Grusch, et al., “Interaction between fatty acid synthase- and ErbB-systems in ovarian cancer cells,” Biochemical and Biophysical Research Communi- cations, vol. 385, no. 3, pp. 454–459, 2009. [46] S. Uddin, A. K. Siraj, M. Al-Rasheed, et al., “Fatty acid synthase and AKT pathway signaling in a subset of papillary thyroid cancers,” Journal of Clinical Endocrinology and Metabolism, vol. 93, no. 10, pp. 4088–4097, 2008. [47] J. Q. Cheng, A. K. Godwin, A. Bellacosa, et al., “AKT2, a putative oncogene encoding a member of a subfamily of protein- serine/threonine kinases, is amplified in human ovarian carcinomas,” Proceedings of the National Academy of Sciences of the United States of America, vol. 89, no. 19, pp. 9267–9271, 1992. [48] J.-N. Li, M. Gorospe, F. J. Chrest, et al., “Pharmacological inhibition of fatty acid synthase activity produces both cytostatic and cytotoxic effects modulated by p53,” Cancer Research, vol. 61, no. 4, pp. 1493–1499, 2001. [49] X. Lu, J. Errington, N. J. Curtin, J. Lunec, and D. R. Newell, “The impact of p53 status on cellular sensitivity to antifolate drugs,” Clinical Cancer Research, vol. 7, no. 7, pp. 2114–2123, [50] P. M. O’Connor, J. Jackman, I. Bae, et al., “Characterization of the p53 tumor suppressor pathway in cell lines of the National Cancer Institute anticancer drug screen and correlations with the growth-inhibitory potency of 123 anticancer agents,” Cancer Research, vol. 57, no. 19, pp. 4285–4300, 1997. [51] R. Salani, R. J. Kurman, R. Giuntoli II, et al., “Assessment of TP53 mutation using purified tissue samples of ovarian serous carcinomas reveals a higher mutation rate than previ- ously reported and does not correlate with drug resistance,” International Journal of Gynecological Cancer,vol. 18, no.3,pp. 487–491, 2008. [52] S. Rossi, E. Graner, P. Febbo, et al., “Fatty acid synthase expression defines distinct molecular signatures in prostate MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 International Journal of Journal of Immunology Research Endocrinology Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com BioMed PPAR Research Research International Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Research and Treatment Cellular Longevity Neurology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Oncology Unpaywall

Expression of Fatty Acid Synthase Depends on NAC1 and Is Associated with Recurrent Ovarian Serous Carcinomas

Download PDF
13 pages

Loading next page...
 
/lp/unpaywall/expression-of-fatty-acid-synthase-depends-on-nac1-and-is-associated-Wrmgm72YRs

References (57)

Publisher
Unpaywall
ISSN
1687-8450
DOI
10.1155/2010/285191
Publisher site
See Article on Publisher Site

Abstract

Hindawi Publishing Corporation Journal of Oncology Volume 2010, Article ID 285191, 12 pages doi:10.1155/2010/285191 Research Article Expression of Fatty Acid Synthase Depends on NAC1 and Is Associated with Recurrent Ovarian Serous Carcinomas 1, 2 3 4, 5 3 3 Stefanie M. Ueda, Kai Lee Yap, Ben Davidson, Yuan Tian, Vivek Murthy, 1, 2 2 3 1, 2 Tian-Li Wang, Kala Visvanathan, Francis P. Kuhajda, Robert E. Bristow, 3 1, 2, 3 Hui Zhang, and Ie-Ming Shih Department of Gynecology and Obstetrics, Johns Hopkins Medical Institutions, Baltimore, MD 21231, USA Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, MD 21231, USA Department of Pathology, Johns Hopkins Medical Institutions, Baltimore, MD 21231, USA Division of Pathology, Norwegian Radium Hospital, Rikshospitalet University Hospital, N-0310 Oslo, Norway Faculty Division Radiumhospitalet, The Medical Faculty, University of Oslo, N-0310 Oslo, Norway Correspondence should be addressed to Ie-Ming Shih, [email protected] Received 2 January 2010; Accepted 21 February 2010 Academic Editor: Ben Davidson Copyright © 2010 Stefanie M. Ueda et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Our previous reports demonstrated that NAC1, a BTB/POZ domain-containing nuclear protein, upregulates in recurrent ovarian serous carcinoma and participates in developing drug resistance in cancer cells. The current study applies quantitative proteomics to identify the proteins controlled by NAC1 by comparing the proteomes of SKOV3 cells with and without expression of a dominant negative NAC1 construct, N130. From the proteins that are downregulated by N130 (upregulated by NAC1), we chose to further characterize fatty acid synthase (FASN). Similar to change in protein level, the FASN transcript level in SKOV3 cells was significantly reduced by N130 induction or by NAC1 knockdown. Immunohistochemistry showed that NAC1 and FASN immunointensities in ovarian serous carcinoma tissues had a highly significant correlation (P< .0001). Moreover, we found that recurrent serous carcinomas exhibited higher FASN immunointensities than their matched primary tumors (P< .001). Multivariate analysis showed that an FASN staining score of >1 in serous carcinomas was associated with a worse overall survival time (P< .01). Finally, C93, a new FASN inhibitor, induced massive apoptosis in carboplatin/paclitaxel resistant ovarian cancer cells. In conclusion, we show that NAC1 is essential for FASN expression in ovarian serous carcinomas and the expression of FASN significantly correlates with tumor recurrence and disease aggressiveness. The dependence of drug resistant tumor cells on FASN suggests a potential application of FASN-based therapeutics for recurrent ovarian cancer patients. 1. Introduction studied ovarian cancer genome and transcriptome and have identified several genes and pathways that are potentially Ovarian cancer is a neoplastic disease that exemplifies involved in this phenotype. One of these genes, NACC1 many of the major issues underlying current chemotherapy encoding NAC1 (or NAC-1) protein, shows significantly regimens in clinical oncology [1, 2]. Although most ovarian higher expression in recurrent chemoresistant ovarian serous carcinomas at advanced stages are responsive to initial carcinomas than in primary untreated tumors [3]. NACC1 carboplatin and paclitaxel treatment, tumor clones resistant belongs to the BTB/POZ domain gene family and contains to these drugs eventually evolve as recurrent diseases. As the BEN domain that potentially mediates protein-DNA and a consequence, the main contributors to the mortality protein-protein interactions during chromatin organization and morbidity of advanced stage ovarian cancer patients and transcription [4]. Biologically, NAC1 has been demon- are chemoresistant tumors. In an effort to elucidate the strated to be an embryonic stem cell marker that controls molecular mechanisms underlying chemoresistance, we have proliferation and pluriopotency in embryonic stem cells 2 Journal of Oncology [5–7]. NAC1 homodimerizes through the highly conserved the same period were also analyzed. For Kaplan-Meier BTB/POZ domain (a.a. from 1–129) [8] and its complex survival analysis, we included 184 primary high-grade serous formation is essential for a variety of its biological functions. carcinoma patients who underwent optimal primary cytore- In our previous study, we found that recurrent ovarian ductive surgery followed by platinum-based chemotherapy serous carcinomas showed a higher NAC1 expression level at the Johns Hopkins Hospital. The paraffintissues were than their matched primary untreated tumors [3, 9]. Ectopic obtained from the surgical pathology repository at the Johns expression of NAC1 increased paclitaxel resistance while Hopkins Hospital and clinical data was obtained from medi- knockdown of NAC1 or disruption of NAC1 homodimer- cal records, including age, race, stage, histologic subtype, and ization sensitized cancer cells to chemotherapeutic drugs date of patient’s current status (alive or deceased). Median [3, 9]. In order to understand how NAC1 contributes to follow-up in those patients was 44.7 months. A subset of 56 drug resistance, we previously compared the gene expression high-grade serous carcinomas including 28 pairs of matched profiles of SKOV3 ovarian cancer cells to those of NAC1 primary and recurrent tumors was analyzed to correlate inactivated SKOV3 cells where the inactivation was induced FASN expression levels and to evaluate primary/recurrent by expression of N130, a mutant protein containing only tumor status. Another subset of 162 high-grade serous the BTB/POZ domain of NAC1 that competitively inhibits carcinomas was analyzed for correlation of NAC1 and FASN NAC1 homodimerization. We found that NAC1 negatively expression levels because the NAC1 immunostaining slides regulated the components of the Gadd45 tumor suppres- from these cases were available from our previous study sor pathway including Gadd45α and its binding protein, [3]. Collection of tissue samples was in accordance with the Gadd45gip1 [9, 10]. However, suppressing the Gadd45 guidelines of the institutional review board. tumor suppressor pathway did not completely rescue ovarian cancer cells after NAC1 inactivation, thus suggesting that 2.2. Quantitative Proteomics. Cell lysates were collected 48 other mechanisms also play a role. hours after induction (removal of doxycyclin) or mock In this study, we identified proteins that are potentially induction [3]. Protein concentration was measured by BCA regulated by NAC1 by applying a quantitative proteomic assay. The same amounts of proteins (1 mg) from each method using tandem mass spectrometry (MS/MS) and condition were reduced, alkylated, and digested with 1 to 50 spectral count [11]. A total of 2914 proteins were identified. Trypsin/protein ratio at 37 C overnight. The peptides were To reduce the sampling error and increase the quantification purified with C18 columns and resuspended in water with a accuracy, 208 proteins identified by at least 20 spectra and final concentration of 10 μg/μL. two unique peptides were quantified to identify candidate For protein identification and quantification, each pep- proteins controlled by N130 expression. By comparing the tide mixture was analyzed twice by the LTQ ion trap protein quantity in the proteomes between N130-induced mass spectrometer (Thermo Finnigan, San Jose, CA). For and noninduced SKOV3 cells, we identified new NAC1 each analysis, 2 μg of peptides were injected into a peptide regulated proteins that could be responsible for the NAC1- cartridge packed with C18 resin, and then passed through mediated drug resistance and for other biological functions. a10cm × 75 μm i.d. microcapillary HPLC (μLC) column Among the proteins found to be downregulated in N130- packed with C18 resin. A linear gradient of acetonitrile from induced cells, we selected fatty acid synthase (FASN) for 5%–32% over 100 minutes at flow rate of ∼300 nL/min was further characterization because it has been shown to be applied. During the LC-MS mode, data was acquired in the associated with tumor progression in a variety of human can- m/z range of 400 and 2000. The MS/MS was also turned on cers and its inhibitors are available for potential translation to collect CID using data dependent mode. studies. MS/MS spectra were searched with SEQUEST against the human IPI protein database (version 2.28). The peptide mass tolerance is 3.0 Da. Other parameters of database searching 2. Methods are as follows: cysteine modification (add cysteine 57) and oxidized methionine (add methionine with 16 Da). The 2.1. Cell Lines and Clinical Tissue Samples. High-grade output files were evaluated by INTERACT [13] and Peptide ovarian carcinoma cell lines, SKOV3, A2780, and OVCAR3, Prophet [14]. The identified peptides with a probability were obtained from the American Type Culture Collection score ≥ 0.9 were used for the spectral count. To determine (Rockville, MD), and a low-grade serous cell line, MPSC1, the number of MS/MS spectra used for identification of was previously established by us [12]. OSE10 was as SV- each proteinindifferent conditions using our in-house 40-immortalized ovarian surface epithelial cell line. All developed software tool and to reduce the error for protein cell lines were maintained in RPMI media with 5% heat- identification and quantification using MS/MS spectra, we inactivated fetal bovine serum (HyClone, Logan, UT) and only quantified proteins identified by at least 20 spectra 2% penicillin/streptomycin (Gibco, Rockville, MD). and 2 independent peptides from the N130-induced and- A total of 427 ovarian carcinoma tumors collected noninduced cells. between January 1990 and December 2006 were arranged in tissue microarrays. These included 269 high-grade serous, 45 low-grade serous, 45 endometrioid, and 68 clear cell 2.3. Western Blot and Real Time PCR. Western blot analysis carcinomas. Fifteen histologically normal ovarian tissues was performed on the protein lysates prepared from ovarian and 20 benign ovarian serous cystadenomas collected over cancer cell lines and OSE10. Similar amounts of total Journal of Oncology 3 proteinfromeachlysatewereseparated on 10% Tris-Glycine- 2.6. Lentivirus Production and Transduction. Two NAC1 SDS polyacrylamide gels (Novex, San Diego, CA) and then targeting shRNA plasmids previously cloned into the pLKO.1 electroblotted to Millipore Immobilon-P polyvinylidene (Sigma) and the packaging lentivirus plasmids were cotrans- difluoride membranes. The membranes were probed with an fected into 293FT cells using Lipofectamine 2000 (Invit- anti-FASN mouse monoclonal antibody (1 : 100) (FASgen, rogen. The sequences of NAC1 targeting shRNAs were Baltimore, MD) followed by a peroxidase-conjugated goat 5 -CCGGCCCAAGTGAGATTG CACATTTCTCGAGAAA- antimouse immunoglobulin (1 : 6,000). Western blots were TGTG CAATCTCACTTGGGTTTTTTG-3 (shRNA-A) and developed by chemiluminescence (Pierce, Rockford, IL) 5 -CCGG CAAGTACTACTGCC AGAACTTCTCGAG AA- utilizing glyceraldehyde-3-phosphate dehydrogenase as the GTTCTGGCAGTAGTACTTGTTTTTTG-3 (shRNA-C). loading control. To determine the mRNA levels of FASN, Five hours after incubation, the transfection reagents were we performed quantitative real-time PCR using a BioRad replaced with 10% FBS RPMI culture media supplemented iCycler. Total RNA was isolated with the TRIzol method with bovine serum albumin. The virus-containing super- (Invitrogen) and cDNA was synthesized from 2 to 5 μgtotal natant was collected, the mixture was centrifuged, and RNA. FASN primer sequences were 5 -CATCCAGATAGG- polybrene was added to a final concentration of 8 μg/mL. CCTCATAGAC-3 (forward) and 5 -CTCCATGAAGTA- Transduction was carried out by adding 0.25 mL of virus GGAGTGGAAG-3 (reverse). The expression of FASN was supernatant to the SKOV3 cells. A second transduction was normalized to that of human amyloid beta precursor protein performed 24 hours later using the same protocol. shRNA or based with threshold cycle numbers calculated from dupli- control (vector only) virus transduced cells were enriched cate measurements. Mean fold expression differences were by adding 2 μg/mL of puromycin. Following this second further normalized to those of ovarian surface epithelium, round of transduction, cells were collected at 48 and 72 hours OSE10. for quantitative real-time PCR analysis of NAC1 and FASN mRNA expression levels. 2.4. Immunohistochemistry. For immunohistochemistry, paraffinsectionsafter deparaffinization were incubated with 3. Results a primary antimouse FASN antibody at a dilution of 1 : 50 in a4 C moist chamber overnight. Negative controls included 3.1. High-Throughput Quantitative Proteomics Identifies benign serous cystadenomas and normal ovaries. Two FASN as an NAC1-Regulated Protein. To determine the pos- independent observers scored the FASN immunoreactivity sible mechanism by which NAC1 promotes drug resistance, using a categorical scoring system from 0 (not detectable) to we used the quantitative proteomics to compare the pro- 3 (intense) with the mean score recorded from triplicates. teomes of N130-induced and-noninduced SKOV3 ovarian Among the FASN stained cases, there were 162 that had been cancer cells using tandem mass spectrometry (MS/MS) and previously stained with an anti-NAC1 antibody [3, 15]. spectral count [11]. We were able to identify proteins that corresponded to a total of 2914 proteins. To reduce the sam- 2.5. Cell Number and Apoptosis Detection. Cancer cells were pling error and increase the quantification accuracy, 208 pro- seeded in 96-well plates at 2.5 × 10 cells/well for 24 hours teins identified by at least 20 spectra and two unique peptides and then were incubated with 100 μL of the FASN inhibitor, were quantified to identify candidate proteins controlled by C93, at concentrations ranging from 2.5 μg/mL to 50 μg/mL N130 expression. Based on the protein expression ratio of for 48 hours. After incubation, the number of viable N130 noninduced to N130 induced cells (N130 OFF versus cells was measured by the CellTiter-Blue assay (Promega, N130 ON), we listed the 19 proteins whose abundances Madison, WI). These numbers were plotted against FASN were at least increased 50% (OFF/ON ≥ 1.5, Table 1)and inhibitor concentrations, and the value of IC (i.e., the identified 18 proteins whose levels were at least increased C93 concentration at which cell growth dropped to 50% of 50% (ON/OFF ≥ 1.5) by N130 induction (Table 1). the control level) was estimated. To detect early apoptotic Cells with induced expression of N130 showed significantly cells, we grew ovariancancercelllinesin6wellplates(5 × higher levels NAC1 expression than N130-noninduced cells 10 cells/well) and treated them with C93 at their IC and (ON/OFF = 76.5), and all the identified peptides from with DMSO of equal concentration. The early apoptotic cells NAC1 matched the N130 (BTB/POZ) domain, indicating the were quantified utilizing the Annexin V-FITC detection kit robustness of N130 induction. Since our primary interest (Biovision, Mountain View, CA) and annexin V-stained cells was to identify the proteins that are upregulated by NAC1 were determined by flow cytometry. (i.e., downregulated by N130), we selected fatty acid synthase After treating cells with C93 or DMSO, cells were (FASN) for validation and characterization since its level trypsinized, washed, and resuspended in a solution contain- in noninduced cells is 1.57 times that in induced cells. We ing 0.6% NP-40, 3.7% formaldehyde, and 11 mg/mL Hoechst made this choice because, more than other proteins in the 33258 in a phosphate buffered solution. The stained cells list, FASN has been associated with tumor progression and were then quantified by the BD LSR cytometer (Becton because specific inhibitors of FASN are available for further Dickinson, Franklin Lakes, NJ). Cell cycle analysis was studies. The levels of FASN are less abundant in N130 performed using the CellQuest software (Becton Dickinson) induced cells than in N130-noninduced cells, suggesting and cells in the subG phase that represent the late phase of that NAC1 inactivation by N130 suppresses FASN protein apoptosis were also measured. expression in SKOV3 cells. 4 Journal of Oncology Table 1: Proteins that are differentially expressed in N130-induced and-noninduced SKOV3 cells. (a) Proteins downregulated by N130 induction ON- ON- OFF- OFF- Total IPI Protein name Off/on spectra peptides spectra peptides spectra Heterogeneous nuclear IPI00013881 4 5 20 7 24 4.62 ribonucleoprotein H Transcription intermediary IPI00012388 8 7 23 9 31 3.07 factor 1-beta IPI00220985 Keratin, type I cytoskeletal 18 8 4 18 5 26 2.35 RNA-binding protein regulatory IPI00298547 10 7 23 7 33 2.30 subunit IPI00299571 Protein disulfide isomerase A6 11 3 21 8 32 1.91 aldolase C, IPI00216976 10 3 19 4 29 1.90 fructose-bisphosphate Heterogeneous nuclear IPI00012074 9 5 17 7 26 1.89 ribonucleoprotein R IPI00013917 40S ribosomal protein S12 10 3 19 3 29 1.85 T-complex protein 1, beta IPI00297779 10 5 18 8 28 1.80 subunit IPI00027434 Transforming protein RhoC 8 5 14 5 22 1.75 IPI00141318 P63 protein 8 6 13 5 21 1.73 IPI00219217 lactate dehydrogenase B 11 6 18 8 28 1.67 IPI00219096 high-mobility group box 1 15 4 25 5 40 1.67 IPI00009236 Caveolin-1 9 2 15 2 24 1.67 Probable ATP-dependent RNA IPI00328343 11 6 18 6 29 1.64 helicase p47 T-complex protein 1, delta IPI00302927 8 3 13 6 21 1.63 subunit IPI00328188 fatty acid synthase 21 7 33 11 54 1.57 IPI00015027 AHNAK-related protein 11 7 17 11 27 1.55 Probable RNA-dependent IPI00017617 11 8 17 7 28 1.50 helicase p68 (b) Proteins upregulated by N130 induction ON- ON- OFF- OFF- Total IPI Protein name On/off spectra peptides spectra peptides spectra IPI00017454 Hypothetical protein FLJ13940 23 4 0 0 23 ? IPI00045207 NAC1 protein 153 9 2 1 155 76.50 IPI00395440 Unknown 26 5 5 4 31 5.13 IPI00107117 Peptidylprolyl isomerase B 14 5 6 3 20 2.33 T-complex protein 1, alpha IPI00290566 14 66320 2.33 subunit similar to Heterogeneous nuclear IPI00176692 27 5 12 6 39 2.25 ribonucleoprotein A1 IPI00221088 ribosomal protein S9 17 6 8 6 25 2.13 IPI00215918 ADP-ribosylation factor 4 17 7 9 7 26 1.89 IPI00015786 Spectrin alpha chain, brain 53 27 30 18 83 1.77 IPI00152412 Hypothetical protein 14 4 8 5 22 1.75 IPI00020984 Calnexin 29 10 17 7 46 1.71 Chloride intracellular channel IPI00010896 17 5 10 4 27 1.70 protein 1 T-complex protein 1, zeta IPI00027626 18 9 11 9 29 1.64 subunit IPI00216587 40S ribosomal protein S8 13 6 8 6 21 1.63 Journal of Oncology 5 (b) Continued. ON- ON- OFF- OFF- Total IPI Protein name On/off spectra peptides spectra peptides spectra IPI00216318 tyrosine 3-monooxygenase 40 10 25 9 65 1.58 60 kDa heat shock protein, IPI00329351 66 21 44 17 110 1.52 mitochondrial IPI00217468 H1 histone family, member 5 21 5 14 4 35 1.50 Malate dehydrogenase, IPI00291006 15 6 10 4 25 1.50 mitochondrial Table 2: Immunointensities of NAC1 and FASN in high-grade obtained from ovarian cancer ascites and tissues for NAC1 serous carcinoma. and FASN and correlated their immunointensities. The specificity of the anti-NAC1 has been demonstrated previ- FASN 2+/3+ FASN 0/1+ ously [3] and the specificity of the anti-FASN monoclonal NAC1 intense 51 34 antibody is shown in Figure 2(a). Western blot analysis NAC1 weak 19 58 shows a single protein band at the molecular mass of FASN Total (n)70 92 protein in all three ovarian cancer cell lines (OVCAR3, A2780, and SKOV3) but not in a low-grade serous carci- noma cell line, MPSC1, nor in ovarian surface epithelial Table 3: FASN immunointensity in primary and recurrent ovarian cells (OSE10). Analyzing 162 high-grade serous carcinomas serous carcinomas. showed a significant positive correlation in NAC1 and Score FASN immunointensities (P< .0001, Fisher Exact test) Tumor 0/1+ 2+/3+ (Table 2). These immunointensities for four representative Primary 18 10 high-grade serous carcinomas are shown in Figure 2(b).Our observations support the view that FASN expression is at Recurrent 4 24 least in part regulated by NAC1. Total (n)22 34 χ P< .0001. 3.3. FASN Expression in Ovarian Serous Tumors. To extend the above immunostaining findings, we investigated FASN 3.2. Validation of NAC1-Dependent FASN Expression. To immunoreactivity in various types of ovarian tumors. All validate the proteomic result above, we used quantitative benign cystadenomas and normal ovarian surface epithe- real-time PCR to determine FASN transcript levels in two lium (n = 35) display undetectable to very low FASN independent experimental systems. In our first system, staining (mean score = 0) whereas ovarian carcinomas of we used the same NAC1 dominant negative (N130) cell various histologic subtypes show FASN immunoreactivity model that was used in the proteomic analysis to assess (mean score ≥ 1) in most cases (Figures 2(c) and 2(d)). whether disruption of NAC1 homodimerization by N130 The FASN staining score of high-grade serous carcino- leads to a decrease in FASN mRNA level. We observed that mas is significantly higher than that of low-grade serous N130 mRNA, as detected by a PCR primer pair that amplifies carcinomas (P< .0001, t-test) and normal ovaries and the N130 region, increases 48 and 72 hours after induction benign cystadenomas (P< .0001). The FASN score in high- (Figure 1(a)). At these same time points, SKOV3 cells with grade serous carcinomas is marginally higher than that in N130 induction exhibit a downregulation of FASN mRNA. clear cell carcinoma (P = .024) and there is no statistical Of note, at 48 hours after N130 induction, the decrease significance of FASN score between high-grade serous and in FASN expression is similar between mRNA (46%) and endometrioid carcinoma (P = .099). Since NAC1 has protein (36%) levels. In our second system, we knocked been shown to be highly expressed in recurrent rather than down NAC1 in SKOV3 cells to determine the FASN levels. primary high-grade ovarian serous carcinomas, we expect Two NAC1 shRNAs (shRNA-A and -C) that target different that FASN expression levels will follow the same pattern. To coding regions of NACC1 were designed and packaged test this possibility, we assessed the FASN immunoreactivity into lentivirus. We found that both shRNAs effectively of matched primary and first recurrent tumors from the reduce NAC1 transcript levels (to less than 10% of control) same individuals. As expected, the FASN staining score is after transduction with shRNA lentivirus (Figure 1(b)). significantly higher in recurrent than in primary tumors Correspondingly, FASN expression levels also decrease as (P< .0001, paired t-test) (Figure 3(a)). Based on a compared to control. The above findings from cell culture 2 × 2 contingency table and chi-square analysis, we also systems suggest that NAC1 expression and its dimerization found that recurrent tumors exhibit a higher percentage of domain are essential for maintaining FASN expression in cases with intense FASN immunoreactivity than primary tumor cells. tumors do (Table 3). Figure 3(b) illustrates representa- In order to assess the biological significance of NAC1- tive pairs of primary and recurrent tumors with FASN dependent FASN expression, we stained tumor samples stain. 6 Journal of Oncology NAC1 (N130) FASN NAC1 FASN (a) (b) Figure 1: NAC1-dependent FASN expression in ovarian cancer cells. (a) N130 mRNA, as detected by a PCR primer pair that amplifies the N130 region, increases after N130 induction. In contrast, FASN mRNA decreases 48 and 72 hours after induction. (b) After shRNA lentivirus transduction,bothshRNAs(-A and-C) effectively reduce NAC1 transcript levels to less than 10% of control and reduce FASN expression. 3.4. Clinical Significance of FASN Expression in High-Grade for parental, paclitaxel, and carboplatin resistant SKOV3 Ovarian Serous Carcinomas. To evaluate the clinical signif- cells, respectively; 7.4 μg/mL, 7.5 μg/mL, and 8.4 μg/mL for icance of FASN expression, we correlated the expression parental, paclitaxel, and carboplatin resistant A2780 cells, levels in primary high-grade serous carcinoma tumors respectively; and 8.6 μg/mL, 6.5 μg/mL, and 8.8 μg/mL for with overall survival of the patient. We found that higher parental, paclitaxel, and carboplatin resistant OVCAR3 cells, FASN staining scores (scores > 1) correlated with worse respectively. When applied to each cell line at its IC overall survival (P< .002) (Figure 3(c)). Patients with concentration, C93 significantly increases the percentage of primary tumors that showed minimal or undetectable FASN annexin V stained cells (representing the early phase of immunoreactivity (score ≤ 1) had a median survival time apoptosis) and the sub-G fraction in cell cycle analyses of 60.4 months (range 1–193 months), whereas those with (representing the late phase) (Figure 4). Thus, C93 induces tumors that showed positive immunoreactivity (score > 1) apoptosis in all three ovarian cancer cell lines. The number of had a median survival time of 36.9 months (range 1–140 annexin V-stained cells increases in a time dependent fashion months, P< .01). After adjusting for age, stage, and race in all cancer cell lines including those resistant to carboplatin in a multivariate analysis, we found that, in patients with and paclitaxel (Figure 4(b)). Moreover, paclitaxel resistant primary serous carcinomas, FASN expression remains an cells are more sensitive to C93 than carboplatin resistant independent marker for prognosis with a hazard ratio of 1.87 cells, especially in the A2780 and SKOV3 cell lines. In all (95% CI: 1.12–3.11, P = .02). three cell lines, the C93-treated group shows a significantly higher percentage of sub-G cells than the DMSO-treated group (P< .01). These sub-G cells can be detected as 3.5. C93 Suppresses Growth of Paclitaxel-Resistant and Car- early as 12 hours after C93 treatment and become more boplatin-Resistant Ovarian Cancer Cells. The above findings pronounced by 48 hours (Figure 4(c)). C93 fails to affect cell suggest that FASN preferentially expresses in recurrent and cycle progression in SKOV3 and OVCAR3 cells by 96 hours in most aggressive types of ovarian serous carcinomas, after treatment but it arrests A2780 cells in the G phase as raising the possibility that FASN expression contributes to early as 24 hours after treatment (data not shown). this phenotype. Thus, to determine if FASN expression is essential for cell growth and survival of high-grade serous carcinoma cells (including those that are resistant 4. Discussion to paclitaxel or carboplatin), we applied C93, a second generation FASN inhibitor, to ovarian cancer cell lines One of the major challenges facing ovarian cancer patients is (including SKOV3, A2780, and OVCAR3). We first used the the development of chemoresistant tumors after cytoreduc- cell number counted by CellTiter Blue assay to determine tion surgery and chemotherapy. In this study, we provide new the IC of C93 for each ovarian cancer cell line. The evidence that homodimerization of NAC1, a drug resistance- IC of C93 was 7.4 μg/mL, 7.4 μg/mL, and 8.7 μg/mL associated nuclear protein, is essential for maintaining FASN mRNA level (% of non-induced control) 48h-cont 48h-N130 72h-cont 72h-N130 mRNA levels (% of pLKO control) 48h-pLKO 48h-shNAC1-A 48h-shNAC1-C 72h-pLKO 72h-shNAC1-A 72h-shNAC1-C Journal of Oncology 7 FASN Tumor A Tumor B Tumor C Tumor D 3+ 0 3+ 1+ NAC1 3+ 0 3+ 1+ GAPDH FASN (a) (b) Normal ovary serous carcinoma (solid tumor) 3.5 2.5 Serous cystadenoma Serous carcinoma 1.5 (ascites) 0.5 (c) (d) Figure 2: Expression of NAC1 and FASN in ovarian tumor tissues. (a) Western blot analysis shows a single protein band that corresponds to FASN protein in all three ovarian cancer cell lines (OVCAR3, A2780, and SKOV3) but not in a low-grade serous carcinoma cell line (MPSC1) nor in ovarian surface epithelial cells (OSE10). (b) Immunoreactivity of NAC1 (upper panels) and FASN (bottom panels) in four representative high-grade serous carcinomas. For each staining, the immunostaining scores are shown in the upper left corner (c)-(d). FASN expression in different types of ovarian carcinomas. (c) FASN immunointensity scores for different types of ovarian carcinoma including high-grade serous, low-grade serous, clear cell, and endometrioid carcinomas. The mean and standard deviation of the immunostaining score are shown. (d) Representative tumor sections of FASN immunostaining are illustrated. FASN immunoreactivity is detectable in high- grade serous carcinoma cells but not in normal ovarian surface epithelium nor in cystadenomas. expression at both protein and mRNA levels in ovarian Mammalian FASN is a ∼260 kD cytoplasmic enzyme that cancer cells. We also demonstrate that FASN expression is responsible for all the steps of de novo fatty acid synthesis. It is highly correlated with the status of recurrent chemore- catalyzes the NADPH-dependent condensation of malonyl- sistant ovarian serous carcinomas and is independently CoA and acetyl-CoA to palmitate [16]. Importantly, normal correlated with poor overall survival. Suppressing FASN adult tissues express minimal amounts of FASN due to enzyme activity with its inhibitor induces apoptosis in cancer the presence of abundant dietary lipids. In contrast, a cells that are resistant to paclitaxel and carboplatin. Thus, variety of tumors including breast, colon, ovary, and prostate molecular studies that illuminate the fundamental properties cancer express elevated FASN levels [17–23], since tumor of chemoresistance should provide new therapeutic targets cells become less sensitive to regulatory nutritional signals for treating recurrent ovarian cancers. and prefer the de novo lipogenesis pathway. Furthermore, Mean FAS IHC score OSE10 High-grade serous MPSC1 OVCAR3 Low-grade serous A2780 Clear cell SKOV3 Endometrioid Cyst and OSE 8 Journal of Oncology Primary Recurrent 4 P< .0001 Patient Patient Primary Recurrent (a) (b) P< .01 0 50 100 150 200 Survival (months) FASH IHC score ≤ 1 FASH IHC score > 1 (c) Figure 3: Clinical significance of FASN expression in high-grade ovarian serous carcinoma. (a) Comparison of FASN immunointensity scores for 28 pairs of matched primary and recurrent tumors from the same patients. Recurrent tumors exhibit elevated expression when compared to primary specimens from the same patient (P< .0001). (b) FASN staining of the primary and recurrent tumors from two representative patients. (c) Kaplan-Meier curve analysis shows that patients whose tumors exhibit higher FASN immunostaining scores have significantly shorter survival times than those whose tumors show undetectable or very low FASN immunoreactivity (60.4 versus 36.9 months, P< .01). FASN expression level highly correlates with the clinical KLF5/SREBP-1 function upstream of FASN [29, 30]. Thus, aggressiveness of tumors [17, 18, 24–27]. at least in ovarian cancer cells, the NAC1 pathway represents Although previous reports have shown the role of FASN another mechanism for controlling FASN expression. Unlike expression in ovarian cancer [18, 28], the current study other members of the BTB/POZ family, NAC1 lacks the provides new findings that should have several biological zinc finger DNA-binding domain. Rather, it has been and clinical implications. First, we use in vitro cell culture reported to act as a transcription corepressor with other studies using NAC1 targeting shRNAs and the dominant BTB/POZ proteins [31]. Moreover, it has also been shown negative NAC1 (N130) approach to demonstrate that FASN to interact with nuclear proteins potentially involved in is one of the proteins regulated at least in part by NAC1. tumorigenesis, including Nanog [7], CoREST [32], and This NAC1-dependent FASN expression in protein level HDAC3 and HDAC4 [33]. Thus, it is possible that FASN is further supported by its mRNA level and the positive expression is indirectly controlled by NAC1 through binding correlation between NAC1 and FASN immunointensities with its specific partner(s). Identification of the NAC1-FASN in ovarian serous carcinoma samples. The mechanism pathway sheds new light on the molecular mechanism by underlying upregulation of FASN in human cancer is not which NAC1 promotes tumor progression. Further studies clear and it likely involves multiple pathways. Previous are required to elucidate the transcriptional regulation of studies have shown that, in prostate cancer, caveolin-1 and FASN by NAC1. FASN IHC score Patients (%) Journal of Oncology 9 SKOV3 80 A2780 OVCAR3 50 40 20 20 0 0 0 0 8 18 24 36 48 0 8 18 24 36 48 0 8 18 24 36 48 Hours of incubation Hours of incubation Hours of incubation DMSO DMSO DMSO C93 C93 C93 (a) 50 60 10 10 0 0 0 8 18 24 36 48 0 8 18 24 36 48 0 8 18 24 36 48 Hours of incubation Hours of incubation Hours of incubation TR TR TR TR CR TR SKOV3 C93 SKOV3 DMSO A2780 C93 A2780 DMSO OVCAR3 C93 OVCAR3 DMSO CR CR CR CR TR CR SKOV3 C93 SKOV3 DMSO A2780 C93 A2780 DMSO OVCAR3 C93 OVCAR3 DMSO (b) SKOV3 OVCAR3 A2780 DMSO C93 (c) Figure 4: C93 treatment promotes apoptosis in ovarian cancer cells including cells that are resistant to carboplatin and paclitaxel. (a) In three ovarian cancer cell lines, C93 at its IC concentration increases the percentage of annexin V stained cells in a time dependent fashion. DMSO was used as the vehicle control. (b) C93 also increases annexin V stained cells in carboplatin resistant (CR) and paclitaxel resistant (TR) cell lines. (c) In all three cell lines, C93 treatment leads to an increase in sub-G cells when compared to the DMSO-treated group (P< .001). The percentage of sub-G cells is measured 48 hours after C93 treatment. The second finding in the current study involves the palmitic acid as a result of FASN overexpression [34]. Despite positive correlation between FASN expression and recur- recent advances in discovering drug-resistance biomarkers rence status in ovarian serous carcinoma tissue. It has been that are associated with recurrent and chemoresistant ovar- shown that ectopic overexpression of FASN results in drug ian cancer [1, 3, 15, 36–41], reversing drug resistance by resistance and that reducing the FASN expression increased targeting these markers remains a tantalizing objective. The the drug sensitivity in breast cancer cell lines [34]. At low problem lies in the lack of reagents that can be used to inhibit concentrations, FASN inhibitor also sensitized tumor cells these genes in clinical studies. Thus, therapeutic targeting with FASN overexpression to chemotherapeutic agents [34, of FASN provides an attractive option since selective small 35]. The FASN-mediated drug resistance appears to be due compound inhibitors for this protein are available. The first to a decrease in drug-induced apoptosis due to abundant generation of FASN inhibitors, including C75 and Orlistat, Annexin V stained cells (%) Annexin V stained cells (%) Annexin V stained cells (%) Annexin V stained cells (%) Sub-G cells (%) Annexin V stained cells (%) Annexin V stained cells (%) 10 Journal of Oncology potently inhibited tumor growth in a mouse xenograft model [3, 15], hence suggesting that the NAC1-FASN pathway con- but the adverse effects associated with these drugs prevented stitutes one of the mechanisms that propels ovarian cancer their further consideration for clinical applications [23]. On progression. How FASN contributes to disease aggressiveness the other hand, C93, the second generation FASN inhibitor in ovarian cancer remains speculative. Besides endowing used in this study, pharmacologically eliminates concomitant drug resistance, FASN may enhance oncogenesis via cellular CPT-1 stimulation and does not induce the anorexia and mechanisms such as enhancing the Wnt [53], c-Met [54], feeding behavior changes in mice that were caused by earlier and proteosome pathways [55]. Moreover, upregulation of generation FASN inhibitors [28, 42]. These features are FASN gives cancer cells a growth and survival advantage by important for any FASN inhibitor before it can be considered blocking apoptosis under hypoxia, a common condition in for clinical testing. Our in vitro studies demonstrate that solid tumors and tumor effusions [56, 57]. C93 affects both carboplatin and paclitaxel resistant ovarian In summary, this study identified candidate proteins cancer cells. Thus, ovarian cancer cells that overexpress controlled by NAC1 and provides new evidence that FASN FASN are molecularly dependent on it for cell survival. This expression is at least in part regulated by NAC1. Thus the observation is significant because FASN inhibitors provide NAC1-FASN pathway may represent a new mechanism for an alternative treatment for ovarian cancer patients who tumor progression that creates ovarian tumor cells that have developed recurrent tumors after initial paclitaxel and are resistant to chemotherapy. Our findings also indicate carboplatin treatment. that FASN is a novel biomarker for recurrent ovarian The antitumor effects of FASN inhibitors, like C93, serous carcinoma and its enzyme activity is essential for are thought to result from depletion of end product fatty the survival of chemoresistant tumor cells. New generation acids with accumulation of toxic intracellular malonyl-CoA FASN inhibitors, like C93, deserve consideration in future and altered production of phospholipids with diminished clinical trials involving advanced ovarian serous carcinomas, membrane synthesis [23, 43]. Alternatively, FASN inhibitors particularly those that are refractory to paclitaxel and plat- may suppress tumors through metabolism-independent inum drugs. Further studies will be required to delineate the mechanisms. For example, FASN inhibition has been shown biological and translational roles of FASN in drug resistance to selectively activate AMP-activated protein kinase (AMPK) in ovarian and perhaps other types of cancers. in ovarian cancer cells causing cytotoxicity while sparing most normal human tissues from these pleiotropic effects of Abbreviations AMPK activation [28]. Moreover, a positive feedback regu- lation has been reported in ovarian carcinoma cells between FASN: Fatty acid synthase. AKT activation and FASN expression [44]. Phosphorylated NAC1: Nucleus accumbens associated 1. AKT significantly correlates with FASN expression and FASN inhibition by either C75 or cerulenin downregulates phosphorylated AKT [44–46]. Therefore, FASN inhibitors Acknowledgment may contribute to antioncogenesis by suppressing tumor- promoting signaling pathways such as AKT, a pathway that This study is in memory of Ms. Sean Patrick, founder is frequently activated in ovarian serous carcinoma [47]. of the HERA Women’s Cancer Foundation, who coura- geously fought against recurrent ovarian cancer. The work Thus, the C93-induced apoptosis in ovarian cancer cells may be related to FASN inactivation and/or suppression of is supported by NIH/NCI RO1CA103937 (IMS), the HERA AKT activity. It has been demonstrated that FASN inhibition Women’s Cancer Foundation, and the Entertainment Indus- initiates apoptosis more effectively in neoplastic cells with try Foundation (SMU). mutant TP53 than in those with wild-type TP53 [27, 48]. Our current finding supports this view as the A2780 cell line References that harbors wild-type TP53 [49] responds to C93 with cell cycle arrest in the G phase in addition to apoptosis, while [1] K. R. Cho and I. M. Shih, “Ovarian cancer,” Annual Review of both the SKOV3 and OVCAR3 cell lines with their deleted Pathology: Mechanisms of Disease, vol. 4, pp. 287–313, 2009. and mutated TP53, respectively, respond to C93 with massive [2] S. B. Kaye, “Reversal of drug resistance in ovarian cancer: apoptosis [50]. Since mutated TP53 appears in the majority wheredowegofromhere?” Journal of Clinical Oncology, vol. of ovarian high-grade serous carcinomas [1, 51], it is likely 26, no. 16, pp. 2616–2618, 2008. that apoptosis is the predominant antitumor mechanism of [3] K. Nakayama, N. Nakayama, B. Davidson, et al., “A BTB/POZ FASN inhibition. protein, NAC-1, is related to tumor recurrence and is essential As with other epithelial cancers of the breast, colon, for tumor growth and survival,” Proceedings of the National and prostate [20, 21, 52], FASN overexpression in ovar- Academy of Sciences of the United States of America, vol. 103, ian cancer, as this study shows, appears to be associated no. 49, pp. 18739–18744, 2006. with the most malignant type, that is, high-grade serous [4] S. Abhiman, L. M. Iyer, and L. Aravind, “BEN: a novel domain carcinoma. More importantly, in patients with this cancer, in chromatin factors and DNA viral proteins,” Bioinformatics, the level of FASN expression significantly correlates with vol. 24, no. 4, pp. 458–461, 2008. worse clinical outcome, supporting the view that FASN [5] J. Kim, J. Chu, X. Shen, J. Wang, and S. H. Orkin, “An extended expression contributes to disease aggressiveness in cancer transcriptional network for pluripotency of embryonic stem cells. A similar observation has been reported for NAC1 cells,” Cell, vol. 132, no. 6, pp. 1049–1061, 2008. Journal of Oncology 11 [6] J. Wang, S. Rao, J. Chu, et al., “A protein interaction network [22] L. Z. Milgraum,L.A.Witters,G.R.Pasternack, andF.P. for pluripotency of embryonic stem cells,” Nature, vol. 444, no. Kuhajda, “Enzymes of the fatty acid synthesis pathway are 7117, pp. 364–368, 2006. highly expressed in in situ breast carcinoma,” Clinical Cancer [7] T. Ma, Z. Wang, Y. Guo, and D. Pei, “The C-terminal Research, vol. 3, no. 11, pp. 2115–2120, 1997. pentapeptide of Nanog tryptophan repeat domain interacts [23] E. S. Pizer, F. D. Wood,H.S.Heine,F.E.Romantsev,G. with Nac1 and regulates stem cell proliferation but not R. Pasternack, and F. P. Kuhajda, “Inhibition of fatty acid pluripotency,” Journal of Biological Chemistry, vol. 284, no. 24, synthesis delays disease progression in a xenograft model of pp. 16071–16081, 2009. ovarian cancer,” Cancer Research, vol. 56, no. 6, pp. 1189–1193, [8] M.A.Stead,S.B.Carr, andS.C.Wright, “Structure of the 1996. human Nac1 POZ domain,” Acta Crystallographica Section F, [24] H. Orita, J. Coulter, E. Tully, F. P. Kuhajda, and E. Gabrielson, vol. 65, no. 5, pp. 445–449, 2009. “Inhibiting fatty acid synthase for chemoprevention of chem- [9] N. Jinawath, C. Vasoontara, K.-L. Yap, et al., “NAC-1, a ically induced lung tumors,” Clinical Cancer Research, vol. 14, potential stem cell pluripotency factor, contributes to pacli- no. 8, pp. 2458–2464, 2008. taxel resistance in ovarian cancer through inactivating Gadd45 [25] V. Jensen, M. Ladekarl, P. Holm-Nielsen, F. Melsen, and F. pathway,” Oncogene, vol. 28, no. 18, pp. 1941–1948, 2009. B. Soerensen, “The prognostic value of oncogenic antigen 519 (OA-519) expression and proliferative activity detected [10] K. Nakayama, N. Nakayama, T.-L. Wang, and I.-M. Shih, “NAC-1 controls cell growth and survival by repressing by antibody MIB-1 in node-negative breast cancer,” Journal of transcription of Gadd45GIP1, a candidate tumor suppressor,” Pathology, vol. 176, no. 4, pp. 343–352, 1995. Cancer Research, vol. 67, no. 17, pp. 8058–8064, 2007. [26] M. S. Shurbaji, J. H. Kalbfleisch, and T. S. Thurmond, [11] H. Liu, R. G. Sadygov, and J. R. Yates III, “A model for random “Immunohistochemical detection of a fatty acid synthase (OA- sampling and estimation of relative protein abundance in 519) as a predictor of progression of prostate cancer,” Human shotgun proteomics,” Analytical Chemistry, vol. 76, no. 14, pp. Pathology, vol. 27, no. 9, pp. 917–921, 1996. 4193–4201, 2004. [27] J. A. Menendez and R. Lupu, “Fatty acid synthase and the [12] G. Pohl, C.-L. Ho, R. J. Kurman, R. Bristow, T.-L. Wang, and I.- lipogenic phenotype in cancer pathogenesis,” Nature Reviews M. Shih, “Inactivation of the mitogen-activated protein kinase Cancer, vol. 7, no. 10, pp. 763–777, 2007. pathway as a potential target-based therapy in ovarian serous [28] W. Zhou, F. H. Wan, L. E. Landree, et al., “Fatty acid synthase tumors with KRAS or BRAF mutations,” Cancer Research, vol. inhibition activates AMP-activated protein kinase in SKOV3 65, no. 5, pp. 1994–2000, 2005. human ovarian cancer cells,” Cancer Research, vol. 67, no. 7, [13] D. K. Han, J. Eng, H. Zhou, and R. Aebersold, “Quantita- pp. 2964–2971, 2007. tive profiling of differentiation-induced microsomal proteins [29] D. Di Vizio, F. Sotgia, T. M. Williams, et al., “Caveolin-1 is using isotope-coded affinity tags and mass spectrometry,” required for the upregulation of fatty acid synthase (FASN), a Nature Biotechnology, vol. 19, no. 10, pp. 946–951, 2001. tumor promoter, during prostate cancer progression,” Cancer [14] A. Keller, A. I. Nesvizhskii, E. Kolker, and R. Aebersold, Biology and Therapy, vol. 6, no. 8, pp. 1263–1268, 2007. “Empirical statistical model to estimate the accuracy of [30] M.-Y.Lee,J.-S. Moon,S.W.Park, Y.-K. Koh, Y.-H.Ahn, peptide identifications made by MS/MS and database search,” and K.-S. Kim, “KLF5 enhances SREBP-1 action in androgen- Analytical Chemistry, vol. 74, no. 20, pp. 5383–5392, 2002. dependent induction of fatty acid synthase in prostate cancer [15] B. Davidson, A. Berner, C. G. Trope’, T.-L. Wang, and I.- cells,” Biochemical Journal, vol. 417, no. 1, pp. 313–322, 2009. M. Shih, “Expression and clinical role of the bric-a-brac [31] L. Korutla, P. Wang, T. G. Jackson, and S. A. Mackler, tramtrack broad complex/poxvirus and zinc protein NAC-1 in “NAC1, a POZ/BTB protein that functions as a corepressor,” ovarian carcinoma effusions,” Human Pathology, vol. 38, no. 7, Neurochemistry International, vol. 54, no. 3-4, pp. 245–252, pp. 1030–1036, 2007. 2009. [16] F. P. Kuhajda, “Fatty acid synthase and cancer: new application [32] L. Korutla, R. Degnan, P. Wang, and S. A. Mackler, “NAC1, a of an old pathway,” Cancer Research, vol. 66, no. 12, pp. 5977– cocaine-regulated POZ/BTB protein interacts with CoREST,” 5980, 2006. Journal of Neurochemistry, vol. 101, no. 3, pp. 611–618, 2007. [17] F. P. Kuhajda, K. Jenner, F. D. Wood, et al., “Fatty acid syn- [33] L. Korutla, P. J. Wang, and S. A. Mackler, “The POZ/BTB thesis: a potential selective target for antineoplastic therapy,” protein NAC1 interacts with two different histone deacetylases Proceedings of the National Academy of Sciences of the United in neuronal-like cultures,” Journal of Neurochemistry, vol. 94, States of America, vol. 91, no. 14, pp. 6379–6383, 1994. no. 3, pp. 786–793, 2005. [18] T. S. Gansler,W.Hardman III, D. A. Hunt,S.Schaffel, and R. [34] H. Liu, Y. Liu, and J.-T. Zhang, “A new mechanism of A. Hennigar, “Increased expression of fatty acid synthase (OA- drug resistance in breast cancer cells: fatty acid synthase 519) in ovarian neoplasms predicts shorter survival,” Human overexpression-mediated palmitate overproduction,” Molecu- Pathology, vol. 28, no. 6, pp. 686–692, 1997. lar Cancer Therapeutics, vol. 7, no. 2, pp. 263–270, 2008. [19] J. I. Epstein, M. Carmichael, and A. W. Partin, “OA-519 (fatty [35] A. Vazquez-Martin, S. Ropero, J. Brunet, R. Colomer, and acid synthase) as an independent predictor of pathologic stage J. A. Menendez, “Inhibition of fatty acid synthase (FASN) in adenocarcinoma of the prostate,” Urology,vol. 45, no.1,pp. synergistically enhances the efficacy of 5-fluorouracil in breast 81–86, 1995. carcinoma cells,” Oncology Reports, vol. 18, no. 4, pp. 973–980, [20] P. L. Alo’, P. Visca, A. Marci, A. Mangoni, C. Botti, and U. Di 2007. Tondo, “Expression of fatty acid synthase (FAS) as a predictor [36] J. Song, I.-M. Shih, R. Salani, D. W. Chan, and Z. Zhang, of recurrence in stage I breast carcinoma patients,” Cancer, vol. “Annexin XI is associated with cisplatin resistance and related 77, no. 3, pp. 474–482, 1996. to tumor recurrence in ovarian cancer patients,” Clinical [21] A. Rashid, E. S. Pizer, M. Moga, et al., “Elevated expression of Cancer Research, vol. 13, no. 22, pp. 6842–6849, 2007. fatty acid synthase and fatty acid synthetic activity in colorectal [37] I.-M. Shih, J. J.-C. Sheu, A. Santillan, et al., “Amplification neoplasia,” American Journal of Pathology, vol. 150, no. 1, pp. of a chromatin remodeling gene, Rsf-1/HBXAP, in ovarian 201–208, 1997. carcinoma,” Proceedings of the National Academy of Sciences of 12 Journal of Oncology the United States of America, vol. 102, no. 39, pp. 14004–14009, cancer,” Molecular Cancer Research, vol. 1, no. 10, pp. 707–715, 2005. 2003. [38] A. Sorrentino, C.-G. Liu, A. Addario, C. Peschle, G. Scambia, [53] M. Fiorentino,G.Zadra,E.Palescandolo,etal., “Overexpres- and C. Ferlini, “Role of microRNAs in drug-resistant ovarian sion of fatty acid synthase is associated with palmitoylation of cancer cells,” Gynecologic Oncology, vol. 111, no. 3, pp. 478– Wnt1 and cytoplasmic stabilization of β-catenin in prostate 486, 2008. cancer,” Laboratory Investigation, vol. 88, no. 12, pp. 1340– [39] D. Gupta, M. Venkatesh, H. Wang, et al., “Expanding the 1348, 2008. rolesfor pregnane Xreceptorincancer: proliferation anddrug [54] D. T. Coleman, R. Bigelow, and J. A. Cardelli, “Inhibi- resistance in ovarian cancer,” Clinical Cancer Research, vol. 14, tion of fatty acid synthase by luteolin post-transcriptionally no. 17, pp. 5332–5340, 2008. down-regulates c-Met expression independent of proteoso- [40] M. Ishibashi, K. Nakayama, S. Yeasmin, et al., “A BTB/POZ mal/lysosomal degradation,” Molecular Cancer Therapeutics, gene, NAC-1, a tumor recurrence-associated gene, as a vol. 8, no. 1, pp. 214–224, 2009. potential target fortaxol resistance in ovarian cancer,” Clinical [55] J. L. Little, F. B. Wheeler, C. Koumenis, and S. J. Kridel, Cancer Research, vol. 14, no. 10, pp. 3149–3155, 2008. “Disruption of crosstalk between the fatty acid synthesis and [41] J. Wang, J.-Y. Zhou, and G. S. Wu, “ERK-dependent MKP-1- proteasome pathways enhances unfolded protein response signaling and cell death,” Molecular Cancer Therapeutics, vol. mediated cisplatin resistance in human ovarian cancer cells,” Cancer Research, vol. 67, no. 24, pp. 11933–11941, 2007. 7, no. 12, pp. 3816–3824, 2008. [42] F. P. Kuhajda, L. E. Landree, and G. V. Ronnett, “The [56] E. Furuta, S. K. Pai, R. Zhan, et al., “Fatty acid synthase gene connections between C75 and obesity drug-target pathways,” is up-regulated by hypoxia via activation of Akt and sterol Trends in Pharmacological Sciences, vol. 26, no. 11, pp. 541– regulatory element binding protein-1,” Cancer Research, vol. 544, 2005. 68, no. 4, pp. 1003–1011, 2008. [43] E. S. Pizer, F. J. Chrest,J.A.DiGiuseppe, andW.F.Han, [57] A. L. Harris, “Hypoxia—a key regulatory factor in tumour “Pharmacological inhibitors of mammalian fatty acid synthase growth,” Nature Reviews Cancer, vol. 2, no. 1, pp. 38–47, 2002. suppress DNA replication and induce apoptosis in tumor cell lines,” Cancer Research, vol. 58, no. 20, pp. 4611–4615, 1998. [44] H. Q. Wang, D. A. Altomare, K. L. Skele, et al., “Positive feedback regulation between AKT activation and fatty acid synthase expression in ovarian carcinoma cells,” Oncogene, vol. 24, no. 22, pp. 3574–3582, 2005. [45] T. W. Grunt, R. Wagner, M. Grusch, et al., “Interaction between fatty acid synthase- and ErbB-systems in ovarian cancer cells,” Biochemical and Biophysical Research Communi- cations, vol. 385, no. 3, pp. 454–459, 2009. [46] S. Uddin, A. K. Siraj, M. Al-Rasheed, et al., “Fatty acid synthase and AKT pathway signaling in a subset of papillary thyroid cancers,” Journal of Clinical Endocrinology and Metabolism, vol. 93, no. 10, pp. 4088–4097, 2008. [47] J. Q. Cheng, A. K. Godwin, A. Bellacosa, et al., “AKT2, a putative oncogene encoding a member of a subfamily of protein- serine/threonine kinases, is amplified in human ovarian carcinomas,” Proceedings of the National Academy of Sciences of the United States of America, vol. 89, no. 19, pp. 9267–9271, 1992. [48] J.-N. Li, M. Gorospe, F. J. Chrest, et al., “Pharmacological inhibition of fatty acid synthase activity produces both cytostatic and cytotoxic effects modulated by p53,” Cancer Research, vol. 61, no. 4, pp. 1493–1499, 2001. [49] X. Lu, J. Errington, N. J. Curtin, J. Lunec, and D. R. Newell, “The impact of p53 status on cellular sensitivity to antifolate drugs,” Clinical Cancer Research, vol. 7, no. 7, pp. 2114–2123, [50] P. M. O’Connor, J. Jackman, I. Bae, et al., “Characterization of the p53 tumor suppressor pathway in cell lines of the National Cancer Institute anticancer drug screen and correlations with the growth-inhibitory potency of 123 anticancer agents,” Cancer Research, vol. 57, no. 19, pp. 4285–4300, 1997. [51] R. Salani, R. J. Kurman, R. Giuntoli II, et al., “Assessment of TP53 mutation using purified tissue samples of ovarian serous carcinomas reveals a higher mutation rate than previ- ously reported and does not correlate with drug resistance,” International Journal of Gynecological Cancer,vol. 18, no.3,pp. 487–491, 2008. [52] S. Rossi, E. Graner, P. Febbo, et al., “Fatty acid synthase expression defines distinct molecular signatures in prostate MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 International Journal of Journal of Immunology Research Endocrinology Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com BioMed PPAR Research Research International Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Research and Treatment Cellular Longevity Neurology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal

Journal of Oncology – Unpaywall

Published: Jan 1, 2010

There are no references for this article.