4-Methylene-2-octyl-5-oxotetrahydrofuran-3-carboxylic Acid (C75), an Inhibitor of Fatty-acid Synthase, Suppresses the Mitochondrial Fatty Acid Synthesis Pathway and Impairs Mitochondrial Function*

Background: C75 is a fatty-acid synthase inhibitor and potential anticancer drug. Results: C75 treatment leads to mitochondrial dysfunction that is rescued by overexpression of β-ketoacyl-acyl carrier protein synthase or lipoic acid. Conclusion: The effect of C75 on mitochondria is caused by inhibition of β-ketoacyl-acyl carrier protein synthase. Significance: The mitochondrial fatty acid synthesis pathway plays an important role in mitochondrial function. 4-Methylene-2-octyl-5-oxotetrahydrofuran-3-carboxylic acid (C75) is a synthetic fatty-acid synthase (FASN) inhibitor with potential therapeutic effects in several cancer models. Human mitochondrial β-ketoacyl-acyl carrier protein synthase (HsmtKAS) is a key enzyme in the newly discovered mitochondrial fatty acid synthesis pathway that can produce the substrate for lipoic acid (LA) synthesis. HsmtKAS shares conserved catalytic domains with FASN, which are responsible for binding to C75. In our study, we explored the possible effect of C75 on HsmtKAS and mitochondrial function. C75 treatment decreased LA content, impaired mitochondrial function, increased reactive oxygen species content, and reduced cell viability. HsmtKAS but not FASN knockdown had an effect that was similar to C75 treatment. In addition, an LA supplement efficiently inhibited C75-induced mitochondrial dysfunction and oxidative stress. Overexpression of HsmtKAS showed cellular protection against low dose C75 addition, whereas there was no protective effect upon high dose C75 addition. In summary, the mitochondrial fatty acid synthesis pathway has a vital role in mitochondrial function. Besides FASN, C75 might also inhibit HsmtKAS, thereby reducing LA production, impairing mitochondrial function, and potentially having toxic effects. LA supplements sufficiently ameliorated the toxicity of C75, showing that a combination of C75 and LA may be a reliable cancer treatment.

been performed in yeast. The mtFAS II system is closely associated with the physiological and biochemical functions of the mitochondria, including mitochondrial fusion and fission, mitochondrial DNA replication, and the antioxidant system (7). A deficiency in any of the mtFAS II genes in yeast leads to RNA processing defects, loss of mitochondrial cytochromes a and b, and defects in cellular lipoic acid (LA) (11)(12)(13). The main biological function of the mtFAS II pathway is the production of the octanoic acid precursor for LA synthesis (12,14,15). LA is a potent antioxidant that improves mitochondrial function, reduces organ dysfunction, and provides beneficial effects for the prevention of several diseases, such as diabetes, cardiovascular disease, and liver disease (16). However, little is known about the mtFAS II genes in mammals. Recent research has shown that defects in mammalian mtFAS II genes result in mitochondrial dysfunction. Transgenic mice for mitochondrial malonyl-CoA acyl carrier protein transacylase have disrupted energy equilibrium and protein lipoylation (17). The overexpression of 2-enoyl thioester reductase causes myocardial dysfunction in mice (18).
Over the past few years, screening of FASN inhibitors to design antitumor drugs has ignored the effects on mtFAS II. Whether these FASN inhibitors have an impact on mtFAS II remains unclear. Human mitochondrial ␤-ketoacyl-acyl carrier protein synthase (HsmtKAS; OXSM) is the key enzyme of the mtFAS II pathway, catalyzing the chain-elongating reaction of the fatty acid synthesis cycle. The catalytic domains are quite conserved between FASN and HsmtKAS. The structure of HsmtKAS shows that it has a highly conserved malonyl-binding pocket for C75, which inhibits enzyme activity (19). These observations prompted the question whether C75 produces side effects in normal cells through HsmtKAS during cancer therapy.
In the present study, we investigated the underlying mechanisms of the toxic effects of C75 on the mtFAS II pathway and mitochondrial function in human embryonic kidney (HEK) 293T cells. In addition, we explored agents that effectively reduce toxic side effects on non-cancer cells that may contribute to future therapeutic cancer treatments.
Cell Culture-HEK293T cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum at 37°C in a 5% CO 2 atmosphere. The cells were cultured for up to 10 generations, and the medium was changed every 2 days. For C75 treatment, cells were initially treated with 10, 50, or 100 M C75 for 2, 6, 12, 24, and 48 h. For NAC treatment, cells were initially treated with 50 M C75 and 10 mM NAC for 2, 6, 12, 24, and 48 h. For the LA supplement experiment, cells were pretreated with 50 M C75 for 24 h after which the medium was discarded and replaced with new medium containing R-LA at 20 or 100 M for another 24 h. To further investigate whether LA would protect cells from C75-induced damage, HEK293T cells were treated with 50 M C75 for 24 h followed by the addition of 100 M R-LA for another 24 h.
Transfection-The transfections were performed using Lipofectamine 2000 according to the supplier's instructions. For the transfection of cells in 6-well plates, HEK293T cells were seeded at 6 ϫ 10 4 cells/well. Lipofectamine 2000 (5 l) was incubated in 250 l of serum-free medium for 5 min. An appropriate amount of siRNA (HsmtKAS siRNA, FASN siRNA, or a combination of both) and the Lipofectamine 2000/medium were combined and incubated for another 20 min. The final mixture was added to each well. After 4 -6 h, the medium in each well was exchanged with fresh HEPES-buffered DMEM medium. For transfection of cells in 96-well plates, HEK293T cells were seeded at 3 ϫ 10 3 cells/well. An appropriate amount of siRNA, 0.5 l of Lipofectamine 2000, and 25 l of medium were combined and then applied to cells as described above. For the HsmtKAS knockdown experiment, cells were transfected with HsmtKAS siRNA for 24 h followed by the addition of 100 M R-LA for another 24 h. For NAC experiment, cells were transfected with siRNA for 24 h followed by the addition of 10 mM NAC for another 24 h. For the HsmtKAS overexpression experiment, cells were transfected with the HsmtKAS overexpression construct (pcDNA3.1-HsmtKAS) and then treated with 50 or 150 M C75 for another 24 h.

JC-1 Assay for Mitochondrial Membrane Potential (MMP)-
The MMP was measured with JC-1 (5 mg/ml stock concentration), a lipophilic, cationic dye that exhibits potential-dependent accumulation in mitochondria indicated by a fluorescence mission shift from green to red. The red/green fluorescence intensity ratio reflects mitochondrial membrane potential. Cells were stained with JC-1 solution at a 1:1000 dilution for 30 min. Cells were rinsed twice with PBS after JC-1 staining and scanned with a microplate fluorometer (Fluoroskan Ascent, Thermo Fisher Scientific Inc.). The MMP was determined at an excitation wavelength of 485 nm and emission wavelengths of 538 and 585 nm to measure green and red JC-1 fluorescence, respectively. Each well was scanned by measuring the intensity of each of 25 squares (of 1-mm 2 area) arranged in a 5 ϫ 5 rectangular array. Data were analyzed with GraphPad Prism using semilog concentration-response analysis.
Cellular Reactive Oxygen Species (ROS) Determination-Cellular ROS were incubated with H 2 DCFDA for 30 min and assayed following the manufacturer's instructions. Cellular ROS were measured at an excitation wavelength of 485 nm and an emission wavelength of 538 mm using a microplate fluorometer (Fluoroskan Ascent, Thermo Fisher Scientific Inc.). The relative H 2 DCFDA fluorescence was normalized to the protein concentration.
Protein Carbonyl Detection-Protein carbonyls were detected by Western blot analysis using the Oxyblot protein oxidation detection kit (Cell Biolabs, San Diego, CA) according to the manufacturer's instructions. The carbonyl groups in the protein side chains were derivatized to 2,4-dinitrophenylhydrazone in a reaction with 2,4-dinitrophenylhydrazine. After the protein samples were incubated with 2,4-dinitrophenylhydrazine for 15-20 min, they were subjected to Western blot analysis. As a control, the same amount of protein for each sample was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and stained with Coomassie Brilliant Blue R-250. The quantification was calculated by a total densitometry of oxyblots over a total densitometry of bluestained gels.
Western Blot Analysis-Cells were suspended in Western blot and immunoprecipitation lysis buffer (Beyotime, Jiangsu, China). The lysates were incubated for 30 min on ice and then centrifuged at 13,000 ϫ g for 15 min at 4°C. The supernatants were collected, and their protein concentrations were measured using the BCA Protein Assay kit (Pierce 23225). The purification of nuclear and cytoplasmic proteins followed the manufacturer's instructions (Beyotime). Next, 20 g of each protein sample was separated by 10% SDS-PAGE and then transferred to a pure nitrocellulose membrane (PerkinElmer Life Sciences). The membranes were blocked with 5% nonfat milk for 1 h at room temperature; washed three times with TBS with Tween 20 for 15 min each; and finally incubated with anti-HsmtKAS, anti-lipoic acid, anti-Nrf2, anti-HO-1, anti-NQO1 (1:1000), or anti-␤-actin (1:5000) antibody at 4°C overnight. The membrane was then incubated with the appropriate anti-rabbit, anti-mouse, or anti-goat secondary antibody at room temperature for 1 h. Chemiluminescence detection was performed using an ECL Western blotting detection kit (Pierce).
Statistical Analysis-The data are shown as the mean Ϯ S.E. of at least three independent experiments. Statistical significance was evaluated using one-way analysis of variance followed by a post hoc test to analyze differences. Statistical significance was set at p Ͻ 0.05.

C75 Induces Mitochondrial Dysfunction in HEK293T Cells-
HEK293T cells were dose-dependently treated with C75 for 2, 6, 12, 24, or 48 h to evaluate its effects on mitochondrial function. The MMP is an essential factor for maintaining mitochondrial function and cellular viability. At 6 h, C75 at 50 or 100 M dose concentration began to induce MMP loss, but cell viability was not affected (Fig. 1B). After 12 h, a serious mitochondrial dysfunction was observed accompanied by increased ROS overproduction (Fig. 1C) and cell viability loss (Fig. 1A). Taken together, MMP loss was assumed to be an early response to C75 addition. A 50 M C75 treatment concentration was used as a toxicity dose for the following assays.
To further determine whether ROS was the major factor to induce cell death, the free radical scavenger NAC was applied in the study. As expected, NAC efficiently removed excess ROS (Fig. 1F). Meanwhile, both the MMP loss (Fig. 1E) and cell viability (Fig. 1D) decrease were inhibited by NAC, suggesting that ROS was a major contributor to C75-induced cell death.
R-LA Supplement Ameliorates C75 Toxicity-Our data showed that 100 M R-LA treatments significantly increased cell viability as well as the MMP after C75 challenge (Fig. 2, A  and B). ROS overproduction induced by C75, a major cause of mitochondrial dysfunction, was also efficiently eliminated by R-LA (Fig. 3A). R-LA also significantly attenuated the protein carbonyl levels that were generated by C75 (Fig. 3B). More importantly, Western blot analysis demonstrated that C75 mainly reduced mitochondrial complex I and spared other complexes. In contrast, the expression level of mitochondrial complex I recovered when R-LA was added (Fig. 3C).
R-LA Normalized the Phase II Antioxidant Enzyme System-Nrf2 is a transcription factor that binds to antioxidant response elements and regulates the antioxidant response (20,21). It is tethered in the cytoplasm by Keap1 protein under normal or unstressed conditions (22). Through the activation process, Nrf2 can translocate into the nucleus and activate transcription of target genes known as phase II enzymes such as HO-1 (23) and NQO1 (24). As shown in Fig. 4, A-C, the mRNA levels of the transcription factor Nrf2, NQO1, and HO-1 were induced to a statistically significant degree after C75 treatment, and these increased levels were returned to normal after R-LA supplementation. Similar observations were also confirmed by Western blot analysis (Fig. 4D). As the key regulator of phase II enzymes, Nrf2 nuclear translocation was increased by C75 and restored to a normal level by R-LA treatment (Fig. 4E).
HsmtKAS Knockdown Impairs Mitochondrial Function-To clarify whether FASN or HsmtKAS plays a role in mitochondrial dysfunction, we developed specific siRNAs for both FASN and HsmtKAS. As shown in Fig. 5A, FASN siRNA led to 80% knockdown of FASN without a significant effect on HsmtKAS mRNA and protein contents. HsmtKAS siRNA resulted in a 70% decrease of both mRNA and protein expression without significant effects on FASN (Fig. 5B). Interestingly, FASN knockdown had no significant effects on mitochondrial function in HEK293T cells, whereas HsmtKAS knockdown had toxic effects on mitochondrial function, including a decrease in cell viability (Fig. 5D) and MMP (Fig. 5E) and an increase in ROS production (Fig. 5C). However, no synergetic effects were observed when both FASN and HsmtKAS were down-regulated simultaneously.
We then applied NAC to further investigate which event is the major factor in HsmtKAS knockdown cells. 10 mM NAC efficiently scavenged overproduced ROS (Fig. 5F) and recovered MMP (Fig. 5H) and cell viability (Fig. 5G). These results indicated that excess ROS induced by HsmtKAS knockdown might contribute to oxidative stress in cell damage and death.
HsmtKAS Knockdown and C75 Treatment Reduces Protein Lipoylation-In addition to generating longer fatty acids, the mtFAS II pathway can produce the octanoyl-acyl carrier protein substrate for endogenous LA synthesis. To investigate the role of C75 on cellular LA synthesis, an anti-lipoic acid antibody was used in Western blot analysis to detect LA binding to two key mitochondrial enzymes, the E2 subunits of the pyruvate dehydrogenase complex (PDC) and ␣-ketoglutarate dehydrogenase (KDH). Both PDC and KDH use LA as a cofactor to form LA-PDC-E2 and LA-KDH-E2, respectively. The LA-PDC-E2   JUNE 13, 2014 • VOLUME 289 • NUMBER 24

JOURNAL OF BIOLOGICAL CHEMISTRY 17187
and LA-KDH-E2 were recognized based on their size: the predicted size of PDC-E2 was 63 kDa, and that of KDH-E2 was 50 kDa. Our results showed that knockdown of HsmtKAS led to a decrease in protein lipoylation (Fig. 6A) as LA association with PDC and KDH decreased to ϳ70%. C75 also affected protein lipoylation; obvious effects were evident after the 50 and 100 M treatments (Fig. 6B).

R-LA Attenuates Mitochondrial Dysfunction Induced by HsmtKAS
Knockdown-Data showed that R-LA could increase the mRNA level of HsmtKAS, especially at doses of 100 and 200 M (Fig. 7A). Consistently, protein expression was also increased by R-LA treatment (Fig. 7B). Meanwhile, HsmtKAS knockdown-induced ROS overproduction was normalized by R-LA (Fig. 7C). In addition, Western blot analysis revealed that the expression level of mitochondrial complex I was significantly decreased by HsmtKAS knockdown (Fig. 7D), which was similar to the results of C75 treatment. As expected, R-LA supplementation efficiently restored the expression level of complex I.
HsmtKAS Overexpression Protects Cells against C75-induced Damage-We developed a pcDNA3.1-HsmtKAS overexpression plasmid. After increasing the HsmtKAS expression level, MMP (Fig. 8A) and ROS (Fig. 8B) levels partially recovered. At 6 h, HsmtKAS protected cells against C75-induced MMP loss. For long term treatment, C75 caused more serious damage due to MMP loss and ROS overproduction, whereas HsmtKAS provided a protective effect on mitochondrial function. Meanwhile, mitochondrial complex I was also protected by HsmtKAS overexpression in C75-injured cells (Fig. 8C). In addition, the activation of Nrf2 and other phase II enzymes induced by C75 was normalized through HsmtKAS overexpression (Fig. 8C). It is interesting that C75 treatment could also decrease the protein expression of HsmtKAS (Fig. 8C), suggesting a regulation effect of C75 on protein expression besides working as an inhibitor. Moreover, overexpression of HsmtKAS did not provide protection against higher dose C75 treatment, which induced nearly 100% cell death (Fig. 8D).

DISCUSSION
The action of C75 is not attributed to a single enzyme because C75 was reported to inhibit FASN, stimulate carnitine palmitoyltransferase 1 (25), activate AMP-activated protein kinase (26,27), and induce peroxisome proliferator-activated receptor-␣ (28). In the present study, C75 dramatically produced excessive ROS in HEK293T cells. Similar ROS production occurred after HsmtKAS knockdown. The induced ROS might be produced by inhibition of HsmtKAS because HsmtKAS overexpression can attenuate ROS. The role of HsmtKAS is distinguished from FASN, which is another of the most important C75 targets. HEK293T cells treated with FASN siRNA had relatively unaffected cellular ROS content. These results suggest that the excessive ROS caused by C75 was mainly due to inhibition of HsmtKAS but not FASN. This finding was consistent with down-regulation of mitochondrial acyl carrier protein in HEK293T cells, which generates excessive ROS by compromising the mtFAS II pathway (29). The release of ROS is thought to occur in the mitochondria to regulate cellular signaling and impair biological macromolecules. As the major ROS producer, mitochondria are also a vulnerable target of ROS; mitochondrial dysfunction-induced ROS overproduction may further damage mitochondria to create a vicious cycle. In response to cellular oxidative stress, one of the antioxidant systems, phase II enzymes, is usually activated to counteract the oxidative stress and protect cell health. In the current study, the mRNA levels of phase II antioxidant enzymes were significantly increased by oxidative stress. It is well known that Nrf2 is a key regulator of phase II antioxidant enzyme expression (30,31). As shown by the results, Nrf2 nuclear translocation was signifi-cantly increased in C75-treated cells. Expression of other well known Nrf2 targets enzymes, NQO1 and HO-1, also followed this trend.
Mitochondria are the primary source of ROS. ROS production can be ascribed to different factors, such as the expression levels of mitochondrial complex I (32,33). Mitochondrial respiratory complexes are one of the important factors that affect cellular ROS abundance. Increasing evidence suggests that the mtFAS II pathway is essential for mitochondrial respiratory function (7,34). Interestingly, one of the mtFAS II genes, 3-hydroxymyristoyl-acyl carrier protein, is a component of the bovine mitochondrial complex I (35). The role of HsmtKAS in mitochondria was further proved in the present study. MMP is one critical factor for maintaining the mitochondrial respiratory chain that is used to assess mitochondrial function. The loss of MMP is associated with cell depletion. In the present study, both C75 and HsmtKAS knockdown significantly affected MMP loss. Importantly, C75 treatment compromised the expression of respiratory complex I. A lesser effect was observed for complex II, but complexes III-V were not affected. The same result occurred when HsmtKAS was knocked down. In contrast, HsmtKAS overexpression was shown to protect mitochondria against C75-induced damage. Therefore, it appears likely that excessive ROS levels induced by C75 are released because mitochondrial complex I is compromised. A similar phenomenon was observed during RNA interference experiments on the mitochondrial acyl carrier protein in which its knockdown caused an ϳ60% reduction in complex I activity.
The activity of complex II was decreased, but complexes III-V were not significantly altered (29). These events increase our understanding of the link between the mtFAS II genes and the respiratory chain. Defects in mtFAS II genes might mainly compromise complex I and consequently lead to excessive ROS generation and loss of MMP.
Many studies have established that deficiency in any of the yeast mtFAS II genes leads to a decrease in the endogenous LA content (7,36). Generally, endogenous LA covalently attaches to two key mitochondrial enzymes, PDC-E2 and KDH-E2, which participate in the oxidative decarboxylation of ␣-keto acids. Down-regulation of HsmtKAS reduces protein lipoylation, which is also observed after C75 treatment, supporting our assumption that C75 could target HsmtKAS in the mtFAS II pathway. Interestingly, R-LA supplementation was able to recover mitochondrial function and eliminated the oxidative response after C75 treatment or HsmtKAS knockdown. As a redox regulator, R-LA is a well known powerful mitochondrial antioxidant (37,38). In addition, we found that R-LA could activate HsmtKAS expression, suggesting that additional lipoic acid stimulated its own production, which is consistent with previous studies that exogenously administered lipoic acid increases lipoic acid synthase expression (39). Likewise, the effect of C75 treatment on HsmtKAS expression might due to decreased R-LA content. Therefore, we assumed that the protection by R-LA might be due to its antioxidant activity, indirect regulation of HsmtKAS, or possibly a combination of the two mechanisms. Many enzymes involved in lipoic acid synthesis FIGURE 5. HsmtKAS knockdown impairs mitochondrial function. HEK293T cells were transfected with FASN siRNA, and the efficiency of FASN and HsmtKAS knockdown was evaluated by real time PCR and Western blot analysis (A). Cells were transfected with HsmtKAS siRNA, and the efficiency of FASN and HsmtKAS knockdown was examined (B). ROS content was measured using H 2 DCFDA staining (C). Cell viability was analyzed by MTT assay (D). The MMP was determined using JC-1 staining (E). After transfection with HsmtKAS siRNA for 24 h, cells were incubated with 10 mM NAC for another 24 h and were examined for ROS content (F), cell viability (G), and MMP level (H). The predicted protein size is marked in the blot. Values are the means Ϯ S.E. (error bars) of at least three independent experiments. *, p Ͻ 0.05; **, p Ͻ 0.01 between the two connected groups.  have been reported; however, knowledge of lipoic acid regulation of those enzymes is limited, and further investigation is needed.
To better understand whether HsmtKAS is the major target of C75, HsmtKAS expression was manipulated in the cells. It is interesting that additional HsmtKAS is not sensitive to C75 like endogenous HsmtKAS and could protect cell survival against C75 toxicity. C75 is a derivative of cerulenin, which inhibited HsmtKAS with an IC 50 value of 300 M (40). In addition, C75 has been reported to inhibit purified human FASN with an IC 50 value of over 100 M (41). Taken together, we assumed the main reason for the non-sensitivity of additional HsmtKAS to C75 might be the high efficiency of HsmtKAS overexpression together with a high K m value of HsmtKAS for C75. Therefore, treatment with a high dose of C75 (150 M) was performed. As expected, C75 induced nearly 100% cell death, which was not prevented by HsmtKAS. However, lack of direct evidence regarding the K m value of HsmtKAS for C75 is a limitation in this study and requires further investigation.
With the experimental design and methodology used in the present study, we observed that C75 has an adverse effect on mitochondrial function, compromising mitochondrial complex I and protein lipoylation and generating excessive ROS. R-LA supplementation plays a protective role against C75-induced damage. These data reveal that HsmtKAS is a new potential target of C75 and a novel regulator of oxidative stress. More attention should be focused on mtFAS II genes during the identification and design of antitumor drugs, especially FASN inhibitors. A combination of R-LA and C75 may provide promising prospects for cancer therapy.