Antiadipogenic Effects of Different Molecular Forms of Conjugated Linoleic Acid on 3T3-L1 Cells: Comparison between Free Fatty Acid and Phosphatidylcholine Forms.

The antiadipogenic activity of conjugated linoleic acids (CLA) in the form of phosphatidylcholine-bound (CLA-PC) or free fatty acids (FFA; CLA-FFA) was evaluated using 3T3-L1 adipocytes. Phosphatidylcholine from soya (soy-PC) was used as the comparison of PC form. Both the lipid accumulation and activity of glycerol-3-phosphate dehydrogenase were measured to determine lipogenesis, whereas the glycerol content was measured to evaluate lipolysis. The CLA uptake also measured to find out the utilization of CLA by the cells. As a results, lipid accumulation in 3T3-L1 adipocytes was inhibited in a dose-dependent manner following treatment with CLA-PC (50-400 μM). Both CLA-PC and soy-PC significantly suppressed lipid accumulation compared with CLA-FFA, even though the amount of CLA in CLA-PC was a half than CLA-FFA. The CLA uptake of PC form was superior to FFA form, however, no difference was noted between CLA-PC and soy-PC. These forms exerted their antiadipogenic activity via the suppression of lipogenesis, and not by increasing lipolysis. Short-term treatment, especially in the middle stage of differentiation, was more effective than long-term treatment; especially for CLA-FFA. The antiadipogenic effect of CLA-PC was partially attributed to the chemical structure of the PC molecule. These results provide important information for the utilization of physiologically functional fatty acids and particularly CLA in the food and medical fields.

mature 3T3-L1 cells treated with CLA had reduced levels of intracellular triacylglycerol and smaller cell size than those treated with a similar linoleic acid.
Most studies on the antiobesity properties of CLA have been conducted using CLA in the form of free fatty acids FFA 3 . The CLA used in most studies were composed of CLA mixtures containing approximately equal amounts of trans-10, cis-12 t10,c12-CLA , and c9,t11-CLA isomers, or other specific isomers. It has been reported that t10,c12-CLA is superior as an antiadipogenic compound in CLA isomers 4 8 . Furthermore, Jeong et al. studied the effect of the molecular form of CLA 9 . They compared the antiobesity properties of CLA in the form of FFA, diglyceride CLA-DG , and triglyceride CLA-TG at concentrations ranging from 10 to 1000 µg/mL. Their results showed that all treatments reduced the proliferation of preconfluent 3T3-L1 cells in a dose-dependent manner, with CLA-DG exerting the strongest effect compared with CLA-FFA and CLA-TG.
Of note, CLA in the form of TG might have limited absorption capacity in the intestine due to their polarity. Due to the lower polarity, TAG does not get through the cellular membrane easily. So, its absorption involves hydrolysis of TAG to release free fatty acids and monoacylglycerol MAG in the intestine lumen 10 . While PC having both hydrophilic and hydrophobic components, bile salts are able to facilitate micelle formation with it, which makes its absorption easily 11 . Hence, coupling a CLA with a polar molecule such as phosphatidylcholine CLA-PC should provide favorable bioavailability 10 . Currently, PC has been widely used for the dissolution of local fat deposits, with many studies reporting a reduction in fat deposits by a subcutaneous injection of PC. In particular, PC has been reported to cause adipocyte lipolysis and apoptosis of adipocytes 12 15 .
To date, no comparative studies have been performed on the antiobesity properties of CLA in the FFA CLA-FFA and PC at the sn-2 position; CLA-PC forms. The objective of this study is to reveal the antiadipogenic effect of the molecular form of CLA FFA and PC on 3T3-L1 adipocytelike cells.

Materials
CLA-FFA is commercially available CLA Jarrow Formula TM , Los Angeles, USA . CLA-PC was prepared via phospholipase A 2 -catalized esterification as previously described 16 . PC from soya soy-PC and glycerophosphocholine GPC were purchased from H. Holstein Co., Ltd. Tokyo, Japan . Choline and linoleic acid LA were purchased from Tokyo Chemical Industry Co., Ltd. Tokyo, Japan . Since tocopherols did not give effect on cell growth and lipid accumulation in 3T3-L1 cells data not shown , the tocopherols in samples were removed using activated charcoal powder, as previously reported 16 . The fatty acid compositions of CLA-FFA, CLA-PC, and soy-PC were analyzed by gas chromatography as previously reported 16 and are summarized in Table 1. All solvents and other chemicals were of analytical grade.

Cell culture
The 3T3-L1 preadipocyte cell line was obtained from the Japanese Collection of Research Bioresources IF050416, Osaka, Japan . Cell culture and differentiation treatment followed the method described by Zebisch et al. 17 with slight modifications. Cells were maintained in fresh Dulbecco s modified Eagle s medium DMEM; Sigma Aldrich, St. Louis, MO containing 10 fetal bovine serum FBS; Biosera, Nuaile, France , as well as 100 U/mL penicillin and 100 µg/mL streptomycin maintenance media MM . Cells were cultured in 6 cm diameter culture dishes at 1.5 10 4 cells/5 mL at 37 in a humidified atmosphere of 95 air and 5 CO 2 . After reaching 80 confluence, cells were passaged.

Differentiation of 3T3-L1 cells
For adipocyte differentiation, cells were cultured in 12-well plates at 1.5 10 4 cells/mL/well Scheme 1a . The medium was replaced every 2-3 d. At day 6 after cell seeding, the medium was replaced with MM containing 0.5 mM 3-isobutyl-1-metylxantine IBMX and 1 µM dexamethasone initiation media IM , and cells were then incubated for another 2 d. After that, the medium was replaced with MM containing 10 µg/mL insulin FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan progression media PM and cells were further incubated for 2 d, after which the medium was replaced with MM. Mature adipocytes were identified 8 d after changing the medium to IM. In this experiment, cells were treated with our experimental samples CLA-FFA, CLA-PC, and soy-PC at the same time of changing the medium IM and PM at concentration 200 µM. All samples were diluted in dimethyl sulfoxide and added to the respective media at concentrations less than 0.1 . We employed two patterns of sample treatments in this study. In the first, treatment was carried out for 4 d through 2 additions Scheme 1a , whereas in the second, treatment was performed for 2 d with a single addition at the same time of changing the medium IM or PM Scheme 1b .

Oil red O staining
Oil red O staining and quantification of relative lipid accumulation in adipocytes were performed Scheme 1 . Briefly, on day 8 of the differentiation stage, mature adipocytes in 12-well plates were rinsed twice with Dulbecco s phosphate-buffered saline D-PBS . Cells were treated and fixed with 1 mL 10 formaldehyde for 20 min and then rinsed twice with D-PBS. Cells were then rinsed with 60 isopropyl alcohol IPA for 1 min. Subsequently, cells were treated with 3 mg/mL Oil red O solution in 60 IPA and fixed for 10-20 min. Cells were then washed with 60 IPA followed by washing with D-PBS. Oil droplet-stained adipocytes were observed under a microscope and lipids stained with Oil Red O pigment were extracted using 100 IPA 0.5 mL . The absorbance of the extract at 490 nm was measured using a microplate reader Varioskan, Flash, Thermo Fisher Scientific, USA . Relative lipid accumulation was determined as the percentage of extracted intracellular lipid content by comparing the treatment with the control.

CLA uptake
CLA uptake analysis was performed to observe CLA in the samples were consumed by the cells during the culture. The CLA content in the media treated with CLA-FFA or CLA-PC was measured before and after the culture. The media was recovered and the CLA in the media was extracted using mixture of chloroform-methanol-water 3:10:5, v/v . Then the chloroform layer was collected and evaporated. The extracted CLA was dissolved by ethanol and the absorbance was read at 233 nm using a microplate reader Varioskan, Flash, Thermo Fisher Scientific, USA . A standard curve of CLA-FFA and CLA-PC were prepared, respectively, and the equation functioned with the concentration of samples were determined CLA-FFA: y 0.0103x 0.2106 R 2 0.9993 ; CLA-PC: y 0.0046x 0.2052 R 2 0.9997 . The CLA uptake was determined as A0-A1, where A0 was CLA concentration detected before culture, and A1 was that after culture. Sample treatment was carried out during two days of differentiation stage, Early stage or Middle stage Scheme 1 .

Glycerol-3-phosphate dehydrogenase GPDH activity
On day 8 of the differentiation stage, the GPDH activity of cells was measured Scheme 1a . Cells were washed twice with 1 mL D-PBS, and then the collected cells were dissolved in 1 mL enzyme extraction solution followed by sonication. After sonication, the mixture was centrifuged at 12,800 g for 5 min at a temperature maintained at 4 . The supernatant was collected and used for analysis. The GPDH activity in cells was measured using a GPDH assay kit Cosmo Bio Co., Ltd., Tokyo, Japan according to the manufacturer s protocol. The protein content was deter-Scheme 1 a. Timeline of treatments during differentiation of 3T3-L1 adipocytes; sample was added twice. b. Timeline of treatments at different growth stages during differentiation of 3T3-L1 adipocytes; sample was added once at day 0 or day 2. MM: maintenance media, IM: initiation media, PM: progression media. mined as described by Lowry et al. 18 for the quantification of GPDH activity per mg of protein. Results were expressed as of the control.

Glycerol content
At day 8 of the differentiation stage, the glycerol content was measured in the media to determine the degree of lipolysis Scheme 1a . The glycerol released in media was measured using a glycerol colorimetric assay kit Cayman Chemical, Michigan, USA according to the manufacturer s protocol. Results were expressed as of the control.

Cell viability
Cell viability was assessed using the trypan blue exclusion assay Scheme 1b . On day 8 of differentiation, cells were rinsed twice with D-PBS, trypsinized, and centrifuged at 1,000 g for 5 min. The supernatant was discarded, and cells were dissolved in 2 mL D-PBS. The cell solution was mixed with trypan blue solution at a ratio of 1:1, and cells were counted using a hemocytometer chamber.

Statistical analysis
All experiments were performed in triplicate. Data were expressed as the mean value with standard deviation s.d. . A significant difference in each treatment was determined using the Scheffe s test p 0.01 .

Dose dependent inhibition of lipid accumulation in
3T3-L1 adipocytes by CLA-PC Figure 1 shows the relative lipid accumulation of 3T3-L1 adipocytes after treatment with various concentrations 50, 100, 200, and 400 µM of CLA-PC. We conducted this experiment to determine the dose dependency of CLA-PC on lipid accumulation and to determine the concentration of samples in the next experiment. The timeline for treatment of samples to the differentiation of 3T3-L1 cells was in accordance with Scheme 1a.
As a result, we noticed that CLA-PC suppressed lipid accumulation in a dose-dependent manner. This finding was in line with some studies showing that CLA-FFA inhibited lipid accumulation in 3T3-L1 adipocytes in a dose-dependent manner 4, 8 . We did not observe any improvement in the suppression of lipid accumulation at concentrations of CLA-PC 200 µM. Therefore, we used a concentration of 200 µM for all samples in the following experiments.

Lipid accumulation in 3T3-L1 adipocytes treated with
CLA-FFA, CLA-PC, or soy-PC To observe the effect of the different molecular forms of CLA, 3T3-L1 preadipocytes were treated with CLA-FFA or CLA-PC at the time of replacing PM and IM during cell culture Scheme 1a . We used soy-PC as a comparison of PC form for CLA-PC.
We noticed that treatment with CLA-PC significantly suppressed lipid accumulation in cells to a greater extent than treatment with CLA-FFA at the same concentration in molecular weight. As shown in Table 1, the amount of CLA in PC form was a half of FFA form. From this result, it was shown that the ability of PC form on suppressing lipid accumulation in 3T3-L1 adipocytes was superior to FFA form, even though the amount of CLA in CLA-PC was lower than CLA-FFA. However, we did not observe any significant difference between the CLA-PC and soy-PC treatments Fig.  2 . These observations were also supported by the morphological appearance of cells Fig. 3 . From these results, it was suggested that PC form could suppress lipid accumulation better than FFA form in 3T3-L1 adipocytes. Jung et al., reported that PC could promote lipolysis and induce apoptosis through TNFα-mediated pathway 15 . In addition, even though the concentration of CLA in CLA-PC is half of CLA-FFA, the uptake of CLA of PC form into 3T3-L1 cells was 20 higher than that of FFA form in both early and middle stage of treatment Fig. 4 . This difference of the bioavailability between FFA and PC form also might be a  reason that PC form shows superior activity.

Effect of lipogenesis and lipolysis on 3T3-L1 adipo-
cytes It is well known that the GPDH activity increases in adipocytes when they start to differentiate and accumulate lipids. Besides, the amount of glycerol released in the culture medium is a marker of lipolysis.
As shown in Fig. 5, the GPDH activity of adipocytes after treatment with CLA-PC and soy-PC was significantly lower than that after treatment with CLA-FFA; however, we did not observe any significant differences between the treatments with CLA-PC and soy-PC. These results suggested that the GPDH activity in 3T3-L1 cells was effectively suppressed by treatment with any of the PC forms. Furthermore, we found that the amount of glycerol in the culture medium did not differ among treatments Fig. 6 . A similar    finding was reported by Kim et al. 19 who found that there was no significant difference on the amount of glycerol released between CLA-FFA-treated and control 3T3-L1 adipocytes. Our results suggested that the inhibition of lipid accumulation in 3T3-L1 cells was not mediated by the lipolysis pathway but through the inhibition of lipogenesis via suppression of the GPDH activity.

Effect of different treatment timing on lipid accumulation and cell viability at of 3T3-L1 cells
To study the effect of the treatment on lipid accumulation at different stages of the differentiation of 3T3-L1 cells, cells were treated with CLA samples as indicated in Scheme 1b. Treatment of cells with each sample was performed only once at the early or middle stage of differentiation, and the lipid accumulation and cell viability of 3T3-L1 adipocytes were monitored. As shown in Fig. 7, at the early stage of differentiation, the lipid accumulation in the CLA-FFA-treated cells was lower than that in the CLA-PC and soy-PC groups Fig. 7a , and the same tendency was observed for cell viability Fig. 7b . At the middle stage of differentiation, there were no significant difference between the samples treated Fig. 7a . We found that the cell viability of 3T3-L1 adipocytes at the middle stage of differentiation was lower after treatment with CLA-FFA compared with CLA-PC and soy-PC treatments Fig. 7b .
This result suggested that the suppression of lipid accumulation by CLA in both forms was higher in the short compared with the long treatment period Figs. 2 and 7a . In addition, we noticed that treatment in the middle stage of differentiation was more effective than that in the early stage Fig. 7a . We also observed that CLA-PC was effective in reducing lipid accumulation in adipocytes, whereas CLA-FFA induced lipid accumulation but decreased the cell number Fig. 7b . This finding was in line with a previous report by Satory and Smith 5 that CLA could inhibit proliferation by decreasing the cell number and promoting lipogenesis and lipid filling in 3T3-L1 adipocytes. In contrast, the effect of CLA-PC on suppressing lipid accumulation in 3T3-L1 adipocytes was shown to be between that of CLA-FFA and soy-PC or almost the same as that of soy-PC. Therefore, CLA-PC might be an effective form of CLA in terms of daily intake compared with the FFA form.

Effect of PC components on lipid accumulation in
3T3-L1 adipocytes It was reported that PC could induce lipolysis and apoptosis in 3T3-L1 adipocytes through the TNFα-mediated pathway 13,15 . To gain insight into the mechanism by which the PC form could suppress lipid accumulation more effectively than the FFA form, we investigated the effect of treatment with the PC component on lipid accumulation in 3T3-L1 adipocytes. In this experiment, we used choline, GPC, and LA as a major fatty acid in soy-PC as samples.
We accordingly found that GPC and LA showed significantly higher lipid suppression in cells than the control Fig. 8 . This result was in line with a previous report that  LA could reduce lipid accumulation in 3T3-L1 adipocytes, even though the effect was lower than that of CLA 4,7 . Based on this, we assumed that the suppression of lipid accumulation was affected by the chemical components of PC, although the mechanisms remain unclear.

Conclusion
In conclusion, CLA-PC was shown to be superior to CLA-FFA in decreasing lipid accumulation in 3T3-L1 adipocytes. This effect was attributed to the suppression of lipogenesis but not lipolysis in adipocytes. Therefore, considering a daily intake routine, the PC form of CLA is promising for the reduction of lipid accumulation in the body compared with the FFA form.