δδ -Tocopherol Slightly Accumulates in the Adipose Tissue of Mice

％ Abstract: This study aimed to compare the distribution of vitamin E analogs, particularly α-tocopherol and δ-tocopherol, in mice fed with a normal diet and a high-fat and high-sucrose diet separately. We used male C57BL/6JJcl strain mice, which were divided into six groups (control [C], Cα, Cδ, high-fat and high-sucrose [H], Hα, and Hδ groups) and bred for 4 weeks. The additional quantity of α-tocopherol or E-mix D (containing 86.7% δ-tocopherol) into diet was 800 mg/kg diet. The final body weight was significantly higher in the H group than in the C group. However, the effects of vitamin E analog intake had no significant difference, with no synergy between vitamin E and diet. Similar results were obtained in epididymal fat weight. Moreover, α-tocopherol was mainly distributed in the liver in both the Cα group and Hα group, whereas δ-tocopherol mostly accumulated in the epididymal fat, in both the Cδ group and Hδ group. Also, δ-tocopherol was detected in all tissues in both groups. Both the α-tocopherol and δ-tocopherol levels in the epididymal fat were significantly lower in the H group than in the C group. In conclusion, our results suggest that a portion of δ-tocopherol was incorporated into the adipose tissue by chylomicron before arriving at the liver, and then it is metabolized in the liver.

respectively, in rat primary hepatocytes and rat liver 10 . This α-TTP catalyzes the secretion of vitamin E analogs via a novel non-Golgi-mediated pathway in rat liver cells; subsequently, α-T is incorporated into VLDL preferentially and transported to various tissues by lipoprotein 11 . Either excess α-T or non-α-T, such as γ-T, δ-T and T3s, is rapidly metabolized and excreted in urine 12 or bile 13 . Therefore, contrary to non-α-T, only α-T can accumulate in each tissue. However, some reports showed novel effects of non-α-T in vitro and in vivo as shown above. In this study, we aimed to investigate the distribution of non-α-T especially δ-T, in the tissues of mice fed with a normal diet and high-fat and high-sucrose diet separately.

Experimental procedure
All experiments were conducted according to the Guide for the Care and Use of Laboratory Animals at Kanagawa Institute of Technology.
We used male C57BL/6JJcl strain mice 3 weeks old, n 31 , which were purchased from CLEA Japan, Inc. These mice were housed individually in plastic cages and were kept in an environment controlled at 23 2 and 55 5 humidity, with 12 h/12 h light/dark cycle. They were initially fed with a basic diet for 1 week to allow them to adapt to the new environment. Thereafter, they were divided into six groups control C , Cα, Cδ, high-fat and high-sucrose H , Hα, and Hδ groups according to their average weight to avoid differences. Table 1 presents the diet composition of each group. The feed and water were supplied ad libitum for 4 weeks. After a 16 h fast, all the mice were sacrificed under isoflurane anesthesia, and the arterial blood and each tissue were extracted for analysis.

Extraction of vitamin E analogs from mice tissues
Vitamin E analogs in each tissue were measured using Ueda s method 14 . A 0.1 g sample of each tissue was homogenized with 0.9 mL of 0.9 NaCl solution wt/vol . The homogenate solution 0.1 mL was then pipetted into a 10 mL centrifuge tube, with 0.1 mL of 2,2,5,7,8-pentamethyl-6-hydroxychroman 100 ng/mL as an internal standard. Subsequently, we added 1.0 mL of ethanolic pyrogallol 6 , wt/vol to each tube while stirring. After adding 0.2 mL of KOH solution 60 , wt/vol to each tube, we saponified the contents at 70 for 30 min. After cooling, vitamin E analogs were extracted using 4.5 mL of NaCl solution 1 , wt/vol and 3.0 mL of 10 ethyl acetate/n-hexane solution and then centrifuged at 3,000 rpm at 4 for 5 min. A 2.0 mL aliquot of the upper layer was evaporated, dissolved in 0.1 mL of n-hexane, and subjected to HPLC.

Statistical analysis
All data were expressed as the mean SD. Vitamin E concentration of each tissue was statistically analyzed by t-test. The final body weight, food intake, energy intake, epididymal fat weight and epididymal fat weight/100 g body weight were statistically analyzed by two-way ANOVA, followed by the Tukey-HSD post-hoc test. All statistical data were analyzed using the SPSS for Windows Tokyo, Japan . Differences were considered significant at p 0.05.

Results
3.1 Final body weight, food intake, energy intake, epididymal fat weights and epididymal fat weights per 100 g body weights in each group Table 2 shows the final body weight, food intake, energy intake, epididymal fat weights and epididymal fat weights per 100 g body weight of mice in each group. The final body weight was significantly higher in the H group than in the C group. However, the effects of vitamin E analogs intake revealed no significant differences, and no synergistic effects were found between diet and vitamin E. Furthermore, food intake was significantly lower in the H group than in the C group. However, therefore, high-fat and high-sucrose diet intake induced obesity for 4 weeks. Because energy intake was markedly higher in the H group than in the C group.
No significant difference was found for each tissue weight brain, heart, lung, liver, kidney, testes, and skeletal muscle , except for the epididymal fat weights in all groups data not shown . However, the epididymal fat weights were significantly higher in the H group than in the C group. However, the effects of vitamin E analog intake were not significantly different. Moreover, similar results were obtained in epididymal fat weight/100 g body weight.
3.2 α-T concentration in the mouse tissue of the C and H groups Figure 1 illustrates the concentration of α-tocopherol in each tissue of the C and H groups. In both groups, only α-T accumulated in all tissues involved. Conversely, δ-T was not detected in all tissues of these two groups. However, the liver had the highest accumulation of α-T in both the Cα and Hα groups. Meanwhile, the α-T level in the epididymal fat of the Hα group was significantly lower than that of the Cα group Fig. 2 . Of note, we did not include the plasma data in the figure because the data had considerably few numbers n 2-3 . Thus, the concentration of α-T in the plasma of each group was as follows: C, 0.26 0.13 µg/mL; Cα, 7.25 0.20 µg/mL; Cδ, 1.34 0.23 µg/mL; H, 0.33 0.01 µg/mL; Hα, 8.22 0.69 µg/mL; and Hδ, 1.25 0.09 µg/mL.

δ-T concentration in the mouse tissue of the C and H groups
In the Cδ and Hδ groups, α-T was detected in all tissues. However, the α-T level tended to be lower in the Hδ group than in the Cδ group in all tissues Fig. 3 . Figure 4 shows the δ-T concentration in each tissue of Table 2 Final body weight, dietary intake, energy intake and epididymal fat weight in each group 1 . Fig. 1 Concentration of α-tocopherol in each tissue of the control and high-fat and high-sucrose groups. C: control diet n 6 , H: high-fat and high-sucrose diet n 5 . The data are presented as mean SD.
the Cδ and Hδ groups. In both groups, δ-T was detected in all tissues. Furthermore, δ-T accumulated most of epididymal fat in two groups. Furthermore, the concentration of δ-T in epididymal fat of Cδ group was significantly higher than that of Hδ group. Even, we did not write the plasma data into the figure because the data had too few numbers n 2-3 . So, the concentration of δ-T in the plasma of each group is as follows: C, not detected; Cα, not detected; Cδ, 0.063 0.003 µg/mL; H, not detected; Hα, not detected; and Hδ, 0.106 0.016 µg/mL.

Discussion
This study investigated the effects of vitamin E analogs on mice fed with a high-fat and high-sucrose diet. We found that the final body weights and epididymal fat weights were significantly higher in the H group than in the C group. However, the effects of vitamin E analog intake were not significantly different, with no synergy between diet and vitamin E Table 2 . Zhao et al. found that the body weight gain of mice fed with a high-fat diet γ-tocotrienol was drastically lower than that of mice fed with merely a high-fat diet 15 . Furthermore, Burdeos et al. reported that the body weight between the high-fat diet group and the high-fat diet 10 mg rice bran tocotrienol group was not significantly different; however, the epididymal fat weight was significantly lower in the high-fat diet 10 mg rice bran tocotrienol group than in the high-fat diet group 16 . Therefore, tocotrienols can potentially prevent or attenuate obesity. On the other hand, we have described that α-T reportedly cannot decrease the body weight of mice fed with a high-fat diet even though α-T increases the expression of the UCP1 and PGC-1α genes of adipose tissues in rats fed with a high-fat diet 6 . Moreover, in the present data, we were not able to prove the anti-obesity effect of δ-T. Therefore, we suggested that the anti-obesity effects of α-T and δ-T are not revealed in this study, probably because of the short feeding period 4 weeks . Hence, further examination is needed.
Next, we measured the distribution of α-T and δ-T in the tissues of mice fed with a diet containing vitamin E analogs. Results showed that α-T accumulates mostly in the liver Fig. 2 . Generally, α-T is transported to the liver and then to various tissues with VLDL again. Therefore, vitamin E stays in the liver once. However, a few numbers of δ-T remain in the liver Fig. 4 . Hence, we presumed that δ-T was metabolized more quickly than α-T in the liver. Regarding vitamin E metabolism, these analogs first undergo   ω-oxidation by the cytochrome P450 CYP enzyme, followed by degradation of the side chain by β-oxidation; finally, they are converted into carboxyethyl hydroxychroman CEHC and excreted into urine 12,17 . In the present study, we did not measure α-CEHC and δ-CEHC in rat urine, but we suggested that most of the δ-T was converted into δ-CEHC and rapidly excreted into urine. In the future, we will measure the metabolites of vitamin E in rat tissues for confirmation.
Interestingly, δ-T accumulated in the epididymal fat more than in the liver, as shown in Fig. 4. Among all vitamin E analogs, α-T accumulates the most in the body. However, vitamin E, except α-T, is characteristically distributed. Ikeda et al. 18 reported that α-T3 and γ-T3 were detected in substantial amounts in the skin of nude mice, hairless mice, and wistar strain rats administered with vitamin E mixtures. These T3s have low affinity with α-TTP, similar to δ-T 10 . Therefore, the distribution of nonα-Ts e.g., δ-T and T3s in the body is low. However, we, as well as other investigators, clarified that the biodistribution of vitamin E analogs is specific. Ikeda et al. investigated on the lymphatic transport of T3s and α-T in thoracic ductcannulated rats; they found that α-T3 had a drastically higher recovery than γ-T, δ-T and α-T and that γ-T, δ-T, and α-T recovered equally 19 . Meanwhile, Traber et al. 20 reported that no significant difference was found between the absorption of α-T and γ-T into the intestinal cells. Therefore, we assumed that the uptake of δ-T into the adipose tissue may occur during transportation by chylomicron. Chylomicron is a kind of lipoprotein that is synthesized in the intestinal cells and transports all dietary lipids into the blood circulation, and its component, triacylglycerol, is hydrolyzed by lipoprotein lipase LPL . LPL is located on the capillary walls as well as in the heart, adipose tissue, spleen, and lungs, and it is inactive in adult liver. Abe et al. showed that serum γ-T and α-T3 concentrations in Triton WR1338, an inhibitor of LPL, groups tended to increase as compared with control group and lowered them in the liver and adrenal gland of rat administered γ-T and α-T3 21 . During this process, tocopherols along with fatty acids are transferred to the tissues by action of LPL. For this transfer of tocopherol to occur, the LPL itself has to bind to the cell membrane 22 . However, it is not clarified about the relations with LPL and vitamin E for the moment. Therefore, we consider that the examination about relations with LPL and vitamin E is necessary in future. Moreover, it is clarified that the metabolism of vitamin E is occurred in the liver mainly 23 . However, it is not known whether vitamin E is metabolized in each tissue except the liver. Hence, we suggested that some of the δ-tocopherol in chylomicron was accumulated into the adipose tissue more than other tissues before arriving at the liver.
We also showed that the δ-T in the epididymal fat was markedly lower in the Hδ group than in the Cδ group Fig.   4 . Similar results were found in the α-T accumulating in the epididymal fat for the comparison with Cα group and Hα group Fig. 2 . We guessed that these results are caused by the decrease of cell number per 1 g of adipose tissue in H group because epididymal fat weight in mice of H group significantly increased than that of C group Table  2 . Unfortunately, we were not able to determine the lipid profiles in plasma. Furthermore, we were not able to measure the lipid content and the cell number in adipose tissue, considering that the data were insufficient. In the future, we will investigate on the uptake of vitamin E in the adipose tissue in conjugation with lipid profiles in plasma and adipose tissue.

Conclusion
In mice separately fed with a normal diet and a high-fat and high-sucrose diet, α-tocopherol mostly accumulated in the liver, whereas δ-tocopherol mostly accumulated in the adipose tissue but in minimal amounts. Therefore, we suggested that some of the δ-tocopherol was incorporated into the adipose tissue by chylomicron before arriving at the liver, and then the remaining δ-tocopherol was metabolized in the liver.