- Email: [email protected]
Fatty Acid Synthase Inhibition in Human Breast Cancer Cells Leads to Malonyl-CoA-Induced Inhibition of Fatty Acid Oxidation and Cytotoxicity
Biochemical and Biophysical Research Communications 285, 217–223 (2001) doi:10.1006/bbrc.2001.5146, available online at http://www.idealibrary.com on
Fatty Acid Synthase Inhibition in Human Breast Cancer Cells Leads to Malonyl-CoA-Induced Inhibition of Fatty Acid Oxidation and Cytotoxicity Jagan N. Thupari, Michael L. Pinn, and Francis P. Kuhajda 1 Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287
Received June 8, 2001
Inhibition of fatty acid synthase (FAS) induces apoptosis in human breast cancer cells in vitro and in vivo without toxicity to proliferating normal cells. We have previously shown that FAS inhibition causes a rapid increase in malonyl-CoA levels identifying malonyl-CoA as a potential trigger of apoptosis. In this study we further investigated the role of malonyl-CoA during FAS inhibition. We have found that: [i] inhibition of FAS with cerulenin causes carnitine palmitoyltransferase-1 (CPT-1) inhibition and fatty acid oxidation inhibition in MCF-7 human breast cancer cells likely mediated by elevation of malonyl-CoA; [ii] cerulenin cytotoxicity is due to the nonphysiological state of increased malonyl-CoA, decreased fatty acid oxidation, and decreased fatty acid synthesis; and [iii] the cytotoxic effect of cerulenin can be mimicked by simultaneous inhibition of CPT-1, with etomoxir, and fatty acid synthesis with TOFA, an acetyl-CoA carboxylase (ACC) inhibitor. This study identifies CPT-1 and ACC as two new potential targets for cancer chemotherapy. © 2001 Academic Press
Key Words: fatty acid synthase; carnitine palmitoyltransferase-1; etomoxir; cerulenin; malonyl-CoA; cancer; acetyl-CoA carboxylase.
Abnormal fatty acid synthesis is among the most prevalent features of human cancer. Breast, prostate, colorectal, and ovarian cancers are among the list of common human tumors known to express high levels of fatty acid synthase (FAS, EC 2.3.1.85) and undergo constitutively elevated levels of fatty acid synthesis (1). Mammalian FAS is a complex multifunctional enzyme This work was presented in part at the 2001 meeting of the American Association of Cancer Research, New Orleans, LA. 1 To whom correspondence should be addressed at Johns Hopkins Bayview Research Campus, Department of Pathology, Building AA, Room 154A, 4940 Eastern Avenue, Baltimore, MD 21224. Fax: 410550-0075. E-mail: [email protected].
that contains seven catalytic domains and a 4⬘phosphopantetheine prosthetic group on a single 260,000 dalton polypeptide (2). Palmitate, a saturated 16-carbon fatty acid is the predominant product of FAS that is synthesized de novo from the substrates acetylCoA, malonyl-CoA, and NADPH (Fig. 1). Since FAS is largely down-regulated by dietary fat in lipogenic tissues such as liver and adipose tissue, but remains highly expressed in many cancers, FAS is an attractive therapeutic target for cancer chemotherapy. Inhibition of FAS in vitro induces apoptosis in a variety of human cancer cells including breast, prostate, colon, and ovarian cell lines (3– 8). Treatment of human breast and prostate cancer xenografts in athymic mice with FAS inhibitors have shown a significant anti-tumor effect without toxicity to proliferating normal tissues such as bone marrow, skin, liver, and gastrointestinal tract (6, 9). Even in animals with high levels of hepatic fatty acid synthesis, such as leptindeficient (ob/ob) mice, both liver histology and serum liver enzyme levels have remained normal following treatment with FAS inhibitors (10). Reversible weight loss with reduction of adipose tissue has been the only consistent alteration in normal tissues of mice treated with FAS inhibitors (10). In addition to a reduction in de novo fatty acid synthesis, FAS inhibition in cancer cells leads to a rapid and profound increase in malonyl-CoA, the predominant substrate for FAS (6). Within 30 min of FAS inhibition, malonyl-CoA levels rise ninefold over controls well in advance of cancer cell apoptosis. Conversely, if the rise in malonyl-CoA is blocked by pretreating the cells with TOFA, an inhibitor of acetylCoA carboxylase (11) (ACC, EC 6.4.1.2) (Fig. 1) the cytotoxic effect of FAS inhibition is markedly reduced. These studies identified malonyl-CoA as a trigger in downstream events leading to apoptosis during FAS inhibition. Malonyl-CoA, in addition to its role as a substrate for FAS, is a key regulator of energy metabolism through
217
0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Schematic representation of the fatty acid synthesis pathway showing the enzyme targets for cerulenin (fatty acid synthase (FAS)) and TOFA (acetyl-CoA carboxylase (ACC)). ACC, the regulatory enzyme of fatty acid synthesis, carboxylates acetyl-CoA to produce malonyl-CoA, the principal substrate for FAS. FAS catalyzes the production of free palmitate from NADPH, acetyl-CoA, and malonyl-CoA. Long chain acyl-CoA synthase esterifies CoA to palmitate producing palmitoyl-CoA. TOFA, an inhibitor of ACC, is esterified to CoA to produce TOFA-CoA which profoundly inhibits ACC. TOFA inhibits fatty acid synthesis by reducing malonyl-CoA. In contrast, cerulenin also inhibits FAS, reducing fatty acid synthesis, but increasing malonyl-CoA levels. Moreover, FAS inhibition leads to reduced levels of palmitoyl-CoA, further stimulating ACC and elevation of malonyl-CoA levels.
its inhibition of fatty acid oxidation (Fig. 2). MalonylCoA inhibits carnitine palmitoyltransferase-1 (CPT-1, EC 2.3.1.21); CPT-1 transesterifies long-chain acylCoA’s to acylcarnitine permitting their entry into the mitochondria for fatty acid oxidation (12). During lipogenesis, high steady-state levels of malonyl-CoA inhibit CPT-1 preventing mitochondrial oxidation of newly synthesized fatty acids. During starvation, a high fat diet, or TOFA treatment, de novo fatty acid synthesis is reduced, malonyl-CoA levels fall, and CPT-1 is activated allowing entry of fatty acid into the mitochondria for oxidation. Inhibition of fatty acid synthesis at the FAS step, reduces de novo fatty acid synthesis but elevates malonyl-CoA levels leading to the nonphysiologic metabolic state of concomitant inhibition of both fatty acid synthesis and fatty acid oxidation. This study further investigates the consequences of FAS inhibition on fatty acid oxidation, CPT-1 activity, and cytotoxicity in human breast cancer cells. EXPERIMENTAL PROCEDURES Carnitine palmitoyltransferase-1 (CPT-1) assay. CPT-1 activity was measured in digitonin permeabilized MCF-7 cells using a protocol outlined by Sleboda et al. with minor modifications (13). MCF-7 cells are plated in DMEM with 10% fetal bovine serum at 1 ⫻ 10 6
cells in 6-well plates in triplicate. Following overnight incubation at 37°C, drugs diluted from a 5 mg/ml stock in DMSO or vehicle are added for the times and concentrations indicated. Following drug or vehicle incubation, the medium is removed and replaced with 700 l of assay medium consisting of 50 mM imidazole, 70 mM KCl, 80 mM sucrose, 1 mM EGTA, 2 mM MgCl 2, 1 mM DTT, 1 mM KCN, 1 mM ATP, 0.1% fatty acid free bovine serum albumin, 70 M palmitoylCoA, 0.25 Ci [methyl- 14C]L-carnitine (6.4 M), 40 g digitonin with or without 20 M malonyl-CoA, drugs at indicated dilutions, or vehicle alone. After incubation for 6 min at 37°C, the reaction is stopped by the addition of 500 l of ice-cold 4 M perchloric acid. Cells are then harvested and centrifuged at 13,000g for 5 min. The pellet is washed with 500 l ice cold 2 mM perchloric acid and centrifuged again. The resulting pellet is resuspended in 800 l dH 2O and extracted with 400 l of butanol. The butanol phase is counted by liquid scintillation and represents the acylcarnitine derivative. Fatty acid oxidation assay. Fatty acid oxidation is measured using methods outlined by Antinozzi et al., with minor modifications (14). 1 ⫻ 10 6 MCF-7 cells are plated in T-25 flasks in triplicate and incubated overnight at 37°C. Drugs are then added as indicated diluted from 5 mg/ml stock in DMSO. After 2 h, medium with drugs are removed and cells are preincubated for 30 min with 1.5 ml of the following buffer: 114 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2PO 4, 1.2 mM MgSO 4, glucose 11 mM. After preincubation, 200 l of assay buffer is added containing 114 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2PO 4, 1.2 mM MgSO 4, glucose 11 mM, 2.5 mM palmitate (containing with 10 Ci of [1- 14C]palmitate) bound to albumin, 0.8 mM L-carnitine, and cells are incubated at 37°C for 2 h. Following the incubation, 400 l of benzothonium hydrochloride is added to the center well to collect released 14CO 2. Immediately, the reaction is stopped by adding 500 l of 7% perchloric acid to the cells. The flasks
218
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Schematic representation of fatty acid transport regulation for -oxidation in the mitochondrion. Oxidation of long chain fatty acids in the mitochondria is highly regulated by a transport system consisting of three proteins, CPT-1, CACT, and CPT-2. CPT-1, regulated physiologically by competitive malonyl-CoA inhibition, transesterifies fatty acids from CoA to carnitine. The acylcarnitines are then transported to CPT-2 via CACT. CPT-2 reverses the transesterification of fatty acids, from acylcarnitines back to acyl-CoA’s for fatty acid oxidation. High levels of malonyl-CoA generated during fatty acid synthesis inhibit CPT-1 preventing oxidation of newly synthesized fatty acids. Etomoxir is an irreversible inhibitor of CPT-1. Structural biological studies have found an association of CPT-1 with BCL-2 on the mitochondrial surface. Porin may also be associated with this fatty acid transporter complex.
with wells are then incubated for 2 h at 37°C after which the benzothonium hydrochloride is removed and counted for 14C. Blanks are prepared by adding 500 l of 7% perchloric acid to the cells prior to the incubation with the assay buffer for 2 h. Cell growth inhibition assay. MCF-7 cells are plated in 24-well plates at 5 ⫻ 10 4 cells per well in DMEM with 10% fetal bovine serum (Hyclone). After overnight incubation at 37°C, drugs were added from stock 5 mg/ml solutions in DMSO. The final concentration of DMSO in the cultures was at or below 0.2%. After either 48 or 72 h, medium was removed, and wells were washed thrice with Hank’s buffered saline. Wells were stained with crystal violet, dried, and solubilized in 10% SDS. One hundred microliter aliquots were transferred to a 96-well plate and read on a Molecular Dynamics plate reader at 490 nm. Clonogenic assay. After overnight incubation at 37°C, 1 ⫻ 10 6 MCF-7 cells were exposed to drugs as indicated for 6 h or vehicle control, washed, detached by trypsin digestion, counted, and plated at 1000 or 500 cells/60-mm plate in triplicate. Colonies were stained with crystal violet and counted 4 – 6 days after plating. Miscellaneous methods and chemicals. MCF-7 cells were obtained from the American Type Culture Collection. TOFA (5(tetradecyloxy)-2-furoic acid) was synthesized at the Johns Hopkins University, by Dr. Craig A. Townsend. Etomoxir was purchased from Dr. H.P.O. Wolfe. Cerulenin and other chemicals were obtained from Sigma. Etomoxir, cerulenin, and TOFA were added at the designated concentrations from a 5 mg/ml stock in DMSO. DMSO concentrations in the final cell culture medium never exceeded 0.1%. [1- 14C]palmitate and [1-methyl 14C] L-carnitine were obtained from Amersham Pharmacia Biotech.
Statistics. Results are expressed as means ⫾ standard error of the mean. Statistical analysis was performed using two-tailed t-tests in Prism 3.0 (Graph Pad).
RESULTS FAS and CPT-1 inhibition but not ACC inhibition leads to reduced FA oxidation. Since FAS inhibition leads to high levels of malonyl-CoA within 30 min, and malonyl-CoA is a potent inhibitor of FA oxidation, we tested whether cerulenin treatment would lead to FA oxidation inhibition in MCF-7 cells. Cerulenin is an irreversible inhibitor of both type I and type II fatty acid synthases (15, 16). After 2 h of cerulenin treatment at 10 g/ml, FA oxidation was reduced by 50% of controls (Fig. 3, upper panel). Importantly, this dose of cerulenin in MCF-7 cells is known to cause: [i] a ⬎50% decrease of fatty acid synthesis within 2 h; [ii] a 900% increase in malonyl-CoA within ⬍1 h; and [iii] ⬎90% cytotoxicity in clonogenic assays (6). Thus, FA oxidation inhibition occurs during fatty acid synthesis inhibition but before the onset of cytotoxicity. Etomoxir, an irreversible inhibitor of CPT-1 (17), also reduced fatty acid oxidation (Fig. 3, middle panel). TOFA, a reversible inhibitor of ACC, had a trend toward stimulation
219
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
of fatty acid oxidation at 5 g/ml that may be secondary to reduced malonyl-CoA levels (Fig. 3, lower panel). Cerulenin induced FA oxidation inhibition is likely mediated by malonyl-CoA elevation. FA oxidation inhibition by cerulenin could be due to direct inhibition of CPT-1 by cerulenin or mediated through increased malonyl-CoA. To discriminate between these possibilities, CPT-1 activity was measured directly in digitonin permeabilized cells in the presence of cerulenin. Cerulenin treatment at 10 g/ml had no direct inhibitory effect on CPT-1 activity in MCF-7 cells (Fig. 4, upper panel) although the same dose caused a 50% reduction in fatty acid oxidation (Fig. 3, upper panel). Exogenously administered malonyl-CoA [20 mM], a known reversible inhibitor of CPT-1 (18), caused an approximately 40% reduction in CPT-1 activity. CPT-1 in MCF-7 cells is inhibitable by malonyl-CoA, but not directly by cerulenin implying that cerulenin inhibits FA oxidation through malonyl-CoA signaling. The K i of malonyl-CoA inhibition of CPT-1 in MCF-7 cells was ⬃20 –30 uM (data not shown). This is consistent with expression of the liver isoform of CPT-1 in MCF-7 cells (19). Etomoxir showed a significant reduction in CPT-1 activity (Fig. 4, middle panel), with TOFA having no significant effect (Fig. 4, lower panel). FA oxidation inhibition is not the sole cause of FAS induced cytotoxicity in cancer cells. To test if FA oxidation inhibition is the primary mechanism inducing cytotoxicity during FAS inhibition, MCF-7 cells were treated with etomoxir at a series of doses (6.25–200 g/ml) that encompassed at least a 50% inhibition of FA oxidation in MCF-7 cells. Etomoxir showed significant growth inhibition only at 200 g/ml after 72 h; lower doses failed to cause significant growth inhibition (Fig. 5). At doses of 200 g/ml or greater, etomoxir looses its specificity as a CPT-1 inhibitor, and is reported to also inhibit CPT-2 (17, 20). Combined FA synthesis and FA oxidation inhibition is cytotoxic to MCF-7 cells. Combined TOFA and etomoxir treatment of MCF-7 cells should produce similar metabolic conditions as cerulenin treatment, and thus should show similar cytotoxicity. Figure 6 shows a clonogenic assay of MCF-7 cells treated with etomoxir alone, TOFA alone, and TOFA and etomoxir combined. TOFA and etomoxir alone had no cytotoxic effect. The combination of TOFA and etomoxir caused about a 50% reduction in clonogenicity
FIG. 3. FAS and CPT-1 inhibition, but not ACC inhibition, reduce fatty acid oxidation in MCF-7 cells. Upper panel, FAS inhibition with cerulenin (10 g/ml) in MCF-7 cells for 1 h resulted in a 50% reduction in fatty acid oxidation (P ⫽ 0.0007, two-tailed t-test).
Middle panel, CPT-1 inhibition with etomoxir (25 g/ml) for 1 h led to an approximately 30% reduction in fatty acid oxidation (P ⫽ 0.03, two-tailed t-test) and a 70% reduction at 50 g/ml (P ⫽ 0.01, two-tailed t-test). Lower panel, TOFA inhibition of ACC has no significant effect on fatty acid oxidation. Error bars represent standard error of the mean in duplicate samples.
220
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DISCUSSION Fatty acid synthesis and fatty acid oxidation are key anabolic and catabolic pathways linked through their common metabolic intermediate, malonyl-CoA. Thus, it is not surprising that rapid and profound alterations in malonyl-CoA levels and CPT-1 activity during FAS inhibition could lead to metabolic chaos in a cancer cell that already exhibits abnormal fatty acid metabolism. The in vitro cytotoxic effects of cerulenin on human cancer cells displaying high levels of fatty acid synthesis, such as MCF-7 cells, have been well established. Using DNA laddering (5) or merocyanine staining (6) as a markers of apoptosis, cerulenin treatment consistently led to apoptosis within 16 –24 h at micromolar concentrations. Apoptosis was preceded by inhibition of fatty acid synthesis by at least 50% within 2 h and a ninefold increase in malonyl-CoA. Pretreatment of the cells with TOFA, inhibited ACC and malonyl-CoA production rescuing the cells from cerulenin induced apoptosis and implicating malonyl-CoA as a trigger for apoptosis in cancer cells. In contrast, human cancer cells with low levels of fatty acid synthesis, such as RKO human colon cancer cells, were resistant to apoptosis following cerulenin treatment. Instead of apoptosis, RKO cells treated with cerulenin underwent a biphasic growth arrest, first during S phase, then again in G 1 and G 2 with accumulation of p53 and p21 (21). Interestingly, RKO cells with mutant p53 exhibited a brisk apoptotic response to cerulenin, similar to cells with high levels of fatty acid synthesis, including the ability to be blocked with TOFA pretreatment. These data indicate that the apoptotic response to cerulenin is dependent upon high levels of fatty acid synthesis and/or p53 mutation. Moreover, malonyl-CoA accumulation appears central to both apoptotic mechanisms but not to growth arrest.
FIG. 4. CPT-1 activity in permeabilized MCF-7 cells treated with malonyl-CoA or inhibitors of FAS, ACC, and CPT-1. Upper panel, cerulenin at 10 g/ml, a dose that reduced fatty acid oxidation by a 50%, had no significant effect on CPT-1 activity. In contrast, malonyl-CoA at 20 M caused a 75% reduction in CPT-1 activity (P ⫽ 0.007, two-tailed t-test). Middle panel, etomoxir, a CPT-1 inhibitor, reduced CPT-1 activity by ⬎80% at 10 g/ml (P ⫽ 0.02, two-tailed t-test). Lower panel, TOFA, an ACC inhibitor had no significant effect on CPT-1 activity. Error bars represent standard error of the mean with duplicate samples.
constituting a significant cytotoxic response. Interestingly, the combination of cerulenin and etomoxir showed no additive effect over cerulenin alone in growth inhibition studies (data not shown). Since cerulenin already inhibits CPT-1 via increased malonyl-CoA levels, the addition of another CPT-1 inhibitor did not potentiate, further supporting the proposed mechanism of action of cerulenin.
FIG. 5. Etomoxir does not reduce the growth of MCF-7 cells in vitro. Only the 200 g/ml dose of etomoxir had a significant growth inhibitory effect on MCF-7 cells (P ⫽ 0.006, two-tailed t-test). At 50 g/ml, etomoxir caused a ⬎50% reduction in fatty acid oxidation, similar to cerulenin. Error bars represent standard error of the mean with quadruplicate samples.
221
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 6. Combined etomoxir and TOFA treatment is cytotoxic to MCF-7 cells in vitro. Cells were treated with etomoxir 100 g/ml, that causes ⬎50% inhibition of fatty acid oxidation, and with TOFA 5 g/ml that inhibit fatty acid synthesis similar to cerulenin. Alone, these inhibitors have no significant effect on MCF-7 clonogenicity. Combined, however, they produce a 50% reduction in clonogenicity compared to control and are significantly more cytotoxic that either TOFA (P ⫽ 0.004, two-tailed t-test) or etomoxir (P ⫽ 0.002, two-tailed t-test) alone. Error bars represent standard error of the mean with triplicate samples.
CPT-1 inhibition has also been implicated in apoptosis in cancer cells. During an investigation of genes involved in cell death of mouse hematopoetic cells, CPT-1 emerged as candidate gene involved in apoptosis. In mouse LyD9 and WEHI-231 cells, etomoxir inhibition of CPT-1 augmented palmitate induced apoptosis (22). Moreover, CPT-1 co-immunoprecipitated with BCL-2 suggesting a physical association between CPT-1 and the mitochondrial apoptotic apparatus (23). These studies all point to CPT-1 as potentially involved potentiating cell death. In our studies as with others, inhibition of CPT-1 alone was insufficient to induce apoptosis (22). Etomoxir treatment leading to the same magnitude of fatty acid oxidation and CPT-1 inhibition as cerulenin failed to growth inhibit or kill MCF-7 cells. However, etomoxir potentiated the anti-tumor effect of inhibition of fatty acid synthesis at the ACC step with TOFA leading to cell death. The combination of fatty acid oxidation inhibition and fatty acid synthesis inhibition is non-physiologic, mimicked the metabolic consequences of FAS inhibition by cerulenin, and was cytotoxic to cancer cells. Interestingly, in the prior studies with mouse hematopoetic cells, etomoxir enhanced the apoptosis induced by palmitate administration which is similar to our combination of etomoxir and TOFA. Palmitate administration would lead to a rapid increase in intracellular palmitoyl-CoA through the activity of acyl-CoA synthase. Like TOFA, palmitoyl-CoA is a potent inhibitor of ACC activity. Palmitoyl-CoA inhibition of ACC
would lead to decreased malonyl-CoA and decreased fatty acid synthesis; this by itself is not cytotoxic. However, in the presence of etomoxir, palmitoyl-CoA inhibition of ACC could cause the same non-physiological state of reduced fatty acid oxidation and fatty acid synthesis as TOFA and etomoxir. Consequently, the combination of palmitate inhibition and etomoxir would have similar metabolic consequences to TOFA and etomoxir, or cerulenin alone—all cytotoxic to cancer cells. Cerulenin induced apoptosis in human cancer cells is due to its simultaneous inhibition of fatty acid synthesis at the FAS step with subsequent elevation of malonyl-CoA, CPT-1 inhibition, and inhibition of fatty acid oxidation. This nonphysiological biochemical state, well tolerated by normal cells both in vitro and in vivo, is rapidly cytotoxic to many types of cancer cells. Our study indicates that CPT-1 inhibitors, such as etomoxir, which have been used in humans for diabetes control (24), may have anti-cancer effects if combined with an inhibitor of ACC. These findings may ultimately provide new strategies for cancer chemotherapy based upon combinations of metabolic inhibitors targeting fatty acid synthesis and oxidation. ACKNOWLEDGMENTS This work was supported by National Institute of Health Grant RO1 CA87850-02. We thank Dr. Townsend for his gift of TOFA for these studies.
222
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES 1. Kuhajda, F. P. (2000) Nutrition 16, 202–208. 2. Wakil, S. (1989) Biochemistry 28, 4523– 4530. 3. Kuhajda, F. P., Jenner, K., Wood, F. D., Hennigar, R. A., Jacobs, L. B., Dick, J. D., and Pasternack, G. R. (1994) Proc. Natl. Acad. Sci. USA 91, 6379 – 6383. 4. Pizer, E. S., Wood, F. D., Heine, H. S., Romantsev, F. R., Pasternack, G. R., and Kuhajda, F. P. (1996) Cancer Res. 56, 1189 – 1193. 5. Pizer, E. S., Jackisch, C., Wood, F. D., Pasternack, G. R., Davidson, N. E., and Kuhajda, F. (1996) Cancer Res. 56, 2745–2747. 6. Pizer, E. S., Thupari, J., Han, W. F., Pinn, M. L., Chrest, F. J., Frehywot, G. L., Townsend, C. A., and Kuhajda, F. P. (2000) Cancer Res. 60, 213–218. 7. Pizer, E. S., Chrest, F. J., DiGiuseppe, J. A., and Han, W. F. (1998) Cancer Res. 58, 4611– 4615. 8. Furuya, Y., Akimoto, S., Yasada, K., and Ito, H. (1997) Anticancer Res. 17, 4589 – 4593. 9. Pizer, E. S., Pflug, B. R., Bova, G. S., Han, W. F., Udan, M. S., and Nelson, J. B. (2000) Proc. Amer. Assoc. Cancer Res. 41, 655. 10. Loftus, T. M., Jaworsky, D. E., Frehywot, G. L., Townsend, C. A., Ronnett, G. V., Lane, M. D., and Kuhajda, F. P. (2000) Science 288, 2379 –2381. 11. Halvorson, D. L., and McCune, S. A. (1984) Lipids 19(11), 851– 856.
12. McGarry, J. D., and Brown, N. F. (1997) Eur. J. Biochem. 244, 1–14. 13. Sleboda, J., Risan, K. A., Spydevold, O., and Bremer, J. (1999) Biochim.Biophys. Acta 1436, 541–549. 14. Antinozzi, P. A., Segall, L., Prentki, M., McGarry, J. D., and Newgard, C. B. (1998) J. Biol. Chem. 273, 16146 –16154. 15. Funabashi, H., Iwaski, S., Okuda, S., and Omura, S. (1983) Tetrahedron. Lett. 24, 2673–2676. 16. Funabashi, H., Kawaguchi, A., Tomoda, H., Omura, S., Okuda, S., and Iwasaki, S. (1989) J. Biochem. 105, 751–755. 17. Lilly, K., Chung, C., Kerner, J., VanRenterghem, R., and Bieber, L. L. (1992) Biochem. Pharmacol. 43, 353–361. 18. Saggerson, E. D., and Carpenter, C. A. (1981) FEBS Lett. 132, 166 –168. 19. Swanson, S. T., Foster, D. W., McGarry, J. D., and Brown, N. F. (1998) Biochem. J. 335, 513–519. 20. Lopaschuk, G. D., McNeil, G. F., and McVeigh, J. J. (1989) Mol. Cell. Biochem. 88, 175–179. 21. Li, J., Gorospe, M., Chrest, F. J., T. S., K., Evans, M. K., Han, W. F., and Pizer, E. S. (2001) Cancer Res. 61, 1493–1499. 22. Paumen, M. B., Ishida, Y., Muramatsu, M., Yamamoto, M., and Honjo, T. (1997) J. Biol. Chem. 272, 3324 –3329 23. Paumen, M. B., Ishida, Y., Han, H., Muramatsu, M., Eguchi, Y., Tsujimoto, Y., and Honjo, T. (1997) Biochem. Biophys. Res. Commun. 232, 523–525. 24. Schmidt-Schweda, S., and Holubarsch, C. (2000) Clin. Sci. 99, 27–35.
223
Fatty Acid Synthase Inhibition in Human Breast Cancer Cells Leads to Malonyl-CoA-Induced Inhibition of Fatty Acid Oxidation and Cytotoxicity Jagan N. Thupari, Michael L. Pinn, and Francis P. Kuhajda 1 Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287
Received June 8, 2001
Inhibition of fatty acid synthase (FAS) induces apoptosis in human breast cancer cells in vitro and in vivo without toxicity to proliferating normal cells. We have previously shown that FAS inhibition causes a rapid increase in malonyl-CoA levels identifying malonyl-CoA as a potential trigger of apoptosis. In this study we further investigated the role of malonyl-CoA during FAS inhibition. We have found that: [i] inhibition of FAS with cerulenin causes carnitine palmitoyltransferase-1 (CPT-1) inhibition and fatty acid oxidation inhibition in MCF-7 human breast cancer cells likely mediated by elevation of malonyl-CoA; [ii] cerulenin cytotoxicity is due to the nonphysiological state of increased malonyl-CoA, decreased fatty acid oxidation, and decreased fatty acid synthesis; and [iii] the cytotoxic effect of cerulenin can be mimicked by simultaneous inhibition of CPT-1, with etomoxir, and fatty acid synthesis with TOFA, an acetyl-CoA carboxylase (ACC) inhibitor. This study identifies CPT-1 and ACC as two new potential targets for cancer chemotherapy. © 2001 Academic Press
Key Words: fatty acid synthase; carnitine palmitoyltransferase-1; etomoxir; cerulenin; malonyl-CoA; cancer; acetyl-CoA carboxylase.
Abnormal fatty acid synthesis is among the most prevalent features of human cancer. Breast, prostate, colorectal, and ovarian cancers are among the list of common human tumors known to express high levels of fatty acid synthase (FAS, EC 2.3.1.85) and undergo constitutively elevated levels of fatty acid synthesis (1). Mammalian FAS is a complex multifunctional enzyme This work was presented in part at the 2001 meeting of the American Association of Cancer Research, New Orleans, LA. 1 To whom correspondence should be addressed at Johns Hopkins Bayview Research Campus, Department of Pathology, Building AA, Room 154A, 4940 Eastern Avenue, Baltimore, MD 21224. Fax: 410550-0075. E-mail: [email protected].
that contains seven catalytic domains and a 4⬘phosphopantetheine prosthetic group on a single 260,000 dalton polypeptide (2). Palmitate, a saturated 16-carbon fatty acid is the predominant product of FAS that is synthesized de novo from the substrates acetylCoA, malonyl-CoA, and NADPH (Fig. 1). Since FAS is largely down-regulated by dietary fat in lipogenic tissues such as liver and adipose tissue, but remains highly expressed in many cancers, FAS is an attractive therapeutic target for cancer chemotherapy. Inhibition of FAS in vitro induces apoptosis in a variety of human cancer cells including breast, prostate, colon, and ovarian cell lines (3– 8). Treatment of human breast and prostate cancer xenografts in athymic mice with FAS inhibitors have shown a significant anti-tumor effect without toxicity to proliferating normal tissues such as bone marrow, skin, liver, and gastrointestinal tract (6, 9). Even in animals with high levels of hepatic fatty acid synthesis, such as leptindeficient (ob/ob) mice, both liver histology and serum liver enzyme levels have remained normal following treatment with FAS inhibitors (10). Reversible weight loss with reduction of adipose tissue has been the only consistent alteration in normal tissues of mice treated with FAS inhibitors (10). In addition to a reduction in de novo fatty acid synthesis, FAS inhibition in cancer cells leads to a rapid and profound increase in malonyl-CoA, the predominant substrate for FAS (6). Within 30 min of FAS inhibition, malonyl-CoA levels rise ninefold over controls well in advance of cancer cell apoptosis. Conversely, if the rise in malonyl-CoA is blocked by pretreating the cells with TOFA, an inhibitor of acetylCoA carboxylase (11) (ACC, EC 6.4.1.2) (Fig. 1) the cytotoxic effect of FAS inhibition is markedly reduced. These studies identified malonyl-CoA as a trigger in downstream events leading to apoptosis during FAS inhibition. Malonyl-CoA, in addition to its role as a substrate for FAS, is a key regulator of energy metabolism through
217
0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Schematic representation of the fatty acid synthesis pathway showing the enzyme targets for cerulenin (fatty acid synthase (FAS)) and TOFA (acetyl-CoA carboxylase (ACC)). ACC, the regulatory enzyme of fatty acid synthesis, carboxylates acetyl-CoA to produce malonyl-CoA, the principal substrate for FAS. FAS catalyzes the production of free palmitate from NADPH, acetyl-CoA, and malonyl-CoA. Long chain acyl-CoA synthase esterifies CoA to palmitate producing palmitoyl-CoA. TOFA, an inhibitor of ACC, is esterified to CoA to produce TOFA-CoA which profoundly inhibits ACC. TOFA inhibits fatty acid synthesis by reducing malonyl-CoA. In contrast, cerulenin also inhibits FAS, reducing fatty acid synthesis, but increasing malonyl-CoA levels. Moreover, FAS inhibition leads to reduced levels of palmitoyl-CoA, further stimulating ACC and elevation of malonyl-CoA levels.
its inhibition of fatty acid oxidation (Fig. 2). MalonylCoA inhibits carnitine palmitoyltransferase-1 (CPT-1, EC 2.3.1.21); CPT-1 transesterifies long-chain acylCoA’s to acylcarnitine permitting their entry into the mitochondria for fatty acid oxidation (12). During lipogenesis, high steady-state levels of malonyl-CoA inhibit CPT-1 preventing mitochondrial oxidation of newly synthesized fatty acids. During starvation, a high fat diet, or TOFA treatment, de novo fatty acid synthesis is reduced, malonyl-CoA levels fall, and CPT-1 is activated allowing entry of fatty acid into the mitochondria for oxidation. Inhibition of fatty acid synthesis at the FAS step, reduces de novo fatty acid synthesis but elevates malonyl-CoA levels leading to the nonphysiologic metabolic state of concomitant inhibition of both fatty acid synthesis and fatty acid oxidation. This study further investigates the consequences of FAS inhibition on fatty acid oxidation, CPT-1 activity, and cytotoxicity in human breast cancer cells. EXPERIMENTAL PROCEDURES Carnitine palmitoyltransferase-1 (CPT-1) assay. CPT-1 activity was measured in digitonin permeabilized MCF-7 cells using a protocol outlined by Sleboda et al. with minor modifications (13). MCF-7 cells are plated in DMEM with 10% fetal bovine serum at 1 ⫻ 10 6
cells in 6-well plates in triplicate. Following overnight incubation at 37°C, drugs diluted from a 5 mg/ml stock in DMSO or vehicle are added for the times and concentrations indicated. Following drug or vehicle incubation, the medium is removed and replaced with 700 l of assay medium consisting of 50 mM imidazole, 70 mM KCl, 80 mM sucrose, 1 mM EGTA, 2 mM MgCl 2, 1 mM DTT, 1 mM KCN, 1 mM ATP, 0.1% fatty acid free bovine serum albumin, 70 M palmitoylCoA, 0.25 Ci [methyl- 14C]L-carnitine (6.4 M), 40 g digitonin with or without 20 M malonyl-CoA, drugs at indicated dilutions, or vehicle alone. After incubation for 6 min at 37°C, the reaction is stopped by the addition of 500 l of ice-cold 4 M perchloric acid. Cells are then harvested and centrifuged at 13,000g for 5 min. The pellet is washed with 500 l ice cold 2 mM perchloric acid and centrifuged again. The resulting pellet is resuspended in 800 l dH 2O and extracted with 400 l of butanol. The butanol phase is counted by liquid scintillation and represents the acylcarnitine derivative. Fatty acid oxidation assay. Fatty acid oxidation is measured using methods outlined by Antinozzi et al., with minor modifications (14). 1 ⫻ 10 6 MCF-7 cells are plated in T-25 flasks in triplicate and incubated overnight at 37°C. Drugs are then added as indicated diluted from 5 mg/ml stock in DMSO. After 2 h, medium with drugs are removed and cells are preincubated for 30 min with 1.5 ml of the following buffer: 114 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2PO 4, 1.2 mM MgSO 4, glucose 11 mM. After preincubation, 200 l of assay buffer is added containing 114 mM NaCl, 4.7 mM KCl, 1.2 mM KH 2PO 4, 1.2 mM MgSO 4, glucose 11 mM, 2.5 mM palmitate (containing with 10 Ci of [1- 14C]palmitate) bound to albumin, 0.8 mM L-carnitine, and cells are incubated at 37°C for 2 h. Following the incubation, 400 l of benzothonium hydrochloride is added to the center well to collect released 14CO 2. Immediately, the reaction is stopped by adding 500 l of 7% perchloric acid to the cells. The flasks
218
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Schematic representation of fatty acid transport regulation for -oxidation in the mitochondrion. Oxidation of long chain fatty acids in the mitochondria is highly regulated by a transport system consisting of three proteins, CPT-1, CACT, and CPT-2. CPT-1, regulated physiologically by competitive malonyl-CoA inhibition, transesterifies fatty acids from CoA to carnitine. The acylcarnitines are then transported to CPT-2 via CACT. CPT-2 reverses the transesterification of fatty acids, from acylcarnitines back to acyl-CoA’s for fatty acid oxidation. High levels of malonyl-CoA generated during fatty acid synthesis inhibit CPT-1 preventing oxidation of newly synthesized fatty acids. Etomoxir is an irreversible inhibitor of CPT-1. Structural biological studies have found an association of CPT-1 with BCL-2 on the mitochondrial surface. Porin may also be associated with this fatty acid transporter complex.
with wells are then incubated for 2 h at 37°C after which the benzothonium hydrochloride is removed and counted for 14C. Blanks are prepared by adding 500 l of 7% perchloric acid to the cells prior to the incubation with the assay buffer for 2 h. Cell growth inhibition assay. MCF-7 cells are plated in 24-well plates at 5 ⫻ 10 4 cells per well in DMEM with 10% fetal bovine serum (Hyclone). After overnight incubation at 37°C, drugs were added from stock 5 mg/ml solutions in DMSO. The final concentration of DMSO in the cultures was at or below 0.2%. After either 48 or 72 h, medium was removed, and wells were washed thrice with Hank’s buffered saline. Wells were stained with crystal violet, dried, and solubilized in 10% SDS. One hundred microliter aliquots were transferred to a 96-well plate and read on a Molecular Dynamics plate reader at 490 nm. Clonogenic assay. After overnight incubation at 37°C, 1 ⫻ 10 6 MCF-7 cells were exposed to drugs as indicated for 6 h or vehicle control, washed, detached by trypsin digestion, counted, and plated at 1000 or 500 cells/60-mm plate in triplicate. Colonies were stained with crystal violet and counted 4 – 6 days after plating. Miscellaneous methods and chemicals. MCF-7 cells were obtained from the American Type Culture Collection. TOFA (5(tetradecyloxy)-2-furoic acid) was synthesized at the Johns Hopkins University, by Dr. Craig A. Townsend. Etomoxir was purchased from Dr. H.P.O. Wolfe. Cerulenin and other chemicals were obtained from Sigma. Etomoxir, cerulenin, and TOFA were added at the designated concentrations from a 5 mg/ml stock in DMSO. DMSO concentrations in the final cell culture medium never exceeded 0.1%. [1- 14C]palmitate and [1-methyl 14C] L-carnitine were obtained from Amersham Pharmacia Biotech.
Statistics. Results are expressed as means ⫾ standard error of the mean. Statistical analysis was performed using two-tailed t-tests in Prism 3.0 (Graph Pad).
RESULTS FAS and CPT-1 inhibition but not ACC inhibition leads to reduced FA oxidation. Since FAS inhibition leads to high levels of malonyl-CoA within 30 min, and malonyl-CoA is a potent inhibitor of FA oxidation, we tested whether cerulenin treatment would lead to FA oxidation inhibition in MCF-7 cells. Cerulenin is an irreversible inhibitor of both type I and type II fatty acid synthases (15, 16). After 2 h of cerulenin treatment at 10 g/ml, FA oxidation was reduced by 50% of controls (Fig. 3, upper panel). Importantly, this dose of cerulenin in MCF-7 cells is known to cause: [i] a ⬎50% decrease of fatty acid synthesis within 2 h; [ii] a 900% increase in malonyl-CoA within ⬍1 h; and [iii] ⬎90% cytotoxicity in clonogenic assays (6). Thus, FA oxidation inhibition occurs during fatty acid synthesis inhibition but before the onset of cytotoxicity. Etomoxir, an irreversible inhibitor of CPT-1 (17), also reduced fatty acid oxidation (Fig. 3, middle panel). TOFA, a reversible inhibitor of ACC, had a trend toward stimulation
219
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
of fatty acid oxidation at 5 g/ml that may be secondary to reduced malonyl-CoA levels (Fig. 3, lower panel). Cerulenin induced FA oxidation inhibition is likely mediated by malonyl-CoA elevation. FA oxidation inhibition by cerulenin could be due to direct inhibition of CPT-1 by cerulenin or mediated through increased malonyl-CoA. To discriminate between these possibilities, CPT-1 activity was measured directly in digitonin permeabilized cells in the presence of cerulenin. Cerulenin treatment at 10 g/ml had no direct inhibitory effect on CPT-1 activity in MCF-7 cells (Fig. 4, upper panel) although the same dose caused a 50% reduction in fatty acid oxidation (Fig. 3, upper panel). Exogenously administered malonyl-CoA [20 mM], a known reversible inhibitor of CPT-1 (18), caused an approximately 40% reduction in CPT-1 activity. CPT-1 in MCF-7 cells is inhibitable by malonyl-CoA, but not directly by cerulenin implying that cerulenin inhibits FA oxidation through malonyl-CoA signaling. The K i of malonyl-CoA inhibition of CPT-1 in MCF-7 cells was ⬃20 –30 uM (data not shown). This is consistent with expression of the liver isoform of CPT-1 in MCF-7 cells (19). Etomoxir showed a significant reduction in CPT-1 activity (Fig. 4, middle panel), with TOFA having no significant effect (Fig. 4, lower panel). FA oxidation inhibition is not the sole cause of FAS induced cytotoxicity in cancer cells. To test if FA oxidation inhibition is the primary mechanism inducing cytotoxicity during FAS inhibition, MCF-7 cells were treated with etomoxir at a series of doses (6.25–200 g/ml) that encompassed at least a 50% inhibition of FA oxidation in MCF-7 cells. Etomoxir showed significant growth inhibition only at 200 g/ml after 72 h; lower doses failed to cause significant growth inhibition (Fig. 5). At doses of 200 g/ml or greater, etomoxir looses its specificity as a CPT-1 inhibitor, and is reported to also inhibit CPT-2 (17, 20). Combined FA synthesis and FA oxidation inhibition is cytotoxic to MCF-7 cells. Combined TOFA and etomoxir treatment of MCF-7 cells should produce similar metabolic conditions as cerulenin treatment, and thus should show similar cytotoxicity. Figure 6 shows a clonogenic assay of MCF-7 cells treated with etomoxir alone, TOFA alone, and TOFA and etomoxir combined. TOFA and etomoxir alone had no cytotoxic effect. The combination of TOFA and etomoxir caused about a 50% reduction in clonogenicity
FIG. 3. FAS and CPT-1 inhibition, but not ACC inhibition, reduce fatty acid oxidation in MCF-7 cells. Upper panel, FAS inhibition with cerulenin (10 g/ml) in MCF-7 cells for 1 h resulted in a 50% reduction in fatty acid oxidation (P ⫽ 0.0007, two-tailed t-test).
Middle panel, CPT-1 inhibition with etomoxir (25 g/ml) for 1 h led to an approximately 30% reduction in fatty acid oxidation (P ⫽ 0.03, two-tailed t-test) and a 70% reduction at 50 g/ml (P ⫽ 0.01, two-tailed t-test). Lower panel, TOFA inhibition of ACC has no significant effect on fatty acid oxidation. Error bars represent standard error of the mean in duplicate samples.
220
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DISCUSSION Fatty acid synthesis and fatty acid oxidation are key anabolic and catabolic pathways linked through their common metabolic intermediate, malonyl-CoA. Thus, it is not surprising that rapid and profound alterations in malonyl-CoA levels and CPT-1 activity during FAS inhibition could lead to metabolic chaos in a cancer cell that already exhibits abnormal fatty acid metabolism. The in vitro cytotoxic effects of cerulenin on human cancer cells displaying high levels of fatty acid synthesis, such as MCF-7 cells, have been well established. Using DNA laddering (5) or merocyanine staining (6) as a markers of apoptosis, cerulenin treatment consistently led to apoptosis within 16 –24 h at micromolar concentrations. Apoptosis was preceded by inhibition of fatty acid synthesis by at least 50% within 2 h and a ninefold increase in malonyl-CoA. Pretreatment of the cells with TOFA, inhibited ACC and malonyl-CoA production rescuing the cells from cerulenin induced apoptosis and implicating malonyl-CoA as a trigger for apoptosis in cancer cells. In contrast, human cancer cells with low levels of fatty acid synthesis, such as RKO human colon cancer cells, were resistant to apoptosis following cerulenin treatment. Instead of apoptosis, RKO cells treated with cerulenin underwent a biphasic growth arrest, first during S phase, then again in G 1 and G 2 with accumulation of p53 and p21 (21). Interestingly, RKO cells with mutant p53 exhibited a brisk apoptotic response to cerulenin, similar to cells with high levels of fatty acid synthesis, including the ability to be blocked with TOFA pretreatment. These data indicate that the apoptotic response to cerulenin is dependent upon high levels of fatty acid synthesis and/or p53 mutation. Moreover, malonyl-CoA accumulation appears central to both apoptotic mechanisms but not to growth arrest.
FIG. 4. CPT-1 activity in permeabilized MCF-7 cells treated with malonyl-CoA or inhibitors of FAS, ACC, and CPT-1. Upper panel, cerulenin at 10 g/ml, a dose that reduced fatty acid oxidation by a 50%, had no significant effect on CPT-1 activity. In contrast, malonyl-CoA at 20 M caused a 75% reduction in CPT-1 activity (P ⫽ 0.007, two-tailed t-test). Middle panel, etomoxir, a CPT-1 inhibitor, reduced CPT-1 activity by ⬎80% at 10 g/ml (P ⫽ 0.02, two-tailed t-test). Lower panel, TOFA, an ACC inhibitor had no significant effect on CPT-1 activity. Error bars represent standard error of the mean with duplicate samples.
constituting a significant cytotoxic response. Interestingly, the combination of cerulenin and etomoxir showed no additive effect over cerulenin alone in growth inhibition studies (data not shown). Since cerulenin already inhibits CPT-1 via increased malonyl-CoA levels, the addition of another CPT-1 inhibitor did not potentiate, further supporting the proposed mechanism of action of cerulenin.
FIG. 5. Etomoxir does not reduce the growth of MCF-7 cells in vitro. Only the 200 g/ml dose of etomoxir had a significant growth inhibitory effect on MCF-7 cells (P ⫽ 0.006, two-tailed t-test). At 50 g/ml, etomoxir caused a ⬎50% reduction in fatty acid oxidation, similar to cerulenin. Error bars represent standard error of the mean with quadruplicate samples.
221
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 6. Combined etomoxir and TOFA treatment is cytotoxic to MCF-7 cells in vitro. Cells were treated with etomoxir 100 g/ml, that causes ⬎50% inhibition of fatty acid oxidation, and with TOFA 5 g/ml that inhibit fatty acid synthesis similar to cerulenin. Alone, these inhibitors have no significant effect on MCF-7 clonogenicity. Combined, however, they produce a 50% reduction in clonogenicity compared to control and are significantly more cytotoxic that either TOFA (P ⫽ 0.004, two-tailed t-test) or etomoxir (P ⫽ 0.002, two-tailed t-test) alone. Error bars represent standard error of the mean with triplicate samples.
CPT-1 inhibition has also been implicated in apoptosis in cancer cells. During an investigation of genes involved in cell death of mouse hematopoetic cells, CPT-1 emerged as candidate gene involved in apoptosis. In mouse LyD9 and WEHI-231 cells, etomoxir inhibition of CPT-1 augmented palmitate induced apoptosis (22). Moreover, CPT-1 co-immunoprecipitated with BCL-2 suggesting a physical association between CPT-1 and the mitochondrial apoptotic apparatus (23). These studies all point to CPT-1 as potentially involved potentiating cell death. In our studies as with others, inhibition of CPT-1 alone was insufficient to induce apoptosis (22). Etomoxir treatment leading to the same magnitude of fatty acid oxidation and CPT-1 inhibition as cerulenin failed to growth inhibit or kill MCF-7 cells. However, etomoxir potentiated the anti-tumor effect of inhibition of fatty acid synthesis at the ACC step with TOFA leading to cell death. The combination of fatty acid oxidation inhibition and fatty acid synthesis inhibition is non-physiologic, mimicked the metabolic consequences of FAS inhibition by cerulenin, and was cytotoxic to cancer cells. Interestingly, in the prior studies with mouse hematopoetic cells, etomoxir enhanced the apoptosis induced by palmitate administration which is similar to our combination of etomoxir and TOFA. Palmitate administration would lead to a rapid increase in intracellular palmitoyl-CoA through the activity of acyl-CoA synthase. Like TOFA, palmitoyl-CoA is a potent inhibitor of ACC activity. Palmitoyl-CoA inhibition of ACC
would lead to decreased malonyl-CoA and decreased fatty acid synthesis; this by itself is not cytotoxic. However, in the presence of etomoxir, palmitoyl-CoA inhibition of ACC could cause the same non-physiological state of reduced fatty acid oxidation and fatty acid synthesis as TOFA and etomoxir. Consequently, the combination of palmitate inhibition and etomoxir would have similar metabolic consequences to TOFA and etomoxir, or cerulenin alone—all cytotoxic to cancer cells. Cerulenin induced apoptosis in human cancer cells is due to its simultaneous inhibition of fatty acid synthesis at the FAS step with subsequent elevation of malonyl-CoA, CPT-1 inhibition, and inhibition of fatty acid oxidation. This nonphysiological biochemical state, well tolerated by normal cells both in vitro and in vivo, is rapidly cytotoxic to many types of cancer cells. Our study indicates that CPT-1 inhibitors, such as etomoxir, which have been used in humans for diabetes control (24), may have anti-cancer effects if combined with an inhibitor of ACC. These findings may ultimately provide new strategies for cancer chemotherapy based upon combinations of metabolic inhibitors targeting fatty acid synthesis and oxidation. ACKNOWLEDGMENTS This work was supported by National Institute of Health Grant RO1 CA87850-02. We thank Dr. Townsend for his gift of TOFA for these studies.
222
Vol. 285, No. 2, 2001
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES 1. Kuhajda, F. P. (2000) Nutrition 16, 202–208. 2. Wakil, S. (1989) Biochemistry 28, 4523– 4530. 3. Kuhajda, F. P., Jenner, K., Wood, F. D., Hennigar, R. A., Jacobs, L. B., Dick, J. D., and Pasternack, G. R. (1994) Proc. Natl. Acad. Sci. USA 91, 6379 – 6383. 4. Pizer, E. S., Wood, F. D., Heine, H. S., Romantsev, F. R., Pasternack, G. R., and Kuhajda, F. P. (1996) Cancer Res. 56, 1189 – 1193. 5. Pizer, E. S., Jackisch, C., Wood, F. D., Pasternack, G. R., Davidson, N. E., and Kuhajda, F. (1996) Cancer Res. 56, 2745–2747. 6. Pizer, E. S., Thupari, J., Han, W. F., Pinn, M. L., Chrest, F. J., Frehywot, G. L., Townsend, C. A., and Kuhajda, F. P. (2000) Cancer Res. 60, 213–218. 7. Pizer, E. S., Chrest, F. J., DiGiuseppe, J. A., and Han, W. F. (1998) Cancer Res. 58, 4611– 4615. 8. Furuya, Y., Akimoto, S., Yasada, K., and Ito, H. (1997) Anticancer Res. 17, 4589 – 4593. 9. Pizer, E. S., Pflug, B. R., Bova, G. S., Han, W. F., Udan, M. S., and Nelson, J. B. (2000) Proc. Amer. Assoc. Cancer Res. 41, 655. 10. Loftus, T. M., Jaworsky, D. E., Frehywot, G. L., Townsend, C. A., Ronnett, G. V., Lane, M. D., and Kuhajda, F. P. (2000) Science 288, 2379 –2381. 11. Halvorson, D. L., and McCune, S. A. (1984) Lipids 19(11), 851– 856.
12. McGarry, J. D., and Brown, N. F. (1997) Eur. J. Biochem. 244, 1–14. 13. Sleboda, J., Risan, K. A., Spydevold, O., and Bremer, J. (1999) Biochim.Biophys. Acta 1436, 541–549. 14. Antinozzi, P. A., Segall, L., Prentki, M., McGarry, J. D., and Newgard, C. B. (1998) J. Biol. Chem. 273, 16146 –16154. 15. Funabashi, H., Iwaski, S., Okuda, S., and Omura, S. (1983) Tetrahedron. Lett. 24, 2673–2676. 16. Funabashi, H., Kawaguchi, A., Tomoda, H., Omura, S., Okuda, S., and Iwasaki, S. (1989) J. Biochem. 105, 751–755. 17. Lilly, K., Chung, C., Kerner, J., VanRenterghem, R., and Bieber, L. L. (1992) Biochem. Pharmacol. 43, 353–361. 18. Saggerson, E. D., and Carpenter, C. A. (1981) FEBS Lett. 132, 166 –168. 19. Swanson, S. T., Foster, D. W., McGarry, J. D., and Brown, N. F. (1998) Biochem. J. 335, 513–519. 20. Lopaschuk, G. D., McNeil, G. F., and McVeigh, J. J. (1989) Mol. Cell. Biochem. 88, 175–179. 21. Li, J., Gorospe, M., Chrest, F. J., T. S., K., Evans, M. K., Han, W. F., and Pizer, E. S. (2001) Cancer Res. 61, 1493–1499. 22. Paumen, M. B., Ishida, Y., Muramatsu, M., Yamamoto, M., and Honjo, T. (1997) J. Biol. Chem. 272, 3324 –3329 23. Paumen, M. B., Ishida, Y., Han, H., Muramatsu, M., Eguchi, Y., Tsujimoto, Y., and Honjo, T. (1997) Biochem. Biophys. Res. Commun. 232, 523–525. 24. Schmidt-Schweda, S., and Holubarsch, C. (2000) Clin. Sci. 99, 27–35.
223