Introduction

Obesity has become one of the most serious health problems in both developed and developing countries, which is related to various diseases including diabetes mellitus, coronary heart disease, sleep-breathing disorders, and certain forms of cancer [1]. Obesity is characterized by an increase adipose tissue mass which is decided by both the number and the size of adipocytes. The increase in the number and the size of adipocytes depends on the differentiation from preadipocytes and the lipid accumulation. Fatty acid synthase (FAS; EC 2.3.1.85) is a critical metabolic enzyme for lipogenesis, and is highly expressed in liver and adipose tissue. It catalyzes the synthesis of saturated fatty acids in cells [2]. Recently, FAS has been considered as an anti-obesity target, because it not only supplies metabolic substrates, but may play an important role in feeding behavior [3]. Intracerebroventricular treatment with FAS inhibitor reduces food intake and body weight in ob/ob mice, and it suggests that FAS may be a potential therapeutic target for obesity and may represent an important link in feeding regulation [4, 5]. Moreover, it has been demonstrated that inhibition of FAS by classic inhibitors, including C75, or siRNA, significantly suppresses the differentiation and lipid accumulation in 3T3–L1 cells, suggesting that FAS is an active and essential participant to maintain the differentiation of preadipocytes [6]. Therefore, FAS inhibitors may be able to reduce body weight by directly acting on adipose tissue.

Curcumin, a diarylheptanoid compound, exists in the spice turmeric. It has been reported to exhibit multitudinous biological activities including antioxidant, anti-inflammatory, antimicrobial, and anticancer activities, and it may have the potential to improve curative effect on diabetes, allergies, arthritis, Alzheimer’s disease, and other various malignant or chronic diseases. These activities are mediated via the regulation of various molecular targets including a lot of enzymes [7]. In addition, it is reported that supplementing high-fat diet of rat with curcumin avoids lipid accumulation in liver and reduces epididymal adipose tissue mass [8], and that curcumin suppresses the adipogenesis and differentiation in adipocytes, and down-regulates PPARγ and C/EBPα. This is consistent with a recent finding that curcumin reduces body fat and lowers weight gain in obese mice but unaffect the food intake of these mice [9]. The anti-obese activity of curcumin is explained, at least in part, by the activation of AMPK that is crucial for the inhibition of the differentiation and growth of adipocytes [10].

Another study shows that curcumin significantly lowers fatty acid biosynthesis in the liver of hamster, one of the rodent species that are very similar to human in lipid metabolism [11], implying that the potential of curcumin in preventing obesity might be related to the inhibition of FAS. However, up to now, there are no studies reporting the direct effect of curcumin on FAS and as to how it affects adipocytes. In this study, we demonstrate, for the first time, that curcumin is a novel effective inhibitor of FAS, and describe its inhibitory characters. Moreover, we investigate the effects of curcumin on the different stages of differentiation in 3T3–L1 cells, and the results show that its effects on adipocytes are highly associated with its inhibition of FAS.

Materials and methods

Materials

Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum were purchased from Gibco BRL. 3-isobutyl-1-methylxanthine, insulin, dexamethasone, oil red O, acetyl-CoA, malonyl-CoA, NADPH, and Curcumin were purchased from Sigma. C75 was provided by The Proctor & Gamble (P&G) Company. The primary antibody for FAS immunoblotting was a mouse anti-FAS monoclonal antibody obtained from BD Biosciences Pharmingen, and horseradish peroxidase-conjugated secondary antibodies were supplied by Santa Cruz. The Trizol reagent and the Superscript First-Strand Synthesis System were obtained from Invitrogen Corp. All other reagents were local products with purity of analytic grade.

Cell culture

3T3–L1 preadipocytes were obtained from Les Laboratoires Servier (France) and were used at passage 8–15. Cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) at 37°C in the presence of 5% CO2. Medium was changed every 2 days. 3T3–L1 preadipocytes were seeded in the plate and grown for 2–4 days for differentiation. Two days after reaching confluence, the medium was changed to DMEM containing 10% FBS supplemented with 0.5 mM 3-isobutyl-1-methylxanthine, 1 μM dexamethasone, and 1.7 μM insulin (day 0). The cells were treated for 2 days (day 2), and then were cultured in DMEM containing 10% FBS and 1.7 μM insulin for another 2 days. Thereafter (day 4), the cells were cultured in DMEM containing 10% FBS up to day 8, and the medium was changed every 2 days. During the differentiation, cells were treated with curcumin or C75 at the concentration and duration indicated in each result, and fresh inhibitor was added whenever a medium change was performed unless stated otherwise.

Cell viability assay

Tests were performed in 96-well plates. Cells were cultured in the plates until confluence; thereafter cells were incubated with either DMSO (1:1000) or increasing concentrations of curcumin for 48 h. The medium was then changed to a fresh one within 1 mg/ml MTT (3-4, 5-dimethylthiazol-2-yl-2, 3-diphenyl tetrazolium bromide). After 4-h incubation at 37°C, the plates were again decanted, and 200 μl of DMSO was added to solubilize the formazan crystals present in viable cells. The plate was analyzed by spectrometry at the wavelength of 490 nm. The wells containing no cells served as a background for the assay. Data were obtained from the average of five experiment wells, and the assay was repeated twice.

Oil red O staining

Intracellular lipid accumulation was determined by oil red O staining at day 8 after adipocyte differentiation. The cells were washed twice with phosphate-buffered saline, and stained with 0.3% (w/v) oil red O solution in 60% (v/v) isopropanol for 1 h. After staining, the cells were washed three times with water to remove excess stain. Stained oil droplets in the cell were dissolved in isopropanol, and spectrophotometrically measured at an absorbance of 520 nm.

Immunoblotting

Cells were washed twice with ice-cold PBS, and then lysed in RIPA lysis buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% (v/v) Nonidet-P40, 0.5% (w/v) sodium deoxycholate, 0.1% (w/v) SDS, 1 mM Na3VO4, 10 μg/ml leupeptin, 50 mM NaF, and 1 mM phenylmethylsulfonyl fluoride) for 30 min on ice. The lysates were cleared by centrifugation in an Eppendorf tube at 18,000g for 30 min at 4°C. Protein content in the supernatant was determined using the BCA protein assay kit, and equal amounts of protein from whole cell lysates were separated by SDS-PAGE and transferred onto a PVDF membrane. The membrane was blocked with 5% non-fat dried milk in TBST for 1 h, and then incubated with primary antibody in TBST containing 5% non-fat dried milk at 4°C over night. Immunoreactive bands were visualized with ECL detection reagent following incubation with horseradish peroxidase-conjugated secondary antibody in TBST for 1 h at the room temperature.

RNA isolation and RT-PCR analysis

Total RNA was extracted with Trizol reagent according to the manufacturer’s instructions and reverse-transcribed by two-step method with the SuperScript First-Strand Synthesis System. The synthesized single-stranded cDNA was used for amplification of a specific target. The mouse β-actin gene was amplified as a loading control. The forward and reverse primers were 5′-GCAAGACATAGACAAAACACCAGTGTGA-3′, 5′-AGCAACCATTGGGTCAGCTCTTGTGA-3′ for mPPARγ; 5′-TGTACCTGGGAGTTGGCGAG-3′, 5′-CTGCTGTTCTTTGCCACGTC-3′ for CD36; and 5′-CCCATCTACGAGGGCTAT-3′ for β-actin. The amplified products were visualized on 1% agarose gels.

Preparation of FAS and substrates

Duck FAS were used. The preparation, storage, and use of FAS were performed as described previously [12]. The final purified enzyme was homogeneous on SDS-PAGE. The enzyme and substrate concentrations were determined by absorption measurements using the extinction coefficients according to the method previously described [12].

Assay of FAS activity

The overall reaction and β-ketoacyl reduction of FAS were determined with an Amersham Pharmacia Ultrospec 4300 pro UV–Vis spectrophotometer at 37°C by following the decrease of NADPH at 340 nm, and the reaction mixture used for these reactions has been described previously [13]. In brief, the assay solution for the overall reaction contained 100 mM potassium phosphate buffer, pH 7.0, 1 mM EDTA, 1 mM dithiolthreitol, 3 μM acetyl-CoA, 10 μM malonyl-CoA, 35 μM NADPH, and 10 μg of FAS in a total volume of 2.0 ml. The β-ketoacyl reduction reaction mixture (2 ml) contained 40 mM ethyl acetoacetate, 35 μM NADPH, 1 mM EDTA, 1 mM dithiolthreitol, and 10 μg of FAS in 10 mM phosphate buffer, pH 7.0.

Cell FAS activity assay

FAS activity in cells was assessed as described previously with some modifications [14]. In brief, cells in 6-well plate were washed with cold PBS and then collected. After ultrasonic disrupted in assay buffer (contained 100 mM potassium phosphate buffer, 1 mM EDTA, and 1 mM dithiolthreitol, pH 7.0) with 0.6 mM PMSF on ice and centrifuged 18,000g for 30 min at 4°C, the supernatant was used to assay as described in Sect. “Assay of FAS activity”. The assay solution without malonyl-CoA served as a background for the assay.

Inhibition studies of curcumin

Fast-binding inhibition is usually reversible and caused by the non-covalent fast combination of inhibitor with enzyme. The fast-binding inhibition was determined by adding curcumin to the reaction system before the addition of FAS solution to initiate the reaction in a total volume of 2 ml. The curcumin was dissolved in dimethyl sulfoxide (DMSO) and added to the reaction mixtures described above. The final concentration of DMSO was under 0.5% (v/v), to avoid the interference with FAS activity [15]. The control activity of FAS in the absence of inhibitor was assigned as A0, and the activity of FAS in the presence of inhibitor was assigned as Ai, and the Ai/A0 was the relative activity. The extent of inhibition by the inhibitors was valued by the half inhibition concentration (IC50). The value of IC50 was obtained from a plot of relative activity versus inhibitor concentration.

Slow-binding inhibition is usually irreversible and caused by the covalent modification of enzyme by inhibitor. The FAS solutions were mixed with curcumin at room temperature and aliquots were taken to measure the remaining activity of FAS at the indicated time intervals to obtain the time-course curves.

Kinetics study of inhibition of FAS by curcumin

Possibility of competitive binding of inhibitor at each substrate-binding-site was determined by measuring the initial reaction rate when holding the concentration of the inhibitor at a constant value and increasing one substrate concentration (the other substrate concentrations were fixed).

Results

Inhibitory effect of curcumin on FAS activity

Curcumin (Fig. 1a) showed a dose-dependent fast-binding inhibition to FAS activity in vitro. It inhibited FAS activity with an IC50 value of 28.6 μM for the overall reaction and 59.1 μM for the β-ketoacyl reduction, respectively (Fig. 1b). The kinetically comparable IC50 values suggest that the fast-binding inhibition by curcumin is highly related to the β-ketoacyl reductase domain.

Fig. 1
figure 1

Fast-binding inhibition of FAS by curcumin. a The structure of curcumin. b The inhibition of FAS in the presence of various concentrations of curcumin was measured: Inhibition of the overall reaction (open circle), inhibition of the β-ketoacyl reduction (filled diamond). Data are the means values from three experiments. Lineweaver–Burk plots for inhibition of FAS by curcumin. c The overall reaction of FAS with Ac-CoA as the variable substrate. The concentrations of curcumin were 0 μM (filled circle), 2.5 μM (filled diamond), 4.375 μM (filled triangle), and 6.25 μM (filled square). d The overall reaction of FAS with Mal-CoA as the variable substrate. The concentrations of curcumin were 0 μM (filled circle), 5.0 μM (filled diamond), 10.0 μM (filled triangle), and 15.0 μM (filled square). e The β-ketoacyl reduction of FAS with NADPH as the variable substrate. The concentration of curcumin were 0 μM (filled circle), 20.0 μM (filled diamond), 35.0 μM (filled triangle), and 45.0 μM (filled square). Lines were derived by least-squares linear regression of the values

Full size image

To further elucidate the mechanism underlying the inhibition of FAS by curcumin, the inhibition kinetics was determined. The overall reaction activity of FAS was measured in the presence of increased concentration of curcumin and one variable substrate concentration (the other substrate concentrations were fixed). The Lineweaver–Burk plotted results showed that curcumin inhibited FAS overall reaction in a mixed manner, containing competitive and noncompetitive inhibition, against both acetyl-CoA (Fig. 1c) and malonyl-CoA (Fig. 1d). The β-ketoacyl reduction activity of FAS was measured in the presence of increased concentration of curcumin and variable NADPH concentration, and the concentration of ethyl acetoacetate was fixed. As shown in Fig. 1e, curcumin inhibited the β-ketoacyl reductase for FAS noncompetitively with respect to NADPH. The inhibition constants obtained from secondary plots are summarized in Table 1.

Table 1 Inhibition type and inhibition constants for the inhibiting of FAS by curcumin for each substrate
Full size table

Curcumin also showed slow-binding inhibition to FAS. The time-course semi-logarithmic plots indicated that curcumin inhibited FAS in an obvious time-dependent fashion at the concentrations of 10, 20, 35, and 50 μM (Fig. 2a). The pseudo-first-order rate constants, k obs, were obtained from the slope of each line in Fig. 2a. The plot of k obs versus [curcumin] was typically hyperbolic line that was replotted as the linear double-reciprocal plot of 1/k obs versus 1/[curcumin] in the inset (Fig. 2b). This result indicated that the inhibition of FAS by curcumin followed the saturated kinetics with a two-step irreversible inhibition mechanism, a rapid equilibrium step to form a reversible enzyme–inhibitor complex (E–I) followed by a slower, irreversible inactivation step as follows [13]:

Fig. 2
figure 2

Time-dependent inhibition of FAS and the effect on the observed first-order rate constant (k obs) with the increasing concentrations of curcumin. a Semi-logarithm plot of the experimental data. R.A. is the residual FAS activity of the overall reaction. The observed first-order rate constants (k obs) were obtained from the linear slopes of the plot. The FAS activities were measured at the indicated time after the addition of various concentrations of curcumin to the FAS (0.8 μM) solutions. The concentrations of curcumin were 0, 10.0 μM (filled diamond), 20.0 μM (filled triangle), 35.0 μM (filled square), and 50.0 μM (open circle). b k obs with the different concentrations of curcumin. The inset is the plot of 1/k obs versus 1/[curcumin]. The dissociation constant, K s, 127.1 μM, and first-order rate constant k, 0.0109 min−1, were calculated from the plot. Lines were derived by least-squares linear regression of the values

Full size image
$$ {\text{E}} + {\text{I}}\overset {Ks} \longleftrightarrow {\text{E}} - {\text{I}}\mathop{\longrightarrow}\limits^{k}{\text{EI}} $$

In which K s is the dissociation constant of the E–I complex at the first step, and k is the rate constant at the second step. The kinetics formula is shown as Eq. 1, and the double-reciprocal formula as in Eq. 2, and, thus, the 1/k obs versus 1/[I] plot is linear.

$$ k_{\text{obs}} = k\left[ {\text{I}} \right]\,/\,\left( {K_{\text{s}} + \,\left[ {\text{I}} \right]} \right) $$
(1)
$$ 1\,/k_{\text{obs}} = K_{\text{s}} /\,\left( {k\left[ {\text{I}} \right]} \right)\, + \, 1\,/k $$
(2)

Ks of 127.1 μM and k of 0.0109 min−1 were calculated from the data in Fig. 2b. Compared with another natural FAS inhibitor, EGCG that inhibited FAS with Ks of 352 μM and k of 0.0168 min−1 [13], curcumin had an apparently lower Ks value but a closer k value.

Suppressive effects of curcumin on lipid accumulation of 3T3–L1 cells

To ensure that the doses of curcumin used were not generally cytotoxic, the concentration-dependent toxic effect of curcumin on confluent undifferentiated 3T3–L1 cells were measured. The results in Fig. 3a showed that curcumin did not affect cell viability up to 25 μM. C75 was previously reported to be non-cytotoxic up to 100 μM in 3T3–L1 cells [6]. Therefore, the concentrations of curcumin and C75 used in this study were not more than 25 and 100 μM, respectively.

Fig. 3
figure 3

Effect of curcumin on 3T3–L1 cell viability and on morphology of the cells during differentiation. a 3T3–L1 cells were incubated with curcumin for 48 h at the concentrations shown. Subsequently, a MTT Cell Viability assay was performed as described under Sect. “Materials and methods”. Data are means ± SD; n = 3. b Cell culture was performed as described in Sect. “Materials and methods”. Photos of differentiated cells were taken at day 8 after oil red O staining

Full size image

During differentiation, 3T3–L1 cells developed visible lipid droplets with staining of oil red O. As shown in Fig. 3b, without treatment, 3T3–L1 cells indicated obvious lipid droplets after differentiation, but 20 μM of curcumin nearly completely blocked the production of lipid droplets. The presence of curcumin also inhibited the morphological changes that were observed in the control differentiated cells. Cells treated with 20 μM of curcumin were morphologically similar to the undifferentiated cells.

Suppressive effect of curcumin on FAS in 3T3–L1 cells during differentiation

Lipid accumulation was measured quantitatively after the lipid droplets were stained with oil red O. As shown in Fig. 4a, 10, 15, 20, and 25 μM of curcumin reduced lipid accumulation during differentiation up to 72, 60, 45, and 34%, respectively, compared with the control differentiated cells, suggesting a dose-dependent inhibitory effect of curcumin on lipid accumulation in adipocytes.

Fig. 4
figure 4

Effect of curcumin during differentiation on lipid accumulation and FAS activity. Bars represent means ± SD. Undiff undifferentiated cells, Diff differentiated cells. Data were normalized to control differentiated cells in experiments conducted with curcumin. P values were obtained using a two-tailed t test. *P < 0.05, **P < 0.01, ***P < 0.001 compared to control differentiated. a On day 8, stained oil droplets in the cell were dissolved in isopropanol, and spectrophotometrically measured at an absorbance of 520 nm. Curcumin inhibited 3T3–L1 intracellular triglyceride accumulation in a dose-dependent manner (n = 3). b Effect of curcumin during differentiation on FAS expression and activity. Cells were differentiated in the presence of 10 or 20 μM curcumin. Cells were collected at day 8 to be used in western blotting or FAS activity assay as described in Sect. “Materials and methods” (n = 3)

Full size image

We also measured the effects of curcumin on the expression of FAS in differentiated adipocytes. Compared with the undifferentiated cells, the differentiated adipocytes showed much higher level of FAS, which were significantly suppressed by curcumin. 20 μM of curcumin almost reduced the expression of FAS to the similar level as in the undifferentiated cells (Fig. 4b top). The increased activity of FAS in differentiated cells was also correspondingly repressed by curcumin (Fig. 4b bottom).

Effects of curcumin on the differentiation, and the mRNA levels of PPARγ and CD36 in 3T3–L1 cells

We further investigated the effects of curcumin on 3T3–L1 cells in the different stages during differentiation. 3T3–L1 cells were treated for 2 days with 20 μM of curcumin or 50 μM of C75 (a FAS inhibitor, as positive control) in three different time intervals: early (days 0–2), middle (days 2–4), and late (days 4–6) stages of differentiation, and then the intracellular lipid accumulation was measured on day 8 by oil red O staining. As shown in Fig. 5a, lipid accumulation was significantly reduced to 49 and 64% by the treatment of curcumin during the early and middle stages, respectively. However, C75 only suppressed lipid accumulation in the early stage (Fig. 5b). These results suggested that curcumin possibly affected the initiation and development of the differentiation of adipocytes during the early-to-middle stages of differentiation, while C75 only suppressed the initiation of the differentiation.

Fig. 5
figure 5

Effect of curcumin and C75 during different stages of 3T3–L1 cell differentiation. Cells were stained with oil red O and measured on day 8. Bars represent means ± SD, n = 3. Undiff undifferentiated cells, Diff differentiated cells. Early means the cells were treated during days 0–2; Middle means days 2–4; and Late means days 4–6. Data were normalized to control differentiated cells in experiments conducted with curcumin or C75. ***P < 0.001 compared to control differentiated. a Effect of 20 μM curcumin on different stages of the differentiation. b Effect of 50 μM C75 on different stages of the differentiation

Full size image

To further explore the mechanism underlying the suppression of adipocyte differentiation by curcumin, the mRNA level of PPARγ and CD36 were examined. The transcription factor PPARγ is regarded as one of the controllers of the differentiation process, and it up-regulated many genes involved in lipogenesis during the differentiation [16]. CD36 is a late marker of adipocyte differentiation and is regulated by PPARγ [17, 18]. As shown in Fig. 6a, the differentiated cells indeed showed significantly higher mRNA levels of PPARγ and CD36 than the undifferentiated adipocytes. An approximate 2.5-fold and twofold increase, respectively, in PPARγ and CD36 mRNA levels was induced in the differentiated adipocytes, which was significantly attenuated by 20 μM of curcumin (P < 0.05) (Fig. 6b, c).

Fig. 6
figure 6

Effect of curcumin during differentiation on PPARγ and CD36 mRNA. Cells were differentiated in the presence of 10 or 20 μM curcumin. On day 8, mRNA was extracted and quantified via RT-PCR, and normalized to actin mRNA. Data were normalized to control differentiated cells. a The gel of control samples of each gene, Undiff undifferentiated, Diff differentiated. b Change of PPARγ mRNA level. c Change of CD36 mRNA level. RT-PCR reactions of each sample were performed in three independent experiments, and three samples were chosen at random from each group. Relative mRNA levels are presented as means ± SD.*P < 0.05 compared to control differentiated; **P < 0.01 compared to control differentiated

Full size image

Discussion

Inhibition of fatty acid synthase reduces food intake and body weight, which suggests that FAS should be a reasonable target for the treatment of anti-obesity [4]. Therefore, more effective FAS inhibitors are needed for the treatment of obesity or health care.

In this article, we reported curcumin as a novel potent FAS inhibitor for the first time. Curcumin are a natural compound existing in the common-used spices, and it showed both fast- and slow-binding inhibitions to FAS. Slow-binding inhibition is usually a kind of irreversible inactivation. Our kinetic study showed that curcumin inhibited FAS in a manner of two-step irreversible inhibition that an irreversible inactivation followed a reversible binding of curcumin to FAS (Fig. 2). In the first step, the reversible specific binding of curcumin to FAS possibly directly led to the suppression of FAS activity. Indeed, the results in Fig. 1 indicated that curcumin showed a concentration-dependent fast-binding inhibition to FAS. Interestingly, curcumin also showed a kinetically comparable inhibition to the β-ketoacyl reduction of FAS, and this suggested that the β-ketoacyl reductase domain might be the main targets of curcumin. However, curcumin had a noncompetitive relationship with respect to NADPH, the substrate of β-ketoacyl reductase domain, and this suggested that curcumin only inhibited the β-ketoacyl reduction of the intermediate substrate of FAS while unaffected the binding of NADPH to the β-ketoacyl reductase domain. In addition, curcumin partially competitively bound to FAS with acetyl-CoA and malonyl-CoA, suggesting that curcumin possibly had a binding-site in the malonyl/acetyl transferase domain of FAS because acetyl-CoA and malonyl-CoA were the two substrates of this domain. Since FAS was a multifunctional enzyme, the product of one enzymatic domain was the substrate of next enzymatic domain of FAS, and thus this process had to keep the same pace in order to synthesize the final product, mainly palmitate. Based on all the above, we proposed a working model for the inhibition of FAS by curcumin: curcumin first bound to the malonyl/acetyl transferase domain, which affected the binding of both acetyl-CoA and malonyl-CoA and thus blocked the following reaction including the β-ketoacyl reduction of the intermediate substrate; following this fast-binding, curcumin then covalently modified FAS.

Compared with the known FAS inhibitors, EGCG and C75 [13], curcumin were generally more potent. Although C75 had a two-phase irreversible inhibition to FAS, it failed to exert a fast-binding inhibition to FAS. EGCG is a natural compound mainly existing in green tea, and it showed both fast-binding and slow-binding inhibition to FAS that were weaker than curcumin. C75 and EGCG already showed potentially applied value in the treatment of obesity [4, 19, 20], and so curcumin should be more promising in obesity prevention or treatment.

C75 and EGCG had been reported to inhibit adipocyte differentiation [6, 21]. Therefore, we tested the effect of curcumin on the differentiation of adipocytes. Curcumin showed a potent suppressive effect on the differentiation of adipocytes. As shown in Fig. 3, at the nontoxic concentration, 20 μM of curcumin nearly completely repressed the differentiation of 3T3–L1 cells and at the same time maintained the morphological shape similar to that of the undifferentiated cells. In contrast, the similar repressive effect on the differentiation of adipocytes was afforded, respectively, by 200 μM of EGCG and 50 μM of C75 [6, 21]. This manifested that curcumin was stronger in its ability to suppress the differentiation of adipocytes than EGCG and C75, corresponding to their FAS inhibitory ability. This prompted us to test the effect of curcumin on FAS activity in adipocytes. Surprisingly, we found that curcumin significantly suppressed the protein level of FAS that should be up-regulated during the differentiation of 3T3–L1 cells, as shown in Fig. 4. This indicated that curcumin could inhibit FAS in 3T3–L1 cells by reducing both protein level and enzymatic activity. The lipid accumulation, the indicator of differentiated adipocytes, was positively correlative to the FAS activity in these cells treated with different concentrations of curcumin, and this strongly suggested that curcumin repressed the differentiation of adipocytes by targeting FAS.

In addition, our results showed that curcumin was able to inhibit the differentiation of 3T3–L1 cells in the early and middle stages, while C75 only affected the differentiation in the early stage. These possibly suggested altered cellular pathways are required for the differentiation in this process. Curcumin could inhibit the initiation and development of the differentiation, while C75 only repressed its initiation. One possible explanation is that curcumin has multiple activities that are related to the inhibition of differentiation and adipogenesis [9, 10]. Anyway, it implied that curcumin was more effective in preventing lipid accumulation than was C75.

PPARγ and CD36 were reported to be required for the differentiation of adipocytes [16, 17, 22]. In this study, the differentiated cells also showed much higher level of PPARγ and CD36 mRNA (Fig. 6a). Interestingly, curcumin reduced the mRNA levels of PPARγ and CD36 of adipocytes during the differentiation. These results were consistent with those from another FAS inhibitor, C75 [6], suggesting that FAS inhibition possibly down-regulated the expression of PPARγ and CD36. This possibly also contributed to the inhibitory effects of FAS inhibitors on the lipid accumulation of adipocytes.

One of the possible mechanisms underlying the effect of curcumin is that curcumin down-regulates the PPARγ expression by inhibiting FAS activity. PPARγ is regarded as an early marker of preadipocyte differentiation, and it can increase FAS expression, while high level of FAS is a late marker of differentiation. There has been some evidence suggesting that FAS activity is required for both PPARγ expression and the differentiation of 3T3–L1 preadipocytes [6]. Palmitate, the main product of FAS, has been reported to increase the transcription of lipogenic genes, including FAS, in preadipocytes as a ligand of PPARγ [23–26]. Therefore, FAS and PPARγ can interact with each other. Based on this, usually FAS inhibitors finally decrease FAS expression via PPARγ, this is to say, FAS inhibitors are able to suppress both the activity and expression of FAS, as shown for curcumin in our study. Both the suppression of PPARγ expression and FAS activity can contribute to the prevention of lipid accumulation in 3T3–L1 cells.

Targeting FAS in adipocytes might represent a new approach to prevent obesity. Therefore, more safe and effective FAS inhibitors are expected to be developed for weight control. Some natural FAS inhibitors, such as EGCG, even have been used in practice. In this study, we demonstrate that curcumin is a novel FAS inhibitor whose structure is totally different from the existing inhibitors. Curcumin exerted a potent effect on the prevention of adipocyte differentiation and lipid accumulation, which was highly associated with its FAS inhibitory ability. Since curcumin widely exists in spices that are eatable, it has promising application potential in preventing obesity, and it may supply some useful idea and new clues in developing drugs in treatment of obesity.