Transformation with Oncogenic Ras and the Simian Virus 40 T Antigens Induces Caspase-Dependent Sensitivity to Fatty Acid Biosynthetic Inhibition

ABSTRACT

Oncogenesis is frequently accompanied by the activation of specific metabolic pathways. One such pathway is fatty acid biosynthesis, whose induction is observed upon transformation of a wide variety of cell types. Here, we explored how defined oncogenic alleles, specifically the simian virus 40 (SV40) T antigens and oncogenic Ras^12V, affect fatty acid metabolism. Our results indicate that SV40/Ras^12V-mediated transformation of fibroblasts induces fatty acid biosynthesis in the absence of significant changes in the concentration of fatty acid biosynthetic enzymes. This oncogene-induced activation of fatty acid biosynthesis was found to be mammalian target of rapamycin (mTOR) dependent, as it was attenuated by rapamycin treatment. Furthermore, SV40/Ras^12V-mediated transformation induced sensitivity to treatment with fatty acid biosynthetic inhibitors. Pharmaceutical inhibition of acetyl-coenzyme A (CoA) carboxylase (ACC), a key fatty acid biosynthetic enzyme, induced caspase-dependent cell death in oncogene-transduced cells. In contrast, isogenic nontransformed cells were resistant to fatty acid biosynthetic inhibition. This oncogene-induced sensitivity to fatty acid biosynthetic inhibition was independent of the cells' growth rates and could be attenuated by supplementing the medium with unsaturated fatty acids. Both the activation of fatty acid biosynthesis and the sensitivity to fatty acid biosynthetic inhibition could be conveyed to nontransformed breast epithelial cells through transduction with oncogenic Ras^12V. Similar to what was observed in the transformed fibroblasts, the Ras^12V-induced sensitivity to fatty acid biosynthetic inhibition was independent of the proliferative status and could be attenuated by supplementing the medium with unsaturated fatty acids. Combined, our results indicate that specific oncogenic alleles can directly confer sensitivity to inhibitors of fatty acid biosynthesis.

IMPORTANCE Viral oncoproteins and cellular mutations drive the transformation of normal cells to the cancerous state. These oncogenic alterations induce metabolic changes and dependencies that can be targeted to kill cancerous cells. Here, we find that the cellular transformation resulting from combined expression of the SV40 early region with an oncogenic Ras allele is sufficient to induce cellular susceptibility to fatty acid biosynthetic inhibition. Inhibition of fatty acid biosynthesis in these cells resulted in programmed cell death, which could be rescued by supplementing the medium with nonsaturated fatty acids. Similar results were observed with the expression of oncogenic Ras in nontransformed breast epithelial cells. Combined, our results suggest that specific oncogenic alleles induce metabolic dependencies that can be exploited to selectively kill cancerous cells.

INTRODUCTION

Cancerous cells frequently exhibit substantial metabolic differences from the tissues that they were derived from. These changes are widely shared and independent of their tissue of origin, highlighting a common cancer cell metabolic program. This program includes activation of glycolysis, i.e., the Warburg effect, induction of nucleotide biosynthesis, and activation of fatty acid biosynthesis (1–3). The oncogenic activation of fatty acid biosynthesis has been known for decades and has been observed in a number of different cancer types, including carcinomas of the liver, breast, and colon (4–6). Furthermore, the activation of fatty acid biosynthesis has since been shown to be critical for tumorigenesis (7, 8), although many questions still remain about both the mechanisms of oncogenic fatty acid biosynthetic activation and their contribution to malignancy.

In addition to the activation of several metabolic pathways, cancerous cells are more susceptible to certain metabolic insults. Cancerous cells become particularly dependent on glucose and glutamine; limitation of these nutrients results in enhanced cancer cell death in comparison to the cell death in normal tissues (9, 10). Similarly, a variety of cancerous cell types undergo cell death upon fatty acid biosynthetic inhibition (11–13). The specific mechanisms that govern cancer cell-specific sensitivity to metabolic challenge are of particular interest as they represent therapeutic targets which could be exploited to more specifically induce cancer cell death. However, confounding the elucidation of these mechanisms has been the lack of comparisons between cancerous cells and nontransformed isogenic cells. The lack of these comparisons in conjunction with the genetic complexity of tumor-derived tissue has prevented the elucidation of the oncogenic events that drive cancer cell sensitivity to metabolic insults.

Here, we utilized a well-described stepwise model of transformation (14) to explore how specific oncogenic alleles affect fatty acid biosynthesis and the sensitivity to fatty acid biosynthetic inhibition. The model consists of parental primary human fibroblasts, telomerase-expressing human fibroblasts, simian virus 40 (SV40) T antigen-immortalized human fibroblasts, and T antigen-Ras^12V-transformed human fibroblasts (14). The SV40 large T antigens (LT) have well-described oncogenic activities, such as inactivation of the Rb and p53 tumor suppressors (15). The SV40 T antigens, in conjunction with Ras^12V expression, are sufficient to induce oncogenic transformation in human fibroblasts (14). The specific mechanisms through which these alleles contribute to oncogenesis and the potential of associated metabolic vulnerabilities are significant given the ubiquity of activating Ras mutations in cancer. Study of the SV40 T antigens has long provided insights into cellular transformation, but the recent findings that a polyomavirus family member of SV40, Merkel cell polyomavirus (MCV), causes Merkel cell carcinoma (MCC) has refocused attention on the important transforming capabilities of polyomavirus gene products (16, 17).

Here, we find that joint expression of the SV40 T antigens and oncogenic Ras^12V substantially elevates fatty acid biosynthesis in comparison to the levels in isogenic primary and immortalized fibroblasts expressing just the SV40 T antigens. Furthermore, the expression of Ras^12V induced sensitivity to inhibition of fatty acid biosynthesis. Fatty acid biosynthetic inhibition resulted in caspase-mediated cell death in the Ras-transformed cells but not in isogenic nontransformed cells. In the Ras-transformed cells, cell death could be rescued by medium supplemented with unsaturated fatty acids. Similarly, inhibition of fatty acid biosynthesis resulted in increased cell death in a basal-like human breast cancer cell line in comparison to the level of cell death in a nontransformed human breast epithelial cell line. Finally, the expression of Ras^12V activated fatty acid biosynthesis and induced sensitivity to fatty acid biosynthetic inhibition in nontransformed human breast epithelial cells. Combined, these results highlight that common oncogenic alleles can drive sensitivity to fatty acid biosynthetic inhibition and provide insight into the pathways responsible for this sensitivity.

MATERIALS AND METHODS

Cell culture.

Ras-transformed BJ fibroblasts (14), telomerase-SV40 T antigen immortalized BJ fibroblasts (14), and primary BJ fibroblasts (ATCC CRL-2522) were courteously provided by Robert Weinberg (Whitehead Institute for Biomedical Research). Primary BJ fibroblasts and their immortalized and transformed derivatives, as well as MDA-MB-231 cells (ATCC CRM-HTB-26), were maintained in Dulbecco modified Eagle medium (DMEM) (Gibco) supplemented with 10% fetal bovine serum and 4.5 g/liter glucose. MCF 10A (ATCC CRL-10317) cells were grown in DMEM–F-12 supplemented with 5% heat-inactivated horse serum, 10 μg/ml insulin, 100 ng/ml cholera toxin, 500 ng/ml hydrocortisone, and 20 ng/ml epidermal growth factor. For all pharmaceutical treatments, cells were grown to equivalent confluence prior to drug administration.

Cloning.

Human telomerase (hTERT) cDNA was amplified by PCR from pWZL-Blast-Flag-HA-hTERT (plasmid number 22396; Addgene) using a pair of Gibson assembly primers (forward, GGAACCAATTCAGTCGACTGGGATCCCGTCCTGCTGCGCACGTG, and reverse, TTTGTACAAGAAAGCTGGGTTCTAGATCAGTCCAGGATGGTCTTGAAGTCTG).

To make pLenti CMV/TO/Hygro/hTERT, hTERT cDNA was cloned via Gibson assembly into pLenti CMV/TO Hygro (plasmid number 17484; Addgene) empty vector doubly digested with BamHI and XbaI.

Lentiviral transduction.

293T cells were seeded at 2 × 10⁶ cells per 10-cm dish and grown for 12 to 24 h. To generate pseudotyped lentivirus, 293T cells were transfected with 2.6 μg lentiviral vector, 0.25 μg vesicular stomatitis virus G protein (VSV-G), and 2.4 μg PAX2 using Fugene 6 (Promega). Twenty-four hours later, the medium was changed and replaced with 4 ml fresh medium, which was collected after a further 24 h and filtered through a 0.45-μm filter. MCF 10A cells were seeded at 1 × 10⁶ cells per 10-cm dish 1 day before transduction. MCF 10A cells were transduced with lentivirus in the presence of 5 μg/ml Polybrene for overnight incubation. The next day, the lentivirus-containing medium was removed and replaced with fresh DMEM–F-12 medium. Seventy-two hours after transduction, MCF 10A cells were selected with 0.7 μg/ml puromycin and 10 μg/ml blasticidin for 3 to 4 days. MCF 10A cells were infected with four different combinations of the two lentiviral vectors, as follows: (i) pLenti-puro-HRas G12V (plasmid number 22262; Addgene) and pLenti-blad-P53 R175H (plasmid number 22262; Addgene), (ii) pLenti-puro-HRas G12V and pLenti-blad-empty (vector control), (iii) pLenti-puro-empty (vector control) and pLenti-blad-P53 R175H, and (iv) pLenti-puro-empty and pLenti-blad-empty. To create pLenti-puro-empty, pLenti-puro-HRas G12V was doubly digested with BamH I and XbaI, and the large fragment was blunted and ligated. pLenti-blad-empty was provided by Hartmut Land, University of Rochester. Primary fibroblasts transduced with pLenti CMV/TO/Hygro/hTERT were selected in 200 μg/ml hygromycin B (Invitrogen) for 1 week. The expression of telomerase in these cells was confirmed by quantitative PCR (qPCR).

Pharmaceuticals.

5-Tetradecyloxy-2-furoic acid (TOFA) (Cayman) was maintained as a 5 mg ml⁻¹ (15.5 mM) stock in dimethyl sulfoxide (DMSO). Rapamycin (Calbiochem) was maintained as a 250 μM stock in DMSO. Z-Val-Ala-Asp-fluoromethylketone (z-VAD-fmk) (Enzo Life Sciences) was maintained as a 5 mM stock in DMSO. Z-Asp-Glu-Val-Asp-phenylnitroaniline (Z-DEVD-pNA) (Enzo) was maintained as a 10 mM stock in DMSO. Fatty acids were first complexed to fatty acid-free bovine serum albumin (BSA; Sigma) as previously described (12). Briefly, 1 volume of 5 mM fatty acid (Sigma) in ethanol was mixed with 4 volumes of 4% BSA in 0.9% NaCl and incubated at 37°C for 1 h to obtain a 1 mM BSA–fatty acid stock solution.

Protein analysis.

For protein normalization, cells were washed with PBS, scraped, and solubilized in lysis buffer containing 150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 1 mM EDTA, 1% NP-40, 0.5% deoxycholate, 0.1% SDS, 10 μg/ml phenylmethylsulfonyl fluoride (PMSF), and 1 μg/ml pepstatin. After 10 min on ice, the lysates were sonicated and centrifuged at 14,000 × g for 5 min to pellet insoluble material. The protein concentration of supernatant fluid was determined by Bradford assay (Bio-Rad).

Immunoblot analysis.

Cells were washed with PBS, scraped, and solubilized in disruption buffer containing 50 mM Tris (pH 7.0), 2% SDS, 5% 2-mercaptoethanol, and 2.75% sucrose. The resulting extracts were sonicated, boiled for 5 min, and centrifuged at 14,000 × g for 5 min to pellet insoluble material. The extracts were then subjected to electrophoresis in an 8 to 10% SDS polyacrylamide gel and transferred to a nitrocellulose sheet. The blots were then stained with Ponceau S to ensure equivalent protein loading and transfer, blocked by incubation in 5% milk in TBST (50 mM Tris-HCl, pH 7.6, 150 mM NaCl, 0.1% Tween 20), and reacted with primary and, subsequently, secondary antibodies. Protein bands were visualized using an enhanced chemiluminescence (ECL) system (Pierce or Bio-Rad) and exposure to film (Kodak) or by using the Molecular Imager Gel Doc XR+ system (Bio-Rad). For protein band quantifications, the Molecular Imager Gel Doc was used and band intensities were integrated by using the Image Lab software. For quantifications, a protein extract dilution series was utilized to ensure the linearity of antibody responses. In certain instances, the brightness and contrast of the resulting images were adjusted in Microsoft PowerPoint to improve image clarity. All such processing was applied uniformly to the entire image. The antibodies used were specific for acetyl-coenzyme A (CoA) carboxylase (ACC; Cell Signaling Technology), fatty acid synthase (FAS; Santa Cruz Biotechnology, Inc.), ATP-citrate lyase (ACL; Cell Signaling Technology), poly(ADP-ribose) polymerase-1 (PARP-1; Santa Cruz Biotechnology, Inc.), α-tubulin (Epitomics or Cell Signaling Technology), glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Cell Signaling Technology), p53 (Santa Cruz Biotechnology, Inc.), Harvey rat sarcoma viral oncogene (HRas; Santa Cruz Biotechnology, Inc.), and SV40 LT (Santa Cruz Biotechnology, Inc.).

Viability/cytotoxicity assays.

Viability and proliferation were analyzed via the trypan blue dye exclusion assay or the tetrazolium dye MTT [3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide] colorimetric assay. Cells were seeded in 12-well plates. For the trypan blue dye exclusion assay, cells were washed with phosphate-buffered saline (PBS) and trypsinized. Following trypsinization, an aliquot of the cell suspension was mixed 1:1 with 0.4% trypan blue solution and counted using a TC10 automated cell counter, following the manufacturer's instructions (Bio-Rad). For the MTT assay, cells were incubated with 0.5 mg/ml MTT for 2 to 4 h at 37°C. The medium was then aspirated and 1 ml 0.04 M HCl in isopropanol was added to each well of a 12-well plate. The plate was then placed on a shaker for 15 min to fully dissolve the formazan, and the absorbance at 570 nm was read with background subtraction at 650 nm.

RNA interference assay.

The following small interfering RNA (siRNA) duplexes were obtained from Dharmacon Research. For ACC1 targeting, 5′ CAAUGGCAUUGCAGCAGUG 3′ and its complementary sequence were utilized. For a non-protein-targeting control, 5′ CGUAAGCGACAUACUUACAUU 3′ and its complement were utilized. Cells were plated in 12-well plates. All cell lines were transfected at approximately 30% confluence with 150 nM siRNA duplex using Oligofectamine (Invitrogen) according to the manufacturer's instructions.

Caspase assays.

Cellular extracts were assayed for caspase activity using a tetrapeptide conjugated to phenylnitroaniline (Z-DEVD-pNA). Cells were plated in 60-mm dishes. At 24 h posttreatment, the cells were scraped in their medium, washed twice with PBS, resuspended in 150 μl of lysis buffer containing 50 mM HEPES (pH 7.4), 0.1% CHAPS {3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate}, 1 mM dithiothreitol, 0.1 mM EDTA, and left on ice for 5 min. The lysates were then centrifuged at 10,000 × g for 10 min at 4°C, and the supernatant tested for DEVDase activity as recommended by the manufacturer (Enzo Life Sciences). After 2 h, the release of chromophore was determined with a spectrophotometer by measuring the absorbance at 405 nm.

Measurement of fatty acid biosynthesis.

Fatty acid biosynthesis was evaluated as described previously (18). Briefly, fatty acid biosynthesis in all cell lines was assayed by measuring the incorporation of [1-¹⁴C]acetic acid (NEN Radiochemicals) into cellular lipids. Cells were plated in 12-well plates. Five hundred microliters of DMEM containing 1 μCi of 56.2 mCi mmol⁻¹ [1-¹⁴C]acetic acid was added to each well and incubated at 37°C for 2 h. The cells were then washed with PBS and dissolved in 0.5 ml of 0.2 M KOH. Samples were then saponified by the addition of 0.3 ml of 50% KOH and 2.0 ml of ethanol, followed by a 2-h incubation at 75°C and then an overnight incubation at room temperature. After saponification, the samples were extracted two times with 4.5 ml of hexane. The hexane-containing fractions were discarded, and the remaining fractions containing fatty acids were acidified to a pH of <2 by the addition of 0.5 ml 12 M HCl. After acidification, the samples were extracted one time with 4.5 ml hexane. The organic fractions were dried under N₂, resuspended in 50 μl of 1:1 chloroform-hexane, dissolved in 1 ml of Ecoscint O liquid scintillation fluid, and assessed for radioactivity using a Beckman LS6500 liquid scintillation counter.

LC-MS/MS methodology.

Cells were plated in 10-cm dishes. At 48 h posttreatment, the medium was aspirated and 80:20 methanol-water (80% methanol) at −80°C was immediately added to quench metabolic activity and extract metabolites. Cells were then incubated at −80°C for 10 min. Following cell quenching, cells were scraped in the dish and kept on dry ice, and the resulting cell suspension vortexed, centrifuged at 3,000 × g for 5 min, and reextracted twice more with 80% methanol at −80°C. After pooling the three extractions, the samples were completely dried under N₂ gas, dissolved in 175 μl 50% methanol, and spun at 13,000 × g for 5 min to remove debris. The liquid chromatography-tandem mass spectrometry (LC-MS/MS) parameters for measurement of malonyl-CoA were as previously described (19), using a malonyl-CoA-specific multiple reaction monitoring (MRM) scan of 852.2 to 808.2 in negative mode with a collision cell energy of 24 eV.

RESULTS

Ras-mediated transformation induces fatty acid biosynthesis.

While the activation of fatty acid biosynthesis (diagrammed in Fig. 1A) is a common characteristic of numerous cancer cell types (20), the mechanisms responsible for the induction are not well understood. We set out to determine whether the expression of specific oncogenes drives fatty acid biosynthetic activation in nontransformed cells. We examined the effects of oncogenic SV40 T antigen and Ras^12V expression in a previously defined stepwise model of transformation, consisting of primary parental BJ fibroblasts subsequently immortalized with telomerase, alone or together with the SV40 early region, and fully transformed with Ras^12V expression (Fig. 1B and C) (14). Ras-transformed fibroblasts exhibited a greater than 2-fold increase in the rate of de novo fatty acid biosynthesis in comparison to the rate in the SV40-immortalized fibroblasts and greater than 10-fold increases in comparison to the rates in telomerase-expressing fibroblasts or primary fibroblasts (Fig. 1C). This induction of fatty acid biosynthesis was not strongly correlated to cellular growth rates, as both the Ras-transformed and the SV40-immortalized fibroblasts grew at similar rates (Fig. 1D). Furthermore, the primary fibroblasts exhibited a maximal growth rate that approximated that of the SV40-immortalized and transformed cells, although their proliferation slowed over time in comparison to the proliferation of those cells, consistent with an increased sensitivity to contact inhibition (Fig. 1D). To determine whether this reduced proliferation rate of the primary cells at high confluence affected the rate of fatty acid biosynthesis, we measured de novo fatty acid biosynthesis in the primary cells at both ∼50% and ∼95% confluence. As shown by the results in Fig. 1E, the confluence of the primary cells did not affect their fatty acid biosynthetic activity. Taken together, these results suggest that stepwise transformation and oncogenic Ras expression can drive fatty acid biosynthetic activation in nontransformed cells and that this activation is not a strict function of increased growth rate.

FIG 1.

Ras-transformed human fibroblasts exhibit elevated levels of fatty acid biosynthesis. (A) Schematic representation of the fatty acid biosynthetic pathway. Abbreviations are as follows: PDH, pyruvate dehydrogenase; MDH, malate dehydrogenase; ME, malic enzyme; ACL, ATP-citrate lyase; ACC, acetyl-CoA carboxylase; FAS, fatty acid synthase. (B) The abundances of SV40 LT, HRas, and α-tubulin in the different cell lines were measured by Western blotting analyses as indicated. Primary, primary human BJ fibroblasts; Primary-ht, telomerase-expressing BJ fibroblasts; Immort., SV40 T antigen-immortalized BJ fibroblasts; Ras Trans., T antigen-Ras^12V-transformed BJ fibroblasts. (C) Cells were labeled with ¹⁴C-acetate for 2 h. Cellular lipids were saponified, extracted, and scintillation counted. All values are normalized to protein concentration. Values are means + standard errors of the means (SEM). Asterisks indicate significant differences from values for Ras-transformed cells (***, P < 0.001). (D) Cell numbers were quantified at the indicated time points after plating. Values are means + SEM. (E) Fatty acid biosynthesis in primary cells at 50% or 95% confluence was measured as described for panel C. (F) The abundances of ACL, ACC, FAS, SCD1, and α-tubulin were measured in lysates from ∼50% or ∼95% confluent cells. Western blotting data are from a representative experiment (n ≥ 3). (G) After 48 h of rapamycin or DMSO treatment, the abundances of ACC, FAS, phosphorylated p70 S6K (P-p70 S6K), p70 S6K, GAPDH, and α-tubulin were measured by Western blotting analyses. (H) After 48 h of rapamycin or DMSO treatment, fatty acid biosynthesis was measured as described for panel C. All values are normalized to the protein concentration. Values are means + SEM. Asterisks indicate significant differences between values for the control and drug-treated cells for each cell line (**, P < 0.01; ****, P < 0.0001).

Numerous types of cancerous tissues induce the expression of lipogenic metabolic enzymes (reviewed in reference 20). The cell growth regulatory kinase mammalian target of rapamycin (mTOR), whose activity is induced by Ras-driven pathways (21–23), has been shown to induce transcriptional activation of lipogenic enzymes (24, 25). We therefore chose to compare the expression of fatty acid biosynthetic enzymes in the Ras-transformed, SV40-immortalized, telomerase-expressing, and primary cells and examine how the levels of these enzymes respond to mTOR inhibition. At ∼50% confluence, the protein levels of ATP-citrate lyase (ACL), acetyl-CoA carboxylase (ACC), fatty acid synthase (FAS), and stearoyl-CoA desaturase 1 (SCD1) were elevated in the Ras-transformed, SV40-immortalized, and telomerase-expressing cells compared to the levels in the primary cells (Fig. 1F). At higher confluence, the protein levels of SCD1 were further induced in the Ras-transformed and SV40-immortalized cells. However, there was little difference in the levels of expression of ACL, ACC, FAS, and SCD1 in the Ras-transformed and SV40-immortalized cells at either 50% or 95% confluence (Fig. 1F). These results indicate that the SV40 early region and, to a lesser extent, telomerase induce the expression of fatty acid biosynthetic enzymes. These results also suggest that the observed 2-fold increase in the rate of de novo fatty acid biosynthesis in Ras-transformed fibroblasts compared to the rate in the SV40-immortalized fibroblasts is independent of lipogenic enzyme expression.

Treatment with rapamycin, an mTORC1 inhibitor, had little impact on the expression of ACC in any of the cells examined but did reduce the protein levels of FAS in the Ras-transformed and SV40-immortalized cells (Fig. 1G). As expected, treatment with rapamycin inhibited the phosphorylation of the mTORC1 substrate p70 S6 kinase (S6K) (Fig. 1G). Consistent with its negative impact on FAS expression, rapamycin treatment reduced fatty acid biosynthesis in Ras-transformed and SV40-immortalized cells (Fig. 1H). Rapamycin also decreased fatty acid biosynthesis in the primary cells, albeit to a lesser extent, without a notable decrease in the levels of fatty acid biosynthetic enzymes (Fig. 1H). These results are consistent with mTORC1's role in fatty acid biosynthetic regulation and also suggest that the induction of fatty acid biosynthesis observed in Ras-transformed cells is not mediated solely by increased mTORC1 activity, as the rapamycin-treated, Ras-transformed cells still exhibit enhanced fatty acid biosynthetic activity compared to the levels in the rapamycin-treated, SV40-immortalized, and primary cells.

The expression of oncogenic Ras induces sensitivity to fatty acid biosynthetic inhibition.

Tumor-derived cell lines have been found to be sensitive to fatty acid biosynthetic inhibition (11–13). However, the mechanisms responsible for this sensitivity are unclear. Given that oncogenic Ras expression induced fatty acid biosynthetic activation, we sought to determine whether oncogenic Ras expression induces sensitivity to treatment with fatty acid biosynthetic inhibitors. Cells were treated with TOFA, a pharmaceutical inhibitor of acetyl-CoA carboxylase (ACC), which catalyzes the carboxylation of acetyl-CoA to form malonyl-CoA, one of the rate-limiting reactions of fatty acid biosynthesis. Treatment with TOFA attenuated fatty acid biosynthesis in the Ras-transformed, SV40-immortalized, and primary cells (Fig. 2A). Furthermore, TOFA treatment dramatically reduced the cellular levels of malonyl-CoA (Fig. 2B). By analyzing cellular toxicity via the trypan blue dye exclusion assay (Fig. 2C) and MTT assay (Fig. 2D), we found that TOFA treatment resulted in a strong dose-dependent decrease in the proliferation of Ras-transformed cells and, to lesser extents, in SV40-immortalized cells and telomerase-expressing cells, whereas the primary fibroblasts were not affected. Additionally, analysis of the rates of cell death indicated that TOFA treatment induced substantial dose-dependent cell death in the Ras-transformed fibroblasts. The SV40-immortalized, telomerase-expressing, and primary fibroblasts were affected but to a much lesser extent (Fig. 2E). These results indicate that oncogenic Ras substantially sensitizes human fibroblasts to pharmacological inhibition of fatty acid biosynthesis, leading to decreased cellular viability and the induction of cell death. Furthermore, this sensitivity to fatty acid biosynthetic inhibition did not directly correlate with the differences in proliferation between these cells (Fig. 1D), suggesting that the growth rate is not responsible for oncogenically increased sensitivity to fatty acid biosynthetic inhibition.

FIG 2.

Expression of oncogenic Ras induces sensitivity to fatty acid biosynthetic inhibition. (A) After 48 h of treatment with TOFA or DMSO, cells were labeled with ¹⁴C-acetate for 2 h. Cellular lipids were extracted, saponified, and scintillation counted. All values are normalized to protein concentration. Values are means + SEM. Differences between values for the control and drug-treated cells for each cell line are significant (P < 0.0001). (B) After 48 h of treatment with TOFA, Ras-transformed cells were processed for LC-MS/MS and the abundance of malonyl-CoA relative to the level in untreated cells was measured as described in Materials and Methods. All values are normalized to protein concentration. Values are means + SEM. (C) After 72 h of TOFA treatment, cells were collected and stained with trypan blue. Viable cells that excluded trypan blue were counted, and the results are presented as the percentages of the value for the DMSO control. Values are means + SEM. (D) After 48 h of TOFA treatment, an MTT assay was performed. Cell viability is represented as the percentage of the value for the DMSO control. (E) Cells were treated as described for panel C, and the percentages of dead cells were calculated after trypan blue staining. Values are means + SEM. (C to E) Asterisks indicate significant differences between values for Ras-transformed cells and other cell types at each TOFA concentration (*, P < 0.05; **, P < 0.01; ***, P < 0.001), and hash marks indicate significant differences between values for primary cells and other cell types at each TOFA concentration (#, P < 0.05; ##, P < 0.01; ###, P < 0.001).

To investigate the mechanism of cell death induced by TOFA treatment, we analyzed cleavage of PARP-1, a nuclear polymerase that is cleaved by caspases upon apoptotic induction. In Ras-transformed fibroblasts, PARP-1 was cleaved upon TOFA treatment (Fig. 3A). This TOFA-induced PARP-1 cleavage could be blocked through treatment with a broad-range caspase inhibitor, z-VAD-fmk (Fig. 3A), indicating that the cleavage was caspase dependent. In contrast, PARP-1 cleavage was not observed in TOFA-treated SV40-immortalized or primary fibroblasts, consistent with a lack of caspase activation and cell death observed upon TOFA treatment in these cells. To extend these observations, we measured caspase activity using a chromogenic caspase substrate. As shown by the results in Fig. 3B, TOFA treatment induced caspase activity in the Ras-transformed fibroblasts by ∼5-fold in comparison to the activity in DMSO-treated controls. Under the same conditions, SV40-immortalized and primary cell lysates exhibited substantially less caspase activity (Fig. 3B). Primary fibroblasts had a higher level of endogenous caspase activity than Ras-transformed or SV40-immortalized fibroblasts, but TOFA treatment had no effect on this activity (Fig. 3B). Similar to the PARP-1 cleavage results, z-VAD-fmk treatment completely ablated TOFA-induced caspase activity in all three cell lines, confirming that caspase-specific cleavage was the reason for the measurements obtained.

FIG 3.

Ras-induced cell death in response to TOFA treatment is mediated through a caspase-dependent mechanism. (A) After 24 h of treatment with DMSO, TOFA, or z-VAD-fmk, the abundances of PARP-1 and α-tubulin were measured by Western blotting analyses. Data shown are from a representative experiment. (B) After 24 h of treatment with DMSO, TOFA, or z-VAD-fmk, cells were harvested and assayed for caspase activity as described in Materials and Methods. All values are normalized to the protein concentration. Values are means + SEM (n = 2). (C) At 24, 48, and 72 h posttreatment with DMSO, TOFA, or z-VAD-fmk, cells were collected and stained with trypan blue. Viable cells that excluded trypan blue were counted, and the results are presented as the percentages of the value for the DMSO control. Values are means + SEM. The asterisk indicates a significant difference between the values for TOFA-treated and TOFA + z-VAD-fmk-treated cells at that time point (*, P < 0.05). (D) The percentages of dead cells were calculated after staining the cells with trypan blue. Values are means + SEM. (B and D) Asterisks indicate significant differences between bracketed values (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Since z-VAD-fmk completely blocked both TOFA-induced PARP-1 cleavage and caspase activity, we wanted to determine whether TOFA-induced cell death is caspase dependent. Treatment with z-VAD-fmk maintained cellular viability at control levels during the first 24 h of treatment (Fig. 3C). After 48 h of treatment, the caspase inhibitor increased the viability of TOFA-treated cells but not to the level of non-TOFA-treated controls (Fig. 3C). Treatment with z-VAD-fmk substantially attenuated cell death after 24 and 48 h of drug treatment and markedly reduced the percentage of dead cells at 72 h after drug treatment (Fig. 3D). Taken together, these data confirm that Ras-transformed cell death in response to fatty acid biosynthetic inhibition is mediated partly through a caspase-dependent mechanism.

Exogenous unsaturated fatty acids rescue the TOFA-induced cell death of Ras-transformed fibroblasts.

To determine whether the Ras-dependent cytotoxic effect of TOFA treatment is the result of depleted fatty acid stores, we attempted to rescue TOFA-induced cell death by adding back exogenous fatty acids. To this end, palmitic (16:0), stearic (18:0), or oleic (18:1) acids were added to Ras-transformed fibroblasts at various concentrations directly after TOFA treatment. As shown by the results in Fig. 4A to D, saturated fatty acids, such as palmitic acid and stearic acid, did not rescue cell viability or block cell death during TOFA treatment at concentrations up to 100 μM. In contrast, the addition of oleic acid, a monounsaturated fatty acid, increased cellular viability during TOFA treatment in a dose-dependent manner, almost completely rescuing viability and ablating TOFA-induced cell death at concentrations of 50 μM and above (Fig. 4E and F). To ensure that the addition of oleic acid did not affect TOFA's ability to inhibit fatty acid biosynthesis, we measured fatty acid biosynthesis after TOFA and oleic acid treatment. The addition of oleic acid did not affect TOFA's inhibition of fatty acid biosynthesis, although oleic acid addition by itself did slightly reduce fatty acid biosynthesis (Fig. 4G). Taken together, our data indicate that exogenous addition of an unsaturated fatty acid rescues TOFA-induced cell death.

FIG 4.

Exogenous unsaturated fatty acids rescue the TOFA-induced cell death of Ras-transformed cells. Ras-transformed fibroblasts were treated with DMSO, TOFA, and palmitic, stearic, or oleic acid as indicated. After 48 h of treatment, cells were collected and stained with trypan blue. (A, C, E) Live cells that excluded trypan blue were counted, and the results are presented as the percentages of the value for the DMSO control. Values are means + SEM. Asterisks indicate significant differences between the values for TOFA-treated and fatty acid- or TOFA + fatty acid-treated cells (*, P < 0.05; **, P < 0.01; ***, P < 0.001). (B, D, F) The percentages of dead cells were calculated after staining the cells with trypan blue. Values are means + SEM. Asterisks indicate significant differences between values for TOFA-treated and control, fatty acid-, or TOFA + fatty acid-treated cells (*, P < 0.05; **, P < 0.01). (G) After 48 h of the indicated TOFA or oleic acid treatments, cells were labeled with ¹⁴C-acetate for 2 h. Cellular lipids were extracted, saponified, and scintillation counted. All values are normalized to the protein concentration. Values are means + SEM. Asterisks indicate significant differences between values for control and treated cells (**, P < 0.01; ****, P < 0.0001).

ACC RNA interference reduces Ras-transformed cell proliferation.

To further confirm the importance of fatty acid biosynthesis for Ras-transformed cell viability and survival, we utilized ACC1-specific siRNA to target ACC expression. Cells were transfected with a control siRNA or an siRNA targeting ACC. Transfection with the ACC1-specific siRNA revealed substantial but not complete suppression of ACC protein expression in Ras-transformed fibroblasts relative to its expression in cells transfected with the control siRNA (Fig. 5A). Transfection with the ACC1 siRNA resulted in an ∼2.7-fold reduction in fatty acid biosynthesis compared to the results for the control siRNA. While substantial, this reduction was much less than the almost total blockage of fatty acid biosynthesis observed after TOFA treatment (Fig. 5B). This observed ACC1 knockdown attenuated cellular proliferation by ∼40 and ∼60% at 4 and 6 days posttransfection, respectively (Fig. 5C). However, the amount of cell death resulting from ACC1 knockdown was not significantly different from the level of cell death in the control (Fig. 5D). These results indicate that RNA interference (RNAi)-mediated targeting of ACC reduces the viability of transformed cells but does not induce substantial cell death. The observed differences in the rates of cell death between pharmacological and RNAi-mediated targeting of ACC1 likely result from the differences in their abilities to inhibit fatty acid biosynthesis.

FIG 5.

ACC RNA interference reduces Ras-transformed cell proliferation and viability. (A) Ras-transformed fibroblasts were transfected with 150 nM siRNA specific for ACC or a nonspecific control (Neg). After 48 h, the abundances of ACC and α-tubulin were measured by Western analysis. Data are from a representative experiment. (B) Ras-transformed fibroblasts were transfected with 150 nM siRNA specific for ACC or a nonspecific control (Neg) or treated with TOFA. Forty-eight hours posttransfection or post-TOFA treatment, cells were labeled with ¹⁴C-acetate for 2 h. Cellular lipids were extracted, saponified, and scintillation counted. All values are normalized to the protein concentration. Values are means + SEM. Asterisks indicate significant differences between bracketed values (*, P < 0.05; ***, P < 0.001). (C) Viable cells that excluded trypan blue are represented as the percentage of the value for the DMSO control. Values are means + SEM. An asterisk indicates a significant difference between the value for day 2 and for a subsequent day (*, P < 0.05). (D) The percentages of dead cells were calculated after staining the cells with trypan blue. Values are means + SEM.

Our results indicate that Ras-induced transformation of SV40-immortalized fibroblasts results in sensitivity to fatty acid biosynthetic inhibition that can be attenuated by supplementing the medium with unsaturated fatty acids. To test whether these findings extend to human tumor-derived cells, we analyzed the effect of fatty acid biosynthetic inhibition in a basal-like breast cancer cell line that harbors a constitutively active k-Ras allele (MDA-MB-231) and compared the results to those for a nontransformed but immortalized fibrocystic breast epithelial cell line (MCF 10A). The fully transformed breast cancer cell line exhibited increased sensitivity to TOFA treatment, i.e., more cell death at lower concentrations, compared to the sensitivity of the immortalized breast epithelial cells (Fig. 6A). The addition of oleic acid also rescued cell death induced by TOFA treatment (Fig. 6B), whereas palmitic acid actually increased TOFA's toxicity (Fig. 6C). These results provide additional evidence that Ras transformation induces sensitivity to fatty acid biosynthetic inhibition which can be reversed with oleic acid supplementation.

FIG 6.

Exogenous unsaturated fatty acids rescue the TOFA-induced cell death of MDA-MB-231 cells. (A) MDA-MB-231 or MCF 10A cells were collected 72 h after treatment with the indicated concentrations of TOFA. The percentages of dead cells were calculated after staining with trypan blue. Values are means + SEM. An asterisk indicates a significant difference between values for MCF 10A and MDA-MB-231 cells at that TOFA concentration (*, P < 0.05). (B and C) MDA-MB-231 cells were treated with DMSO or the indicated dose of TOFA, oleic acid (B), or palmitic acid (C). After 72 h, the cells were collected, and the percentages of dead cells were calculated after staining with trypan blue. Values are means + SEM. Asterisks indicate significant differences between values for TOFA-treated and control, fatty acid-, or TOFA + fatty acid-treated cells (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Oncogenic Ras expression in nontransformed breast epithelial cells induces sensitivity to fatty acid biosynthetic inhibition.

Given that the transformed MDA-MB-231 cells and the nontransformed MCF 10A cells are derived from different genetic backgrounds and require different culture conditions, we sought to examine the impact of Ras^12V expression in human breast epithelial cells. To this end, MCF 10A cells were lentivirally transduced separately with constitutively active HRas^12V or dominant-negative p53^175H or with both in conjunction (Fig. 7A). We found that shortly after transduction, the expression of HRas^12V was sufficient to induce fatty acid biosynthesis; however, p53^175H expression in conjunction with HRas^12V further upregulated the biosynthesis of fatty acids (Fig. 7B). Consistent with the observations in human fibroblasts (Fig. 1F), the induction of fatty acid biosynthesis associated with Ras expression appeared to be independent of changes in the expression of the lipogenic enzymes ACC, FAS, and ACL (Fig. 7A).

FIG 7.

Oncogenic Ras upregulates de novo fatty acid biosynthesis in MCF 10A cells without inducing the expression of lipogenic enzymes. (A) MCF 10A cells were transduced with Ras^12V and p53^175H, Ras^12V and vector 2 (control vector for p53^175H), vector 1 (control vector for Ras^12V) and p53^175H, or vector 1 and vector 2. The abundances of ACC, FAS, ACL, HRas, p53, and α-tubulin were measured by Western blotting. The data are from a representative experiment (n = 2). (B) Cells were labeled with ¹⁴C-acetate for 2 h. Cellular lipids were saponified, extracted, and scintillation counted. All values are normalized to the protein concentration. Values are means + SEM. Asterisks indicate significant differences relative to the value for MCF 10A cells transduced with control vectors (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Next we wanted to determine whether the expression of oncogenic Ras influenced the sensitivity of MCF 10A cells to fatty acid biosynthetic inhibition. Treatment with TOFA revealed that oncogenic Ras expression was sufficient to induce sensitivity of MCF 10A cells to fatty acid biosynthetic inhibition, which was further enhanced by p53^175H expression (Fig. 8A). Importantly, neither oncogenic Ras nor dominant-negative p53 affected the proliferation rate of MCF 10A cells (Fig. 8B). To confirm that the sensitivity of Ras-expressing MCF 10A cells to fatty acid biosynthetic inhibition was due to a limiting supply of fatty acids, we attempted to rescue TOFA-induced cell death by adding back oleic acid. As observed with human fibroblasts, the addition of exogenous unsaturated fatty acid was sufficient to restore the viability of Ras-transduced MCF 10A cells upon fatty acid biosynthetic inhibition (Fig. 8C). Together, these results indicate that oncogenic Ras expression is sufficient to induce sensitivity to fatty acid biosynthetic inhibition in human breast epithelial cells.

FIG 8.

Oncogenic Ras induces sensitivity to fatty acid biosynthetic inhibition in MCF 10A cells without enhancing proliferation. (A) MCF 10A cells were transduced with Ras^12V and p53^175H, Ras^12V and vector 2 (control vector for p53^175H), vector 1 (control vector for Ras^12V) and p53^175H, or vector 1 and vector 2. After 48 h of TOFA treatment, cell viability was assessed via MTT assay as described in Materials and Methods, and the results are presented as the percentages of the value for the DMSO control. Asterisks indicate significant differences relative to the value for MCF 10A cells transduced with control vectors (**, P < 0.01; ***, P < 0.001). (B) Cell numbers were quantified after plating on day 1. (C) MCF 10A cells transduced with HRas^12V and p53^175H or HRas^12V and vector 2 were treated with 10 μg/ml TOFA alone, 50 μM oleic acid alone, or both for 48 h. Cell viability was assessed via MTT assay, and the results are presented as the percentages of the value for the DMSO control. Asterisks indicate significant differences relative to the value for the TOFA-treated cells (**, P < 0.01; ***, P < 0.001) (all values are means + SEM).

DISCUSSION

Oncogenic transformation frequently results in a metabolic shift which activates glycolysis, nucleotide biosynthesis, and fatty acid biosynthesis (1–3). The activation of fatty acid biosynthesis has been shown to be critical for tumorigenesis (7, 8). The mechanisms of fatty acid biosynthetic activation and the means through which it contributes to oncogenesis are largely unknown, severely hampering our ability to design therapeutic strategies that take advantage of this metabolic dependency. Here, we report that oncogenic Ras induces fatty acid biosynthesis in both human SV40-immortalized fibroblasts and human mammary epithelial cells. Furthermore, this induction of fatty acid biosynthesis was independent of any effect on proliferation. Finally, the expression of oncogenic Ras induced a caspase-dependent sensitivity to fatty acid biosynthetic inhibition. Similar results were obtained in mammary epithelial cells, suggesting that specific oncogenic mutations can confer susceptibility to fatty acid biosynthetic inhibition which may provide promising new avenues for therapeutic intervention.

We find that stepwise oncogenic transformation induces a concomitant stepwise increase in fatty acid biosynthesis. Interestingly, in comparing the immortalized and Ras-transformed fibroblasts, there was no significant difference between the levels of key fatty acid biosynthetic enzymes, including ACL, ACC, and FAS. Previous reports indicate that tumors frequently induce the accumulation of fatty acid biosynthetic enzymes, including FAS and ACC (25, 26). Signaling through mTOR-linked pathways appears to be at least partially responsible for inducing the expression of these enzymes through the sterol response element binding protein (SREBP) family of transcription factors (24, 27), and SV40 proteins have also been found to induce mTOR activation early during infection (28). Consistent with an important role for activation of fatty acid biosynthesis, our results indicate that inhibition of mTORC1 through rapamycin treatment reduces the levels of FAS and attenuates fatty acid biosynthesis in both Ras-transformed and immortalized cells. However, the equivalent concentrations of fatty acid biosynthetic enzymes in Ras-transformed and immortalized cells suggest that additional levels of posttranslational regulation may be activated by oncogenic Ras to induce fatty acid biosynthesis.

Cancerous cells often respond differently than nontransformed cells to metabolic challenges. Oncogenesis frequently upregulates autophagy, which can confer protection from specific nutrient stresses (reviewed in reference 29). The SV40 small T antigen has been found to induce autophagy, resulting in resistance to glucose starvation (30). In contrast, other oncogenic alleles, such as Ras^12V, have been found to inhibit autophagy, and many cancer-derived cell lines are more sensitive to glucose and glutamine withdrawal than nontransformed cells (9, 10). AKT activation, which is also induced by the SV40 T antigens (31–33), is thought to be partially responsible, as AKT induction has been shown to convey glucose dependence to cancerous cells (34, 35). Our results indicate that the combined expression of SV40 T antigens and oncogenic Ras^12V induces sensitivity to fatty acid biosynthetic inhibition via TOFA in nontransformed cells. Furthermore, we find that this sensitivity can be rescued by supplementing the medium with unsaturated fatty acids but not with saturated fatty acids. The reason for this difference with respect to fatty acid saturation and rescue of fatty acid biosynthetic inhibition is not clear. One possibility is that transformed cells require longer unsaturated fatty acids and that supplemented saturated fatty acids are not accessible to cellular fatty acid elongases or desaturases. The finding that oleic acid is a predominant cellular fatty acid (36) is consistent with a cancer cell requirement for long-chain fatty acids. Furthermore, oncogenic Ras expression has been correlated with an increase in very-long-chain fatty acids (36). Finally, targeting cellular desaturases and depleting monounsaturated fatty acids has also recently been found to kill cancer cells (37). These findings suggest that fatty acid-modifying enzymes and the mechanisms of their oncogenic activation might be fertile areas for research into specifically killing cancerous cells.

Our finding that defined oncogenes can drive sensitivity to fatty acid biosynthetic inhibition in isogenic cell lines has important implications. The increased cancer cell toxicity that occurs upon metabolic challenge is a central tenet of many anticancer chemotherapies. Despite their clinical importance, the underlying mechanisms governing these cancer cell-specific toxicities are largely unknown. The typically large number of cancer cell mutations makes identifying links between specific mutations and oncogenic sensitivity very difficult. This highlights an advantage of genetically defined models of oncogenesis, as these causal links are more readily identifiable when comparing isogenic cells that differ only in a relatively small number of defined ways. Interrogation of how oncogenic alleles change cellular responses to metabolic challenge will likely highlight mechanisms that govern differential responses to these challenges in cancer cells, i.e., those responsible for increased toxicity. These mechanisms can therefore be directly targeted for more specific killing of cancerous cells.

In summary, we find that the expression of Ras^12V drives an increased sensitivity to pharmaceutical inhibition of fatty acid biosynthesis which can be rescued by supplementing the medium with unsaturated fatty acids. The demonstration that defined oncogenic alleles can induce cancer-associated metabolic sensitivity will help facilitate the identification of specific mechanisms that govern the responses to metabolic challenge. Understanding the inner workings of these responses is essential for the ability to manipulate them in therapeutically beneficial ways, that is, to specifically kill cancerous cells.