Original papers
Effect of dietary fats and soybean phospholipid on hepatic fatty acid metabolism in mice
2024 Volume 30 Issue 1 Pages 83-95
Details
2024 Volume 30 Issue 1 Pages 83-95
The combined effect of γ-linolenic acid (GLA)-rich evening primrose oil and soybean phospholipid on lipid metabolism was investigated. Male ICR mice were fed diets containing 100 g/kg of coconut, safflower or GLA oil, supplemented with 31 g/kg of soybean oil or 50 g/kg of soybean phospholipid, for 21 d. All experimental diets provided the same amount of fatty acids. The diets containing GLA oil significantly reduced hepatic triacylglycerol levels compared with those containing coconut oil. The values obtained from diets containing GLA oil also tended to be lower than those obtained from the diets containing safflower oil. Soybean phospholipid, regardless of the type of dietary fat, reduced hepatic triacylglycerol levels. Therefore, the combination of GLA oil and soybean phospholipid was effective in causing a marked decrease in hepatic triacylglycerol levels. Changes in hepatic fatty acid oxidation and lipogenesis may be responsible for the observed results.
Alterations in hepatic fatty acid synthesis (Fukuda and Ontko, 1984) and oxidation (Ide and Ontko, 1981; Ide et al., 1982) modify the amounts of fatty acids available for triacylglycerol synthesis, which consequently alters very-low-density lipoprotein production by the liver. Therefore, a change in the rate of these metabolic processes modifies hepatic triacylglycerol levels and serum lipid concentrations.
It is well established that different types of dietary fat are critical in determining serum lipid levels in humans (Hegsted et al., 1993). Numerous animal studies in rats and mice have also shown that different types of dietary fat strongly influence serum and liver lipid levels. Findings from our previous studies have shown that different types of dietary fat characteristically alter hepatic fatty acid synthesis and oxidation in different ways in rats and mice (Kumamoto and Ide, 1998; Ide et al., 2000; Takahashi et al., 2000; Ide et al., 2004; Ide et al., 2017; Ide and Origuchi, 2019; Ide and Origuchi, 2020). Therefore, it is conceivable that alterations in these pathways are involved in the dietary fat-dependent changes in serum and liver lipid levels. Oils that are rich in certain polyunsaturated fatty acids have a marked influence on these metabolic pathways. Regarding the physiological activities of γ-linolenic acid (GLA)-rich oils, our previous studies (Kumamoto and Ide, 1998; Takahashi et al., 2000; Ide et al., 2017; Ide and Origuchi, 2020) showed that dietary fats rich in GLA increased hepatic fatty acid oxidation in rats. The effects were pronounced for peroxisomal fatty acid oxidation; however, the increases observed in these studies were rather small (20–80 %). We recently demonstrated that dietary GLA causes considerably more pronounced increases hepatic fatty acid oxidation in mice (Ide and Origuchi, 2019; Ide and Origuchi, 2020). Accordingly, evening primrose oil, which is rich in GLA (42.6 %, GLA oil) was compared to palm oil (a saturated fat) and safflower oil (an oil rich in linoleic acid but lacking GLA). The findings showed that evening primrose oil increased the rate of hepatic peroxisomal fatty acid oxidation by 4- to 5-fold at a dietary level of 10 %. It is considered that the strong GLA oil-dependent increase observed in hepatic fatty acid oxidation may be effective in ameliorating metabolic disorders in organisms. Indeed, GLA oil strongly reduced serum triacylglycerol levels, cholesterol, and phospholipid levels in the very-low-density lipoprotein + low-density lipoprotein fraction in hyperlipidemic apoE-null mice (Ide and Origuchi, 2019). Regarding the physiological activity of GLA oil on hepatic lipogenesis, this oil, compared to a saturated fat (palm oil), reduced the activity and mRNA levels of numerous lipogenic enzymes in both rats (Ide et al., 2017) and mice (Ide and Origuchi, 2019). This lowering effect of GLA oil was almost comparable to that observed with linoleic acid-rich safflower oil in both rats and mice.
Serum and liver lipid-lowering effects of dietary soybean phospholipid have been reported previously in animal and human studies (Cohn et al., 2008; Pandey et al., 2008). We have previously shown that soybean phospholipid markedly decreases the activity and mRNA levels of lipogenic enzymes in rat liver in a dose-dependent manner (Ide et al., 1992; Rouyer et al., 1999). However, soybean phospholipid had no effect on hepatic fatty acid oxidation in rats. Although few studies have examined the physiological activities of soybean phospholipid on lipid metabolism in mice, one study showed that soybean phospholipid reduced serum levels of triacylglycerol and cholesterol in mice fed a high-fat diet (Lee et al., 2014). Therefore, it is conceivable that soybean phospholipid exerts similar physiological effects on hepatic fatty acid metabolism in mice.
It is widely accepted that oils that are rich in GLA and soybean phospholipid are dietary factors that profoundly affect hepatic fatty acid metabolism, and therefore, these factors exert a lipid-lowering effect. It is expected that the combination of these compounds in the diet would have a profound effect on hepatic fatty acid metabolism and thus be effective in reducing serum and tissue lipid levels, as well as the incidence of atherosclerosis. We therefore investigated the combined effect of a GLA-rich oil and soybean phospholipid on lipid metabolism in mice.
Animals and diets Male ICR mice, purchased from Charles River Japan (Kanagawa, Japan) at 4 weeks of age, were housed individually in animal cages in a room with controlled temperature (20–22 °C) and lighting (lights on from 07:00 to 19:00) and fed a commercial diet (type NMF; Oriental Yeast Co., Tokyo, Japan). After 14 d of acclimatization, mice were fed purified experimental diets supplemented with either 3.1 °% of soybean oil (Nacalai Tesque Inc., Kyoto, Japan) or 5.0 % of soybean phospholipid (a gift from Taiyo Kgaku Co., Yokkaichi, Japan), together with 10.0 °% of coconut oil (a saturated fat, purchased from Nacalai Tesque Inc., Kyoto, Japan), safflower oil (rich in linoleic acid, a gift from Nisshin OilliO Group, Ltd., Tokyo, Japan), or evening primrose oil rich in γ-linolenic acid (GLA oil) for 21 d. The compositions of the experimental diets are shown in Table 1. The compositions of vitamin and mineral mixtures (AIN-93-VX and AIN-93G-MX, respectively) obtained from Oriental Yeast Co. (Tokyo, Japan) were the same as those reported previously (Reeves et al., 1993). GLA oil containing 39.9 % GLA prepared by the selective hydrolysis of evening primrose oil containing 10.5 °% GLA using lipase of Candida cylindracea origin was a gift from Tama Biochemical Co., Ltd., Tokyo, Japan. Soybean oil and soybean phospholipid at the levels added provided comparable amounts of fatty acids (3.02 and 3.03 %, respectively) in the diets. The phospholipid composition of soybean phospholipid was (in mol %): lysophosphatidylcholine, 10.7; phosphatidylcholine, 34.1; phosphatidylinositol,18.0; and phosphatidylethanolamine, 37.2. The fatty acid content was 605 mg/g. The fatty acid compositions of dietary fats and soybean phospholipid were analyzed by gas-liquid chromatography using a FAMEWAXTM column (30 m × 0.25 mm, Restek USA, Bellefonte, PA) and are shown in Table 2. The absorption rates of fatty acids from all of the dietary lipids used in this study are considered to be very high and comparable (McKimmie et al., 2013; Mu and Høy, 2004; Ramirez et al., 2001; Watanabe and Tsujino, 2022). Animals had free access to food and water throughout the experimental period. This study was approved by our university's animal ethics review board (Approval No. 1904, June 19, 2019), and we followed the university's guidelines for the care and use of laboratory animals.
| Ingredients (g/100 g) | Coconut oil+ Soybean oil | Coconut oil+ Soybean PL | Safflower oil+ Soybean oil | Safflower oil+ Soybean PL | GLA oil+ Soybean oil | GLA oil+ Soybean PL |
|---|---|---|---|---|---|---|
| Casein | 20.00 | 20.00 | 20.00 | 20.00 | 20.00 | 20.00 |
| Coconut oil | 10.00 | 10.00 | - | - | - | - |
| Safflower oil | - | - | 10.00 | 10.00 | - | - |
| GLA oil* | - | - | - | - | 10.00 | 10.00 |
| Soybean oil | 3.10 | 0.00 | 3.10 | 0 00 | 3.10 | 0.00 |
| Soybean PL** | - | 5.00 | - | 5.00 | - | 5.00 |
| Starch | 15.00 | 15.00 | 15.00 | 15.00 | 15.00 | 15.00 |
| Cellulose | 2.00 | 2.00 | 2.00 | 2.00 | 2.00 | 2.00 |
| Vitamin mixture | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Mineral mixture | 3.50 | 3.50 | 3.50 | 3.50 | 3.50 | 3.50 |
| Choline bitartrate | 0.25 | 0.25 | 0.25 | 0.25 | 0.25 | 0.25 |
| L-Cystin | 0.30 | 0.30 | 0.30 | 0.30 | 0.30 | 0.30 |
| Sucrose | 44.85 | 42.95 | 44.85 | 42.95 | 44.85 | 42.95 |
| Fattt acid (wt%) | Dietary lipids | ||||
|---|---|---|---|---|---|
| Coconut oil | Saffiower oil | GLA oil* | Soybean PL** | Soybean oil | |
| 8:0 | 1.9 | - | - | - | - |
| 10:0 | 3.3 | - | - | - | - |
| 12:0 | 53.0 | - | 0.2 | - | - |
| 14:0 | 21.3 | 0.1 | - | 0.1 | 0.1 |
| 16:0 | 9.7 | 5.8 | 4.5 | 17.0 | 9.1 |
| 16:1 (n-7) | 0.2 | 0.1 | 0.1 | 0.2 | 0.1 |
| 18:0 | 2.8 | 1.7 | 2.1 | 3.1 | 2.9 |
| 18:1 (n-9) | 6.1 | 12.5 | 6.0 | 5.6 | 25.1 |
| 18:2 (n-6) | 1.7 | 79.7 | 47.2 | 64.1 | 54.8 |
| 18:3 (n-6) | - | - | 39.9 | - | - |
| 18:3 (n-3) | - | 0.3 | - | 9.9 | 7.9 |
Enzyme assays At the end of the experiment, animals were anesthetized with isoflurane and euthanized by bleeding from the inferior vena cava between 09:00 and 11:00 h, after which the livers were quickly excised. A portion of each liver (approximately 0.6 g) was homogenized in 10 volumes of 0.25 M sucrose containing 1 mM EDTA and 3 mM Tris-HC1 (pH 7.2). A portion of the homogenate (4 ml) was centrifuged at 200 000 g for 30 min. The activities of lipogenic enzymes and enzymes involved in fatty acid oxidation were measured spectrophotometrically using 200 000 g supernatant of the liver homogenate and the whole liver homogenate as the enzyme source, respectively (Ide et al., 2004).
RNA analysis Liver RNA was extracted, and mRNA abundance was analyzed by quantitative real-time PCR as described elsewhere (Ide, 2005). mRNA abundance was calculated as a ratio to β actin level in each cDNA sample and expressed as a fold-change, with a value of 1 assigned to mice fed a diet containing coconut oil and soybean oil.
Western blotting A portion of the liver (0.3 g) was homogenized in 10 volumes of 0.25 M sucrose containing 1 mM EDTA, 3 mM Tris-HCl (pH 7.2) and 1 % (v/v) of a protease inhibitor cocktail obtained from Nacalai Tesque, Inc., Kyoto, Japan. The homogenates were centrifuged at 1 000 g for 10 min and the supernatants obtained were then centrifuged at 8 000 g for 10 min. Finally, the supernatants were centrifuged at 200 000 g for 30 min to separate cytosolic and microsomal fractions. Microsomes were suspended in 1 mL of 0.25 M sucrose containing 1 mM EDTA, 3 mM Tris-HCl (pH 7.2), and 1 % (v/v) of a protease inhibitor cocktail. Cytosolic fractions were used for western blot analysis of protein levels of acetyl-CoA carboxylase α, ATP citrate lyase, and pyruvate kinase using rabbit immunoglobulin G specific for the respective proteins. Antibodies specific for acetyl-CoA carboxylase α, ATP-citrate lyase, and pyruvate kinase were obtained from GeneTex, Inc. (Irvine, CA, USA). Microsomal fractions were probed for the precursor form of sterol regulatory element binding protein (SREBP)-1 using rabbit anti-SREBP-1 serum. The rabbit anti-SREBP-1 serum was the same as that used in our previous study (Ide et al., 2001). For immunoblot analysis, cytosolic and microsomal fractions were subjected to 5–15 % gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to polyvinylidene difluoride membranes, incubated with specific antibodies, and then reacted with anti-rabbit immunoglobulin G coupled to horseradish peroxidase (Cytiva, Tokyo, Japan). Chemiluminescence was generated using a kit provided by Cytiva (ECL Prime Western Blotting Detection Reagents) and quantified using a C-DiGit blot scanner equipped with Image Studio for C-DiGit software (LI-COR, Lincoln, NE, USA). Protein levels were expressed as a fold-change, with a value of 1 assigned to mice fed a diet containing coconut oil and soybean oil.
Analyses of serum and liver components Serum triacylglycerol, cholesterol, phospholipid, and glucose concentrations were measured using commercially available enzyme kits (Wako Pure Chemical, Osaka, Japan). Liver triacylglycerol (Fletcher, 1968), phospholipid (Rouser et al., 1970), and cholesterol (Ide et al., 1982) concentrations were determined as described previously.
Statistical analysis Microsoft Excel add-in software (Excel Statistics 2010; Social Survey Research Information Co., Tokyo, Japan) was used for statistical analysis. Data were expressed as means and their standard errors. Levene's test and the Kolmogorov-Smirnov test were used to evaluate the homogeneity of variance and normality of the distribution of the observations, respectively. If the variances were heterogeneous and/or the distributions were not normal, they were logarithmically transformed. As transformations were successful in making the variance of the observations constant and normalizing the data distribution, the transformed values were used for subsequent statistical analyses. The data were analyzed by two-way ANOVA to establish the effect of dietary fat types and soybean phospholipid or any interaction between these two factors. When the effect of fat types was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the effect of the respective fat. When the interaction was significant, the data were reanalyzed by one-way ANOVA and Tukey's post-hoc test. Differences were considered significant when p < 0.05.
Animal growth and liver weight Two-way ANOVA revealed a significant interaction between dietary fat and soybean phospholipid on average daily food intake (Table 3). Accordingly, we reanalyzed the data by one-way ANOVA. However, the test indicated that there was no significant difference among the groups. In addition, dietary fat types and soybean phospholipid did not affect body weight at the time of euthanasia and growth during the experimental period. Liver weights were significantly higher in mice fed GLA oil than in those fed coconut oil or safflower oil. Soybean phospholipid did not affect this parameter in mice fed different fats.
| Dietary oils | Two-way ANOVA (Pp value) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Soybean PL** | Coconut oil | Safflower oil | GLA oil* | Fat type | Soybean PL | Fat type× Soybean PL | Pot hoc test For fat effect (p<0.05) | |||
| − | + | − | + | − | + | |||||
| Food intake (g/day) | 4.8 ± 0.1 | 5.1 ± 0.2 | 4.9±0.1 | 4.7 ±0.1 | 4.7 ±0.1 | 4.9 ±0.1 | NS | NS | <0.05 | - |
| Body weight (g) | ||||||||||
| 0 d | 31.0 ± 0.5 | 30.5 ± 0.7 | 30.7 ±0.7 | 30.8 ± 0.7 | 31.1 ± 0.7 | 30.6 ±0.1 | NS | NS | NS | - |
| 21 d | 38.7 ± 0.9 | 38.6 ± 1.3 | 39.0 ± 0.9 | 36.9 ± 1.2 | 38.1 ± 1.0 | 38.2 ± 0.9 | NS | NS | NS | - |
| Growth (g/21 d) | 7.7 ±0.5 | 8.1 ± 1.2 | 8.2 ±0.4 | 6.1 ± 0.6 | 7.0 ± 0.7 | 7.6 ± 0.4 | NS | NS | NS | - |
| Liver weight (g/100 g body weight) | 5.37 ±0.17 | 5.29 ± 0.11 | 5.36 ± 0.19 | 5.47 ± 0.07 | 6.53 ± 0.11 | 6.75 ± 0.12 | <0.01 | NS | NS | C = S<G |
Values are the mean SEM (n = 7). Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, samowcr oil; G, GLA oil When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test.
Effect of dietary fat type and soybean phospholipid on hepatic fatty acid oxidation The activities of hepatic enzymes involved in fatty acid oxidation are shown in Fig. 1. The results of two-way ANOVA are shown in each panel for each enzyme. When two-way ANOVA showed that the effect of dietary fat was significant and the interaction of two experimental factors, i.e., dietary fats and soybean phospholipid, was not significant, post hoc tests were performed to clarify significant differences among the effects of the respective fats, and the results are also shown in each panel. Except in a few cases, GLA oil significantly increased the activities of many enzymes involved in hepatic fatty acid oxidation compared to other fats (Fig. 1). Soybean phospholipid had only a marginal effect on the activities of enzymes involved in hepatic fatty acid oxidation. Soybean phospholipid significantly, but only slightly, decreased the rate of peroxisomal palmitoyl-CoA oxidation regardless of dietary fat type. Soybean phospholipid also slightly decreased the activities of 3-hydroxyacyl-CoA dehydrogenase and 3-ketoacyl-CoA thiolase in mice fed GLA oil, but not in those fed other oils. Also, soybean phospholipid slightly increased the activities of carnitine acyltransferase measured with C16 substrate in mice fed coconut oil and safflower oil, but not in those fed GLA oil. However, soybean phospholipid was irrelevant in modulating the activities of acyl-CoA oxidase, carnitine acyltransferase measured with C2 and C8 substrates, and enoyl-CoA hydratase.
Effect of dietary fat type and soybean phospholipid on the activities of enzymes involved in hepatic fatty acid oxidation in mouse liver.
Values are the mean ± SEM, n = 7. Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, safflower oil; G, GLA oil. When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test. The results of two-way ANOVA and Tukey's post-hoc test for the effect of fat type are shown in each panel. Means without a common letter are different (p < 0.05). NS, not significant. GLA oil, an oil of evening primrose origin rich in γ-linolenic acid.
Similar effects of dietary fats and soybean phospholipid were confirmed on the mRNA levels of enzymes involved in mitochondrial and peroxisomal fatty acid oxidation pathways (data not shown).
Effect of dietary fat type and soybean phospholipid on hepatic fatty acid synthesis Safflower oil and GLA oil, compared with coconut oil, significantly reduced the activities of fatty acid synthase to the same level among mice fed soybean phospholipid-free diets (Fig. 2). Soybean phospholipid significantly reduced the activity of this enzyme in mice fed coconut oil and GLA oil, but not in those fed safflower oil. Thus, compared with coconut oil, GLA oil, but not safflower oil, significantly reduced the activities of this enzyme in mice fed diets containing soybean phospholipid. Both types of dietary fats and soybean phospholipid significantly modified the activities of ATP-citrate lyase and pyruvate kinase. Compared with coconut oil, safflower oil and GLA oil were equally effective in reducing the activity of this enzyme. For pyruvate kinase, GLA oil significantly reduced the activity of this enzyme compared to coconut oil, but safflower oil did not. Soybean phospholipid, regardless of dietary fat type, significantly reduced the activity of ATP-citrate lyase and pyruvate kinase.
Effect of dietary fat type and soybean phospholipid on lipogenic enzyme activities in mouse liver.
Values are the mean ± SEM, n = 7. Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, safflower oil; G, GLA oil. When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test. The results of two-way ANOVA and Tukey's post-hoc test for the effect of fat type are shown in each panel. Means without a common letter are different (p < 0.05). NS, not significant. GLA oil, an oil of evening primrose origin rich in γ-linolenic acid.
Fig. 3 shows the mRNA levels of proteins related to lipogenesis. There are two types of acetyl-CoA carboxylase, i.e., alpha and beta. The alpha form, but not the beta form, appears to be involved in fatty acid synthesis in cytosol (Abu-Elheiga et al., 2001). There are four isoforms of mammalian pyruvate kinase. L-Pyruvate kinase is an enzyme expressed in the liver (Noguchi et al., 1992). Adiponutrin is a protein thought to be involved in the regulation of lipogenesis (Jenkins et al., 2004).
Effect of dietary fat type and soybean phospholipid on mRNA levels of lipogenic enzymes in mouse liver.
Values are mean ± SEM, n = 7. Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, safflower oil; G, GLA oil. When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test. The results of two-way ANOVA and Tukey's post-hoc test for the effect of fat type are shown in each panel. Means without a common letter are different (p < 0.05). NS, not significant. GLA oil, an oil of evening primrose origin rich in γ-linolenic acid.
Two-way ANOVA revealed that soybean phospholipid, independent of dietary fat type significantly reduced mRNA levels of the proteins related to lipogenesis. Two-way ANOVA also showed that dietary fat type significantly modified these parameters. Subsequent post-hoc tests showed that safflower oil and GLA oil significantly decreased mRNA levels of acetyl-CoA carboxylase α, fatty acid synthase, ATP-citrate lyase, and adiponutrin compared with coconut oil. GLA oil, but not safflower oil significantly decreased the mRNA levels of L-pyruvate kinase. The levels of these parameters, except for L-pyruvate kinase, were comparable between mice fed safflower and GLA oils. L-Pyruvate kinase mRNA levels were significantly lower in mice fed GLA oil than in those fed safflower oil.
We also analyzed mRNA levels of transcription factors involved in the regulation of hepatic fatty acid synthesis (Fig. 4). Soybean phospholipid regardless of dietary fat type, significantly decreased the mRNA levels of carbohydrate response element binding protein (ChREBP)β and sterol regulatory element binding protein (SREBP)-1c. Regarding the effect of dietary fat type, both safflower oil and GLA oil significantly decreased the levels of SREBP-1c compared to coconut oil. Compared with coconut oil, GLA oil also significantly decreased the levels of ChREBPβ, but the safflower oil-dependent decrease was not significant. Dietary fat type and soybean phospholipid were totally ineffective in modulating ChREBPa and LXRa mRNA levels.
Effect of dietary fat type and soybean phospholipid on mRNA levels of transcription factors involved in the regulation of lipogenesis in mouse liver.
Values are mean ± SEM, n = 7. Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, safflower oil; G, GLA oil. When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test. The results of two-way ANOVA and Tukey's post-hoc test for the effect of fat type are shown in each panel. Means without a common letter are different (p < 0.05). NS, not significant. GLA oil, an oil of evening primrose origin rich in γ-linolenic acid.
The protein levels of some lipogenic enzymes as well as the precursor form of SREBP-1 are shown in Fig. 5. Regardless of the dietary fat type, soybean phospholipid significantly decreased the protein levels of acetyl-CoA carboxylase α, ATP-citrate lyase, L-pyruvate kinase, and the precursor form of SREBP-1. Dietary fat type also modified these parameters. Both safflower oil and GLA oil significantly decreased protein levels of acetyl-CoA carboxylase α, ATP citrate lyase, and the precursor form of SREBP-1 compared to coconut oil. These oils were equally effective in reducing the levels. Regarding the L-pyruvate kinase protein level, GLA oil significantly lowered the level compared to both coconut oil and safflower oil. However, compared to coconut oil, safflower oil did not reduce the level of L-pyruvate kinase. Attempts to detect the mature form of SREBP-1 using total homogenate were not successful, likely because of the very low concentrations of mature SREBP-1 and its instability.
Effect of dietary fat type and soybean phospholipid on protein levels of enzymes involved in lipogenesis and the immature form of SREBP-1 in mouse liver.
Values are mean ± SEM, n = 7. Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not significant, the data were analyzed by Tukey's post-hoc test to clarify the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, safflower oil; G, GLA oil. When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test. The results of two-way ANOVA and Tukey's post-hoc test for the effect of fat type are shown in each panel. Means without a common letter are different (p < 0.05). NS, not significant. GLA oil, an oil of evening primrose origin rich in γ-linolenic acid.
Effect of dietary fat type and soybean phospholipid on serum and liver lipids GLA oil significantly lowered serum triacylglycerol concentrations compared to both coconut and safflower oils (Table 4). No significant difference was observed between mice fed coconut oil and those fed safflower oil. However, dietary phospholipid was ineffective in reducing serum triacylglycerol levels. Two-way ANOVA indicated that dietary fat type significantly modified serum cholesterol concentration. However, subsequent pot-hoc test failed to detect any significant difference in this parameter among mice fed different fats. Again, dietary soybean phospholipid did not affect this parameter. Both safflower oil and GLA oil significantly lowered the serum concentration of phospholipid compared to coconut oil. No significant difference in this parameter was observed between mice fed safflower oil and those fed GLA oil. Soybean phospholipid, regardless of dietary fat type, significantly lowered serum phospholipid levels.
| Dietary oils | Two-way ANOVA (Pp value) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Soybean PL** | Coconut oil | Safflower oil | GLA oil* | Fat type | Soybean PL | Fat type × Soybean PL | Pot hoc test For fat effect (p < 0.05) | |||
| - | + | - | + | - | + | |||||
| Serum lipids(μmol/dL) | ||||||||||
| Triacylglycerol | 207 ± 21 | 196 ± 49 | 168 ± 9 | 122 ± 19 | 87.8 ± 6.6 | 82.8 ± 8.5 | < 0.01 | NS | NS | C = S > G |
| Cholesterol | 469 ± 40 | 512 ± 34 | 441 ± 23 | 404 ± 24 | 480 ± 25 | 491 ± 30 | < 0.05 | NS | NS | C = S = G |
| Phospholipid | 452 ± 31 | 368 ± 13 | 412 ± 16 | 297 ± 14 | 366 ± 18 | 323 ± 20 | < 0.01 | < 0.01 | NS | C > S = G |
| Liver lipids(μmol/g) | ||||||||||
| Triacylglycerol | 41.2 ± 4.9 | 30.4 ± 3.2 | 34.1 ± 5.3 | 22.9 ± 2.9 | 25.6 ± 2.1 | 18.8 ± 2.8 | < 0.01 | < 0.01 | NS | C = S > G |
| Cholesterol | 4.77 ± 0.17bc | 4.34 ± 0.15b | 5.40 ± 0.26e | 4.09 ± 0.17b | 4.08 ± 0.09ab | 3.37 ± 0.11a | < 0.01 | < 0.01 | < 0.05 | - |
| Phospholipid | 42.9 ± 0.9 | 45.4 ± 1.2 | 42.3 ± 0.9 | 42.6 ± 0.7 | 46.1 ± 0.7 | 47.8 ± 1.6 | < 0.01 | NS | NS | C = S < G |
Values are the mean ±SEM (n = 7). Data were analyzed by two-way ANOVA. When the effect of fat type was significant and the interaction was not signification, the data were analyzed by Tukey's post hoc test to clanfy the differential effect of each fat. The following abbreviations were used to indicate the results of the post-hoc test for fat effect. C, coconut oil; S, safflower oil; G, GLA oil. When the interaction was significant, the data were reanalyzed by one-way ANOVA followed by Tukey's post-hoc test.
GLA oil significantly reduced hepatic triacylglycerol concentration compared to coconut oil, but safflower oil did not. No significant differences in this parameter were found between mice fed safflower oil and mice fed GLA oil. Soybean phospholipid, regardless of dietary fat type, significantly reduced hepatic triacylglycerol levels. GLA oil, compared with safflower oil, but not coconut oil, significantly reduced hepatic cholesterol concentration in mice fed soybean phospholipid-free diets. In mice fed diets containing soybean phospholipid, GLA oil reduced the level compared with both coconut oil and safflower oil. A soybean phospholipid-dependent decrease in this parameter was observed in mice fed safflower oil, but not in those fed other oils. GLA oil significantly increased hepatic phospholipid concentration compared to both coconut oil and safflower oil. However, no significant difference in this parameter was observed between mice fed coconut oil and those fed safflower oil. Dietary soybean phospholipid did not affect this parameter regardless of dietary fat type.
Combined effect of dietary fats and soybean phospholipid on hepatic fatty acid oxidation The findings of the present study corroborated those of previous studies (Ide and Origuchi, 2019; Ide and Origuchi, 2020) and showed that GLA oil significantly increased the hepatic activity and mRNA levels of many fatty acid oxidation enzymes in mice. It is considered that GLA oil increases hepatic fatty acid oxidation through upregulation of the PPARa signaling pathway, and it is possible that GLA or its metabolites act as a ligand to activate PPARa. Indeed, previous studies have shown that different types of polyunsaturated fatty acids act as ligands and activators of PPARa (Forman et al., 1997; Kliewer et al., 1997; Krey et al., 1997). However, all studies showed that different polyunsaturated fatty acids (linoleic, linolenic, arachidonic, eicosapentaenoic, and docosahexaenoic acids) strongly activate PPARα to the same extent. This observation is not consistent with the results of animal studies that dietary fats rich in α- and γ-linolenic acids and eicosapentaenoic and docosahexaenoic acids, but not those rich in linoleic and arachidonic acids, increased hepatic fatty acid oxidation. (Ide et al., 2000; Ide et al., 2004; Ide et al, 2012; Ide et al., 2017; Ide and Origuchi, 2019; Ide and Origuchi, 2020). Some studies also demonstrated that arachidonate metabolites (eicosanoids) act as strong activators of PPARa. Again, this is inconsistent with the finding of a previous animal study (Ide et al., 2012) which showed that dietary arachidonic acid is ineffective in increasing hepatic fatty acid oxidation.
We have previously shown that soybean phospholipid is not effective in modulating hepatic fatty acid oxidation in rats (Ide, 2014). In the present study using mice as experimental animals, soybean phospholipid-dependent changes in the activities of hepatic fatty acid oxidation enzymes were observed in only two cases. That is, soybean phospholipid significantly decreased the activities of 3-hydroxyacyl-CoA dehydrogenase and 3-ketoacyl-CoA thiolase in mice fed GLA oil, but not in other conditions. In addition, soybean phospholipid was ineffective in modulating mRNA levels of peroxisomal and mitochondrial enzymes involved in hepatic fatty acid oxidation (data not shown). It appears that soybean phospholipid is rather ineffective in modulating hepatic fatty acid oxidation, not only in rats but also in mice.
Combined effect of dietary fats and soybean phospholipid on hepatic lipogenesis Previous studies have shown that soybean phospholipid decreased hepatic lipogenesis in rats (Ide et al., 1992; Rouyer et al., 1999; Ide, 2014). The physiological activity of soybean phospholipid in decreasing hepatic lipogenesis was confirmed in the present study using ICR mice. We previously reported that dietary fats rich in polyunsaturated fatty acids decrease hepatic lipogenesis compared with saturated fats, in rats (Kumamoto and Ide, 1998; Ide et al., 2000; Ide et al., 2017) and mice (Ide et al., 2004; Ide and Origuchi, 2019). The findings of these studies suggested that linoleic acid and GLA are equivalent in their ability to reduce hepatic lipogenesis. Consistent with these previous studies, the present study showed that safflower oil composed mainly of linoleic acid and no GLA and GLA oil composed mainly of equivalent amounts of linoleic acid and GLA were equally effective in reducing the activities of fatty acid synthase and ATP-citrate lyase. Similar responses were observed with the mRNA levels of acetyl-CoA carboxylase α, fatty acid synthase, ATP citrate lyase, and adiponutrin, and with the protein levels of acetyl-CoA carboxylase α and ATP citrate lyase. However, the results obtained for pyruvate kinase were somewhat different. GLA oil strongly decreased the activity, mRNA and protein levels of pyruvate kinase, but safflower oil failed to cause significant changes in these values. The observed changes in the expression of lipogenic enzymes may have been due to the combined effects of signals mediated by several transcription factors.
Numerous lipogenic enzymes are under the control of SREBP-1. ChREBPs also transactivate the genes of many proteins involved in glucose metabolism and lipogenesis. There are two isoforms of ChREBP, α and β (Iizuka, 2017), and ChREBPα is mainly localized in the cytosol. Upon stimulation, cytosolic ChREBPα is translocated to the nucleus where it upregulates genes involved in glucose metabolism and lipogenesis. ChREBPα also targets the ChREBPβ gene to stimulate its expression. The transcriptional activity required to stimulate the gene expression of enzymes involved in glucose metabolism and lipogenesis is markedly higher with the β-isoform than with the α-isoform. In addition, LXRα also regulates lipogenesis. This transcription factor is a dominant activator of the SREBP-1c promoter (Repa et al., 2000). Also, LXRα regulates fatty acid synthase expression through direct interaction with the fatty acid synthase promoter (Joseph et al., 2002).
In the present study, dietary phospholipid, regardless of the type of dietary fat, significantly decreased the mRNA expression of SREBP-1c. In addition, safflower oil and GLA oil decreased the mRNA expression of SREBP-1c compared to coconut oil. Furthermore, western blot analysis showed that changes in the protein levels of the SREBP-1 precursor by soybean phospholipid and dietary fats were strongly correlated with changes in the mRNA levels of this transcription factor. The findings showed that dietary fat and soy phospholipid both affect the gene expression of this transcription factor and consequently modulate hepatic lipogenesis. Indeed, the contours of changes in the activity of fatty acid synthase and ATP-citrate lyase, and the mRNA levels of many proteins involved in lipogenesis, including acetyl-CoA carboxylase a, fatty acid synthase, ATP-citrate lyase, and adiponutrin, as well as the protein levels of acetyl-CoA carboxylase α and ATP-citrate lyase, reflected the changes in the mRNA and protein levels of SREBP-1. It is possible that not only the changes in gene expression of SREBP-1, but also the changes in the proteolytic process required to convert the precursor form of this transcription factor to the mature form are involved in the regulation by soybean phospholipid and dietary fat type. This needs to be clarified in the future studies by determining the protein level of the mature form in the nucleus. Indeed, a previous study (Yahagi et al., 1999) reported that a dietary fish oil-dependent decrease in lipogenesis was accompanied by a large decrease in the amount of the mature form of SREBP-1 in the nucleus, but only marginal decreases in the mRNA levels and protein levels of the membrane-bound precursor form of this transcription factor were observed.
As observed for SREBP-1c mRNA levels, the changes in ChREBPβ mRNA levels by soybean phospholipid and dietary fats generally reflected the activity, mRNA, and protein levels of the lipogenic enzymes. This observation suggests that the ChREBP pathway, in addition to the SREBP-1 pathway, is involved in the regulation of hepatic lipogenesis by soybean phospholipid and dietary fat type. However, the dietary fat-dependent changes in ChREBPβ mRNA levels were somewhat different from those observed for SREBP-1c. Compared to coconut oil, safflower oil and GLA oil reduced SREBP-1c mRNA levels to the same extent. GLA oil also significantly reduced ChREBPβ mRNA levels compared to coconut oil, but the safflower oil-dependent decreases in this parameter were insignificant. This may be the reason for the different responses of pyruvate kinase activity and L-pyruvate kinase mRNA and protein levels to the respective dietary fats. That is, GLA oil strongly reduced these parameters compared to coconut oil, but safflower oil did not. It has been reported that L-pyruvate kinase is under the control of ChREBP, but SREBP-1 is not involved in regulating the gene expression of this enzyme (Iizuka, 2017). Furthermore, it has been reported that a PPARa agonist down-regulates the gene expression of L-pyruvate kinase (Xu et al., 2006). It is plausible that the greater decrease in pyruvate kinase gene expression observed with GLA oil is a consequence of the activation of PPARα.
Combined effect of dietary fat type and soybean phospholipid on serum and hepatic lipid levels Alterations in hepatic fatty acid synthesis and oxidation alter the availability of fatty acids for triacylglycerol synthesis, which in turn, alters the production of very-low-density lipoprotein by the liver (Fukuda and Ontko, 1984; Ide and Ontko, 1981; Windmueller and Spaeth, 1967); therefore, a change in the rate of these metabolic processes is critical in determining serum lipid concentrations. In the current study, dietary fat type modified serum triacylglycerol. The values were lowest with GLA oil, intermediate with safflower oil, and highest with coconut oil. The strong effect of GLA oil on this parameter is considered to be a consequence of the combined effect of the reduction of hepatic lipogenesis and the marked increase in hepatic fatty acid oxidation. The effects of dietary fat type on serum cholesterol and phospholipid levels were less clear. That is, both safflower oil and GLA oil reduced serum phospholipid levels compared with coconut oil, but the levels were comparable between mice fed safflower oil and GLA oil. Dietary fat type was also not effective in modulating serum cholesterol concentrations. This is not surprising because very-low-density lipoproteins are rich in triacylglycerol, but they contain less cholesterol and phospholipids, and these lipids are mainly distributed in low- and high-density lipoproteins. It is suggested that GLA oil decreased the production of very-low-density lipoproteins by the liver and thus decreased serum triacylglycerol concentrations, but exerted a different effect on the metabolism of low- and high-density lipoproteins and thus had minimal effect on serum cholesterol and phospholipid concentrations. Dietary soybean phospholipid significantly reduced the serum phospholipid concentrations. Unexpectedly, however, soybean phospholipid did not affect serum concentrations of triacylglycerol and cholesterol. We previously reported that soybean phospholipid effectively reduced serum concentrations of triacylglycerol, cholesterol, and phospholipid in Sprague-Dawley rats (Ide, 2014). The response of serum lipid concentrations to dietary soybean phospholipid may vary between species; this needs to be clarified in future studies.
Hepatic triacylglycerol concentration was lowest with GLA oil, intermediate with safflower oil, and highest with coconut oil. In addition, soybean phospholipid decreased the level of triacylglycerol levels regardless of the type of dietary fat. These changes are considered to be the result of dietary alterations in hepatic lipogenesis and fatty acid oxidation. Although the changes were not always significant, GLA oil lowered hepatic cholesterol concentrations compared to the other oils. Dietary soybean phospholipids also lowered hepatic cholesterol levels, although a significant difference was found only in mice fed safflower oil. Therefore, it is possible that dietary fat types and soybean phospholipid affect not only fatty acid metabolism, but also cholesterol metabolism in the liver. The diets containing GLA oil, compared with those containing coconut and safflower oils, increased hepatic phospholipid levels and were accompanied by an increase in liver weight, which may reflect the proliferation of mitochondria and peroxisomes. Previous studies have shown that agonists of PPARa increase hepatic phospholipid levels (Yanagita et al., 1987).
Fatty acid desaturase 2 (Δ desaturase), which converts linoleic acid to γ-linolenic acid, is considered a rate-limiting step in the pathway for the synthesis of eicosapolyenoic acids (dihomo-γ-linolenic acid and arachidonic acid), which serve as substrates for eicosanoid production (Nakamura and Nara, 2003). It is therefore plausible that the intake of γ-linolenic acid bypasses the fatty acid desaturase 2 step and thus enhances the production of various eicosanoids by increasing the levels of eicosapolyenoic acids. It is possible that the increase in eicosanoid production is a factor that accounts for the triacylglycerol-lowering effect of GLA oil. However, there is evidence that eicosanoids, especially the 2-series prostaglandins derived from arachidonic acid, cause hepatic triacylglycerol accumulation (Wang et al., 2021). This is not consistent with the present observation that GLA oil decreased hepatic triacylglycerol concentrations.
Compared with coconut oil and safflower oil, GLA oil markedly increased the activities of enzymes involved in hepatic fatty acid oxidation; however, soybean phospholipid was not involved in altering these parameters regardless of dietary fat type. Compared with coconut oil, GLA oil and safflower oil decreased parameters related to hepatic lipogenesis. In addition, soybean phospholipid reduced these parameters independently of dietary fat type. SREBP-1- and ChREBP-signaling pathways may be involved in these changes in hepatic lipogenesis. GLA oil significantly reduced hepatic triacylglycerol levels compared to coconut oil. Although the difference was not significant, the levels in mice fed GLA oil were also lower than those obtained with safflower oil. Soybean phospholipid reduced hepatic triacylglycerol levels regardless of dietary fat type. GLA oil also strongly reduced serum triacylglycerol levels, but unexpectedly, soybean phospholipid did not affect this parameter. Therefore, the combination of GLA oil and soybean phospholipid was at least effective in reducing hepatic triacylglycerol levels. The enhancement of hepatic fatty acid oxidation by GLA oil and reduction of hepatic fatty acid synthesis by GLA oil and soybean phospholipid may account for this reduction.
Acknowledgements Expert technical assistance received from Mses. Izumi Origuchi, Saki Arat, Chisato Kima, Miho Takano, Tomomi Shimozaki, Haruka Takahashi, Madoka Nakagawa, and Misaki Matsuzaki was greatly appreciated.
Conflict of interest There are no conflicts of interest to declare.
