Lipids (2012) 47:495–503 DOI 10.1007/s11745-012-3667-0
ORIGINAL ARTICLE
Mechanisms Underlying Decreased Hepatic Triacylglycerol and Cholesterol by Dietary Bitter Melon Extract in the Rat Gamarallage V. K. Senanayake • Nobuhiro Fukuda • Shoko Nshizono • Yu-Ming Wang • Koji Nagao • Teruyoshi Yanagita • Masako Iwamoto • Hideaki Ohta
Received: 30 October 2011 / Accepted: 7 March 2012 / Published online: 29 March 2012 Ó AOCS 2012
Abstract In these studies, we focused on finding the mechanism(s) underlying the bitter melon (Momordica charantia L.) methanol fraction (MF)-dependent reduction in the concentration of hepatic triacylglycerol (TAG) and cholesterol in the rat. Rats were fed diets containing low (5 %) fat for 2 weeks (experiment 1), or low (5 %) and high (15 %) fat for a longer period of 8 weeks (experiment 2). MF was supplemented at 1 % level in both experiments. After feeding, rats were sacrificed, and their livers were
A preliminary part of this study was presented at 11th European Nutrition Conference, Spain, 2011. G. V. K. Senanayake N. Fukuda Department of Biochemistry and Applied Biosciences, Faculty of Agriculture, University of Miyazaki, Miyazaki 889-2192, Japan Present Address: G. V. K. Senanayake (&) Richardson Center for Functional Foods and Nutraceuticals, University of Manitoba, 196 Innovation Drive, Winnipeg, MB R3T 2N2, Canada e-mail:
[email protected] S. Nshizono Center for Collaborative Research and Community Cooperation, University of Miyazaki, Miyazaki 889-2192, Japan Y.-M. Wang K. Nagao T. Yanagita Department of Applied Biochemistry and Food Science, Faculty of Agriculture, Saga University, Saga 840-8502, Japan M. Iwamoto H. Ohta Faculty of Nutritional Science, Nakamura Gakuen University, Fukuoka 814-0198, Japan
prepared as slices and hepatocytes, followed by incubation with [1(2)-14C] acetate or [1-14C] oleic acid (18:1 n-6). Under these conditions, we found that rats fed diets containing MF, as compared to those without MF, showed: (1) no adverse effects on food intake and growth, (2) a decreased hepatic TAG and total cholesterol, irrespective of the difference in dietary fat level or feeding period, and (3) a decreased incorporation of [1(2)-14C] acetate and [1-14C] oleic acid into TAG of liver slices and hepatocytes. MFsupplemented rats also showed no altered incorporation of labeled acetate into cholesterol and cholesterol ester, an increased fecal excretion of neutral steroids, but not of acidic steroids, and an enhanced mRNA abundance of carnitine palmitoylacyltransferase I, which is the rate-limiting enzyme for fatty acid oxidation. These results suggest that dietary MF decreases hepatic TAG synthesis while enhancing fatty acid oxidation, thereby reducing the concentration of hepatic TAG. The liver cholesterol-lowering effect of MF, however, is probably mediated through an increased fecal excretion of neutral steroids, without an effect on cholesterogenesis. Keywords Momordica charantia Liver triacylglycerol-lowering effect Liver cholesterol-lowering effect Lipid metabolism Abbreviations CPT I Carnitine palmitoylacyltransferase I DPS Digitonin precipitable sterol MF Methanol fraction of Momordica charantia MTP Microsomal transfer protein PPAR Peroxisome proliferator-activated receptor STZ Streptozotocin TAG Triacylglycerol(s)
123
496
Introduction Bitter melon (Momordica charantia L.), which is cultivated in countries located in tropical and subtropical areas such as India, Asia, and South America has been traditionally used as a folk medicine for the treatment of diabetes. More recently, favorable effects on glucose metabolism have been demonstrated in cell culture, animal models and human studies [1–8]. It has also been reported that feeding of bitter melon and/or its extract was beneficial in preventing lipid disorders such as hyperlipidemia and diabetic dyslipidemia in both normal rats and in rats with chemically induced diabetes [9]. In a previous series of experiments, we also reported, to our knowledge for the first time, that the freeze-dried powder of bitter melon and its methanol fraction (MF) have potent liver triacylglycerol (TAG)- and cholesterol-lowering activities in rats fed diets with and without cholesterol supplementation [10, 11]. We have also previously examined whether effects of dietary MF on lipid metabolism are reproduced in hamsters fed diets with and without cholesterol supplementation, and found that MF had a potent and dose-dependent lipid-lowering activity in serum, but not in liver, especially in the animals fed diets with cholesterol [12]. Together these data show that bitter melon and its extracts, such as MF, contain bioactive components which favorably modulate TAG and cholesterol metabolism in experimental animals, although there are species-specific differences. To date, very little information has been reported about the mechanism(s) responsible for the reduced concentration of serum and/or hepatic lipids observed with bitter melon. Bitter melon juice appears to inhibit the synthesis and secretion of TAG in the form of apoB lipoprotein, as well as decrease microsomal transfer protein (MTP) mRNA expression in HepG2 cells [13, 14]. On the other hand, it has been reported that bitter melon extract activates the peroxisome proliferator-activated receptor (PPARa), and upregulates the expression of acyl CoA oxidase gene in rat hepatoma cells H4IIEC3 [15]. These observations suggest that the triglyceride-lowering effect of bitter melon extract(s) is due to a reciprocal response of TAG synthesis and fatty acid oxidation. There is also a possibility, as reported by Oishi et al. [16] that bitter melon-dependent hypotriglyceridemic action is exerted through an inhibitory effect on digestive enzyme activity in the intestinal tract. Regarding the mechanism(s) of the hypocholesterolemic and/or liver cholesterol-lowering effects of bitter melon, reduced cholesterol synthesis in the liver and/or increased fecal excretion of steroids seem plausible. However, so far, no information is available to confirm or refute this hypothesis. In the present studies, we demonstrated for the first time that dietary MF has favorable and independent effects on
123
Lipids (2012) 47:495–503
TAG and cholesterol metabolism. We also carried out a more detailed mechanistic examination of the effects of dietary MF on TAG and cholesterol metabolism, by using in vitro techniques of liver slices and hepatocytes in the presence of radiolabelled acetate and oleic acid (18:1n-6).
Materials and Methods The bitter melon methanol fraction (MF) was prepared according to the method described previously [10–12]. [1(2)-14C] sodium acetate (2.18 GBq/mmol) and [1-14C] oleic acid (18:1 n-9, 2.5 GBq/mmol) were purchased from Amersham Life Sciences (Buckinghamshire, England), and other chemicals were from Wako Chemicals (Osaka, Japan). Male Sprague–Dawley rats, purchased from a local breeder (Kyudo Co., Kumamoto, Japan), were acclimatized for several days in a temperature and light controlled room (22–24 °C, light on 0700–1900 h). During this period they were fed a pelleted stock chow (Type CE-2, Clea Japan, Tokyo, Japan). They were then divided into two groups (experiment 1) or four groups (experiment 2) with equal body weight (body weight 158 ± 2 g in experiment 1 and 86 ± 1 g in experiment 2). The control diet in experiment 1 was prepared according to the recommendations of the American Institute of Nutrition [17], and contained (by weight): casein, 20 %; corn oil, 5 %; vitamin mixture (AIN 76), 1 %; mineral mixture (AIN 76), 3.5 %; DL–methionine, 0.3 %; choline bitartrate, 0.2 %; cellulose, 5 %; corn starch, 15 %; and sucrose to 100 %. The control diet in experiment 2 was the same as in experiment 1, except for dietary fat levels. In experiment 2, diets were prepared with either 5 % fat (lard 4 % and corn oil 1 %) or 15 % fat (lard 14 % and corn oil 1 %). The fat level in the latter diet was increased at the expense of sucrose. The experimental diet was prepared by adding MF at the 1 % level to the control diets at the expense of sucrose in both experiment 1 and experiment 2. Rats had free access to the diets and water for 2 weeks in experiment 1, and for a longer period of 8 weeks in experiment 2. Food intake and body weight were recorded every other day. In experiment 1, approximately half of the animals were subjected to the liver slice experiments (seven rats for control and eight rats for experimental groups) and the others to hepatocyte experiments (six rats each for control and experimental groups). In experiment 1, the animals were decapitated between 9:00 a.m. and 12:00 p.m., blood was collected, and serum was harvested after centrifugation at 4 °C. The livers were removed immediately, rinsed, dried on filter paper, and weighed. A portion of each liver was cut into small blocks (1.5 9 1.5 9 1 cm), sliced (0.5 mm thickness) using a tissue slicer, and immediately transferred to opaque flasks containing ice-cold and oxygenated Krebs–Ringer-Bicarbonate buffer (pH 7.4) and
Lipids (2012) 47:495–503
[1(2)-14C] sodium acetate. Incubation was carried out for 2 h at 37 °C after flushing with a mixture of O2:CO2 (95:5, by vol), and was terminated by addition of 0.1 ml of acetic acid. In the hepatocyte experiments, the rats from control and experimental groups were anesthetized with pentobarbital sodium (64.8 mg/kg BW), and their livers were isolated by a two-step collagenase digestion method [18]. Briefly, anesthetized rats were cleaned and sanitized and placed inside a bio-safety cabinet. The hepatic portal vein was cannulated and the liver was perfused in situ with Krebs-Henseleit buffer (pH 7.4) using a peristaltic pump. A cut was made in the abdominal vena cava just above the level of the kidneys to have an outlet for the drainage of the perfusate. The perfusion was continued until all the lobes of the liver were visibly cleared of blood and blanched. Then the abdominal vena cava was clamped and the thoracic vena cava was cannulated through right atrium and a collagenase solution (0.05 % w/v) in Krebs–Henseleit buffer (pH 7.4) was re-circulated for 10–20 min until the liver lobes gets mildly dispersed. Liver lobes were collected into a sterile beaker with the buffer and sliced. The undispersed parts were filtered out using sterile gauze and the cells were collected into a centrifuge tube. Hepatocytes thus obtained were washed in Hanks buffer and re-suspended in KrebsHenseleit buffer (pH 7.4) with 10 mM glucose and 2 % bovine serum albumin (fatty acid free, Wako Chemical Co., Japan). All incubations were performed in duplicate for 1 h at 37 °C, in the presence of either [1(2)-14C] sodium acetate or [1-14C] oleic acid, under CO2:O2 (5:95, by vol) gas. After incubation, cells and medium were separated by low-speed centrifugation (600 rpm for 1 min), followed by washing three times with PBS; cells were then homogenized with a Teflon homogenizer and stored at -80 °C until further analysis for radioactivity in the lipid fractions. Cell viability, as determined by the trypan blue exclusion test, always exceeded 85 %. All experimental protocols and procedures were approved by the Animal Care and Use Committee of the University of Miyazaki, Japan. The lipids in the liver and serum were extracted and purified according to the method of Folch et al. [19]. TAG, cholesterol, and phospholipid contents in the lipid extract were measured chemically [10–12]. In the liver slice and hepatocyte experiments, lipids were also extracted by the method of Folch et al. [19] and various lipid fractions were separated by thin layer chromatography on Silica Gel 60G, with a solvent mixture of n-hexane/diethyl ether/acetic acid (80:20:1, by vol). The bands corresponding to cholesterol ester, TAG, free fatty acid, partial glycerides (mono- and di-glycerides), and phospholipid were identified with iodine vapor and were scraped into counting vials containing a toluene-based scintillation cocktail. Radioactivity derived from [1(2)-14C] acetate and [1-14C] oleic acid in
497
these lipid fractions was measured in a Wallac 1409 DSA liquid scintillation counter (Wallac Co., The Netherlands). For [1-14C] acetate incorporation into total fatty acid and digitonin precipitable sterol (DPS), the lipid extract was saponified in methanolic KOH at 75 °C for 30 min, and nonsaponifiable matter was extracted with n-hexane three times. After removal of the nonsaponifiable matter, the extracts were acidified with concentrated HCl, and the fatty acids were extracted with n-hexane three times. DPS were isolated and purified according to the procedure described previously [20, 21]. Radioactivities for total fatty acid were counted in a toluene-based scintillation cocktail and those for DPS were counted in 1 ml of methanol and Insta Gel. Feces collected for 2 days before the end of experiment 2 were lyophilized and analyzed for neutral and acidic steroids, as described in detail previously [22]. In experiment 2, the mRNA expression level of liver carnitine palmitoylacyltransferase I (CPT I) was also measured. Total RNA was extracted from 300 mg of liver, using a TRI ZOL Reagent (Invitrogen, Tokyo, Japan). Extracted RNA was treated with DNAse to remove any genomic DNA (gDNA) TaqManÒ Gene Expression Assays TM (Applied Biosystems, Tokyo, Japan), Assays-on-Demand , Gene Expression Products (Rn00580702_m1 for CPT I, Hs99999901_s1 for 18S RNA; Applied Biosystems, Tokyo, Japan) were used for quantitative real-time RT-PCR analysis of CPT I and 18S RNA expression in the liver. The manufacturer specifies that the assay probes spans an exon junction in the case of CPT I primers and that the assay will not detect gDNA in the case of 18S RNA primers. The amplification was performed with a real-time PCR system (ABI Prism 7000 Sequence Detection System, Applied Biosystems). Results were expressed as a relative value after normalization to 18S RNA expression. Data were analyzed by Student’s t test in experiment 1, and by one-way analysis of variance, followed by multiple comparisons with Tukey–Kramer’s test (StatView software, Sas Institute, USA) in experiment 2. The statistical significance of the difference of the means in these two experiments was evaluated at the level of p \ 0.05. In addition, the significance of the interaction of the two treatments was also analyzed by two-factor ANOVA (StatView software, Sas Institute, USA).
Results Effects of Dietary Bitter Melon Methanol Fraction (MF) on Lipid Metabolism After 2 Weeks of Administration (Experiment I) Effects of dietary MF, fed for 2 weeks, on growth parameters and the concentration of serum and liver lipids are
123
498
Lipids (2012) 47:495–503
Table 1 Effects of Momordica charantia methanol fraction (MF) on growth and lipid parameters Control (7)
Table 2 Effects of Momordica charantia methanol fraction (MF) on incorporation of [1(2)-14C] into lipids of liver slices Control (7)
1 % MF (8)
1 % MF (8)
Incorporation of [1(2)-14C]acetate into lipids (liver slice experiment)a
Growth parameter Initial body weight (g)
158 ± 2
158 ± 3
Triacylglycerol
5.14 ± 0.92
2.41 ± 0.81*
Final body weight (g)
275 ± 5
271 ± 4
Partial glycerides
0.97 ± 0.10
0.89 ± 0.24
Food intake (g/day)
21.4 ± 0.5
21.1 ± 0.5
Cholesterol
1.56 ± 0.21
1.82 ± 0.22
Liver weight (g/100 g BW)
5.76 ± 0.17
5.46 ± 0.09
Cholesterol ester
0.37 ± 0.02
0.38 ± 0.02
Phospholipids
4.49 ± 0.31
6.03 ± 0.61*
Free fatty acid
6.97 ± 0.93
10.1 ± 1.63
Serum lipids (mg/dL) Triacylglycerol Cholesterol
257 ± 27
260 ± 30
Total
86.8 ± 4.1
95.7 ± 4.6
Free
21.0 ± 2.0
22.0 ± 2.3
Ester (%)
75.8 ± 1.9
77.0 ± 2.0
HDL
44.6 ± 2.1
48.0 ± 1.7
218 ± 12
218 ± 7
52.2 ± 6.3
23.8 ± 2.8*
2.80 ± 0.11
2.11 ± 0.07*
Phospholipids
The rats were fed diets with and without MF for 2 weeks * Significantly different compared to the control group at p \ 0.05 a
Incorporation of radiolabelled acetate into lipid fractions of liver slices are expressed as kBq/g liver. The values are given as means ± SE of seven rats for control and eight rats for MF group
Liver lipids (mg/g liver) Triglyceride Cholesterol Total Free
1.39 ± 0.07
1.35 ± 0.04
Ester (%)
50.2 ± 2.3
35.4 ± 3.3*
Phospholipids
27.6 ± 1.5
29.5 ± 1.4
The values are given as means ± SE of seven to eight rats per group as indicated in parentheses. The rats were fed diets with and without MF for 2 weeks * Significantly different from the control group at p \ 0.05
summarized in Table 1. Although dietary MF had no effect on food intake and growth, relative liver weight tended to be lower in rats fed diets with MF than in those fed diets without MF. Effects of dietary MF on serum lipid parameters were marginal; however, dietary MF significantly decreased the concentration of hepatic TAG and total cholesterol by 54.4 and 24.6 %, respectively. The percentage of cholesterol ester in the liver was significantly lowered in rats fed diets with MF as compared to those fed diets without MF. However, the free cholesterol level remained unchanged between the groups, indicating that the marked reduction in hepatic total cholesterol in rats fed diets with MF is due to decreased cholesterol ester. Liver phospholipid concentration was comparable between the groups. Since hepatic lipid synthesis is one of the regulatory factors responsible for determining the lipid concentration, we first estimated hepatic lipid synthesis by measuring the rate of incorporation of radiolabelled acetate and oleic acid into lipids of both liver slices and hepatocytes prepared from rats fed diets supplemented with and without MF. As shown in Table 2, in the liver slice experiment, the incorporation of [1(2)-14C] acetate into TAG was 53.1 % lower in rats fed diets with MF compared to those fed diets
123
without MF. The extent of reduction in its incorporation was similar to that of the reduction in liver TAG concentration (54 % reduction in TAG concentration and 53 % reduction in incorporation of [1(2)-14C] acetate into TAG). On the other hand, incorporation of [1(2)-14C] acetate into free- and esterified-cholesterol was comparable between the groups, while incorporation of labeled acetate into phospholipids was significantly elevated (1.4-fold) following feeding of MF. Incorporation into free fatty acid fraction tended to be higher in rats fed diets with MF than in those fed diets without MF, but the difference was not significant. In the hepatocyte experiment, incorporation of [1(2)-14C] acetate into TAG was 47.3 % lower in liver cells prepared from rats fed diets with MF as compared to those without MF (Fig. 1), while incorporation into both free and esterified cholesterol was comparable between the groups (data not shown). Any effect of dietary MF on the incorporation of [1(2)-14C] acetate into free fatty acid fraction was also not apparent (data not shown). In this experiment, we further examined the effects of dietary MF on the metabolism of exogenous free fatty acid by using hepatocytes; for this, we used [1-14C] oleic acid as an exogenous substrate complexed with fatty acid free albumin, and administered to the cells prepared from livers of rats fed with and without dietary MF; dietary MF caused a significantly lowered incorporation of [1-14C] oleate into TAG fraction (21 % decrease relative to rats fed an MF-free diet) (Fig. 1). On the other hand, no significant differences were observed in incorporation into phospholipids and cholesterol esters between the two groups (data not shown). These results suggest that dietary MF appears to have a potent and specific inhibitory effect on the esterification step of TAG synthesis, but not on de novo synthesis of fatty acids.
Lipids (2012) 47:495–503
499
Fig. 1 Effect of bitter melon methanol extract (MF) on hepatic TAG synthesis (hepatocyte experiment). Hepatocytes from livers of rats fed diets with and without MF for 2 weeks were incubated in the presence of either [1(2)-14C] sodium acetate or [1-14C] oleic acid for 1 h. The values are giver as means ± SE of 6 rats per group. *Significantly different from the control group at p \ 0.05
Effects of Dietary MF on Lipid Metabolism After 8 Weeks of Administration (Experiment 2) It is known that the level and type of dietary fats are one of the modifying factors for the concentration of serum and liver lipids; in general, high fat diet compared to low fat causes elevated serum and liver lipids. In experiment 2, we examined the effects of dietary MF in rats fed on different fat levels (5 and 15 % fat level) and for prolonged feeding period of 8 weeks. Food intake was comparable in the four groups of rats, as shown in Table 3. However, the final body weight was significantly greater in rats fed high-fat diets than in those fed low-fat diets. Supplementation of MF in the high fat diet, but not in the low fat diet, tended to reduce weight gain. Relative liver weight tended to be lower in rats fed diets with MF, whether rats were fed the high-fat or the low-fat diet. Perirenal and epididymal adipose tissues were significantly heavier in rats fed high-fat diets than in those fed low-fat diets, although supplementation of MF did not influence these parameters. Serum TAG levels were significantly elevated in rats fed high-fat diets compared with those fed low fat diets. Dietary MF in combination with the high-fat diet, but not with the low-fat diet, reduced serum TAG. Neither the concentrations of total- and HDL-cholesterol nor that of phospholipids were influenced by the dietary fat levels or MF supplementation. Hepatic TAG concentrations were elevated by 1.5-fold in rats fed the high-fat diet compared to those fed the lowfat diet, while supplementation of MF in the low and highfat diet groups caused reductions in TAG concentration of 44.7 and 52.3 % respectively. The high-fat diet was also associated with an increased concentration of total cholesterol (1.4-fold elevation). Supplementation of MF in the low- and high-fat diets caused reductions of 19.0 and
40.6 %, respectively, in this lipid molecule. The reduction in the concentration of total cholesterol was due to decreases in the concentrations of both free and esterified cholesterol. Phospholipid concentrations were comparable among rats fed high-fat and low-fat diets containing MF and those fed diets without MF. Fecal excretion of neutral and acidic steroids is shown in Table 4. Fecal weights tended to be heavier in rats fed high-fat diets than in those fed low-fat diets, a further increase in fecal weights was noted in MF-supplemented rats. Although the excretion of bile acids into feces was not different among the groups, that of neutral steroids (coprostanol and cholesterol) was significantly increased in rats fed diets supplemented with MF. The extent of the increase was greater in the high-fat diet group (1.8-fold increase in low-fat diet and 2.3-fold increase in high-fat diet groups, respectively). In experiment 2, as in experiment 1, we estimated the hepatic lipid synthesis by measuring the incorporation of [1(2)-14C] acetate into various lipid fractions. Incorporation of labeled acetate into all lipid fractions in the liver was significantly lower in liver slices prepared from rats fed high-fat diets compared to those fed low-fat diets (Table 5). The extent of this reduction was 65.3 % for TAG and 48.8 % for phospholipids. Supplementation of MF in the low fat, but not the high fat diet group resulted in a typical profile of [1(2)-14C] acetate incorporation into various lipid fractions. Thus, dietary MF caused a significant reduction (45.7 %) in incorporation of labeled acetate into the TAG fraction but incorporation into other fractions (partial glycerides, free- and esterified-cholesterol, and phospholipids) was not affected. In addition, there was no change of incorporation into the free fatty acid fraction between the control and MF groups. To
123
500
Lipids (2012) 47:495–503
Table 3 Effect of dietary Momordica charantia methanol fraction (MF) on growth and serum lipid parameters in rats Low fat
High fat
ANOVA
Control (7)
MF (7)
Control (8)
MF (8)
Fat
MF
Interaction
Initial body weight (g)
85.4 ± 1.91
85.3 ± 1.61
86.1 ± 1.63
86.2 ± 1.74
NS
NS
NS
Final body weight (g)
483 ± 24.1a
506 ± 9.24a,b
558 ± 17.3b
523 ± 12.1a,b
p \ 0.05
NS
NS
Growth parameter
Food intake (g/day)
25.5 ± 1.21
25.3 ± 0.43
25.0 ± 1.22
23.7 ± 0.71
NS
NS
NS
Liver weight (g/100 g BW)
4.24 ± 0.14
4.11 ± 0.09
4.25 ± 0.16
4.04 ± 0.14
NS
NS
NS
Perirenal adipose (g/100 g BW)
2.50 ± 0.42
2.97 ± 0.22
3.51 ± 0.17
3.41 ± 0.34
p \ 0.05
NS
NS
Testicular adipose (g/100 g BW) Serum lipids (mg/dL)
1.92 ± 0.25
2.02 ± 0.11
2.38 ± 0.11
2.41 ± 0.16
p \ 0.05
NS
NS
185 ± 28a
297 ± 35a,b
385 ± 61b
287 ± 30a,b
p \ 0.05
NS
p \ 0.05
Triacylglycerols Cholesterol Total
120 ± 13.2
110 ± 11.4
101 ± 6.4
100 ± 6.1
NS
NS
NS
HDL
69.0 ± 5.92
71.0 ± 4.41
71.0 ± 3.52
72.3 ± 5.61
NS
NS
NS
233 ± 15.2
248 ± 13.1
226 ± 8.45
243 ± 12.1
NS
NS
NS
Phospholipids
The rats were fed diets with and without MF for 8 weeks and the values are mean ± SE of seven to eight rats per group as indicated in parentheses a,b Values not sharing common superscript letters are significantly different at p \ 0.05
Table 4 Effect of Momordica charantia methanol fraction (MF) on liver lipids and fecal steroid excretion in rats Low fat
High fat
ANOVA
Control (7)
MF (7)
Control (8)
MF (8)
Fat
MF
Interaction
60.9 ± 14.1a,b
33.7 ± 3.04a
93.2 ± 19.2b
44.5 ± 5.41a
NS
p \ 0.05
NS
3.36 ± 0.46a,b
2.72 ± 0.06a
4.61 ± 0.63b
2.74 ± 0.26a
NS
p \ 0.05
NS
1.29 ± 0.07b
NS
p \ 0.05
NS
Liver lipids (mg/g) Triacylglycerols Cholesterol Total
a
a,b
a,b
Free
1.85 ± 0.23
1.61 ± 0.10
1.70 ± 0.14
Ester (%)
41.8 ± 6.61
40.6 ± 3.93
60.0 ± 4.53
49.2 ± 6.71
p \ 0.05
NS
NS
Phospholipids Fecal steroid excretion
27.9 ± 1.12
28.1 ± 1.11
27.8 ± 0.62
29.7 ± 0.64
NS
NS
NS
Fecal weight (g/2d)
4.29 ± 0.17a
4.67 ± 0.28a,b
4.76 ± 0.23a,b
5.26 ± 0.19b
p \ 0.05
NS(0.051)
NS
7.67 ± 0.55
7.27 ± 1.18
8.93 ± 1.28
8.31 ± 1.61
NS
NS
NS
p \ 0.05
p \ 0.0001
NS
Steroid excretion (mg/2d) Bile acids Neutral steroids
a
13.4 ± 1.53
b
24.4 ± 2.01
14.2 ± 1.83
a
32.5 ± 2.31
c
The rats were fed diets with and without MF for 8 weeks and the values are mean ± SE of seven to eight rats per group as indicated in parentheses a,b,c
Values not sharing common superscript letters are significantly different at p \ 0.05
investigate this unexpected finding, we estimated de novo synthesis of fatty acid and cholesterol in liver slices by measuring incorporation of [1(2)-14C] acetate into total free fatty acid and DPS. Increased dietary fat levels significantly lowered the incorporation of radiolabelled acetate into the total fatty acids and DPS, while there were no significant differences in the incorporation of acetate into total fatty acids due to dietary MF (23.2 ± 3.88 kBq/g for low-fat without MF vs. 21.7 ± 1.53 kBq/g for low-fat
123
with MF, 10.6 ± 1.73 kBq/g for high-fat without MF vs. 15.1 ± 1.45 for high-fat with MF) and DPS (0.17 ± 0.02 kBq/g for low fat without MF vs. 0.23 ± 0.03 kBq/g for low fat with MF, and 0.05 ± 0.01 kBq/g for high fat without MF vs. 0.10 ± 0.02 kBq/g for high fat with MF). We also checked the expression of CPT I mRNA and found enhanced mRNA abundance following feeding of MF particularly in the low-fat group (data not shown).
Lipids (2012) 47:495–503
501
Table 5 Effect of dietary Momordica charantia methanol fraction (MF) on incorporation of [1(2)-14C] acetate into liver lipids in rats Low fat
High fat
Control (7)
MF (7)
ANOVA
Control (8)
MF (8)
Fat (p value)
MF
Interaction
Incorporation of [1(2)-14C] acetate into lipids in liver slices (KBq/g liver) Triglycerides
5.91 ± 1.31a
3.21 ± 0.42b
2.05 ± 0.33b
2.45 ± 0.43b
\0.05
NS
p \ 0.05
Partial glycerides
1.08 ± 0.18a
1.14 ± 0.09a
0.45 ± 0.08b
0.69 ± 0.14a,b
\0.05
NS
NS
Free cholesterol
0.15 ± 0.03a,b
0.26 ± 0.05a
0.09 ± 0.02b
0.15 ± 0.03a,b
\0.05
p \ 0.05
NS
a
b
0.15 ± 0.01a,b
\0.05
NS
NS
b
Cholesterol esters
0.23 ± 0.03
Phospholipids
6.61 ± 0.98a
7.08 ± 0.48a
3.38 ± 0.51b
4.65 ± 0.78a,b
\0.05
NS
NS
a
a
b
4.71 ± 0.95a,b
\0.05
NS
NS
Free fatty acid
5.95 ± 0.92
0.22 ± 0.02 7.03 ± 0.48
0.13 ± 0.02 2.81 ± 0.57
The rats were fed diets with and without MF for 8 weeks and the values are means ± SE of seven to eight rats as indicated in parentheses. Incorporation of radiolabelled acetate into lipid fractions are expressed as kBq/g liver a,b
Values not sharing common superscript letters in the same row are significantly different at p \ 0.05
Discussion In the present studies, we show that the methanol fraction (MF) prepared from bitter melon (Koimidori variety), at a 1.0 % dietary level, has potent liver TAG- and cholesterollowering activity in rats fed diets containing low (5 %) and high (15 %) fat. In a series of experiments with rats fed diets containing 5 % fat supplemented with or without cholesterol, we had previously reported that feeding of freezedried powder prepared from various varieties of bitter melon, especially that from Koimidori, at a 1.0 % dietary level, exhibits potent liver TAG- and cholesterol-lowering potentials, and that its major active principle(s) was localized almost exclusively in the fraction extracted by methanol from Koimidori freeze-dried powder [10, 11]. On the other hand, Ahmed et al. studied the effects of long term feeding (10 wk) of bitter melon juice on plasma and liver lipid profiles in normal and streptozotocin (STZ)-induced Type I diabetic rats, and found that this dietary ingredient exhibits a potent hypolipidemic effects in the STZ-induced diabetic rat [9]. In addition, they also observed the liver TAG-lowering effects of bitter melon juice in non-diabetic normal rat, but not in diabetic rat [9]. These results suggest that bitter melon and/or its extract may be effective in preventing fatty liver and hyperlipidemia caused by highcholesterol and high-fat diets, as well as the hyperlipidemia induced by diabetes mellitus. As most studies investigating the biological activity of bitter melon are focused on effects pertaining to plasma glucose and diabetes, or on the mechanism of its hypoglycemic activity [23–25], information regarding the mechanism(s) of its action on serum and/or hepatic lipids is sparse. The present studies therefore focused on understanding the mechanism(s) responsible for the dietary MFdependent reduction in the concentration of liver TAG and cholesterol. The liver is a major organ regulating the concentrations of serum and/or liver TAG and cholesterol,
through various synthetic and degradative processes (e.g., fatty acid synthesis and oxidation for determining the TAG level, and cholesterogenesis and its elimination into the feces as neutral and acidic steroids for cholesterol level). In the present studies, we calculated the de novo synthesis of lipids by measuring the rate of incorporation of radiolabelled acetate into lipids in liver slices and hepatocytes prepared from rats fed diets with and without MF. In addition, we calculated TAG synthesis by measuring the rate of esterification of exogenous [1-14C] oleate into the TAG fraction in hepatocytes prepared from these same animals. The appearance of radioactive carbon in hepatic lipids reflects the activation of acetate added to the liver slices and/or hepatocytes to its coenzyme A ester and their subsequent incorporation into glycerolipids and cholesterol. Thus, measuring the rate of incorporation of radioactive acetate into hepatic lipids in the liver slices and hepatocytes is an index of hepatic lipid synthesis and yields valuable information regarding this process, in spite of some limitations such as dilution [26, 27]. In addition, incorporation of [1(2)-14C] acetate and [1-14C] oleic acid into hepatic lipids into the liver slices and hepatocytes prepared from rats fed diets with and without MF indicates that these in vitro models of the liver function adequately with respect to lipid synthesis. As shown in Table 1, a marked decrease in TAG concentration in the livers of rats fed diets with MF was concomitantly associated with a decreased incorporation of [1(2)-14C] acetate into this lipid molecule; the extent of reduction was similar. Further, in experiment 2, this response was reproduced in the liver slices of rats fed, for a longer period of 8 weeks, diets with MF in combination with a low, but not with a high, fat content. On the other hand, incorporation of [1(2)-14C] acetate into free fatty acid fraction (Tables 2, 4) as well as that into the total fatty acids was not different between the liver slices prepared from rats fed diets with and without MF, suggesting that MF may not be an inhibitor of de novo
123
502
fatty acid synthesis. On the other hand, the decrease in the esterification process, as estimated by incorporation of labeled oleate into TAG in hepatocytes, suggested that MF exerts an inhibitory effect on TAG synthesis, rather than fatty acid synthesis. There is no data regarding the effect of bitter melon on fatty acid synthesis, but a few studies have been reported for TAG synthesis. For example, Nerurkar et al. recently reported that HepG2 cells treated with bitter melon juice exhibited reduced cellular TAG synthesis and secretion of TAG into the medium [13, 14]. They also observed that in HepG2 cells treated with bitter melon juice, mRNA expression of MTP was inhibited. MTP plays a pivotal role in the assembly and secretion of apoB-containing lipoproteins, and induces sterol regulatory elementbinding protein-Ic, which is one of the nuclear transcription factors responsible for modulating lipid and lipoprotein metabolism. The results of these in vitro studies further confirmed in vivo experiments showing that this juice lowers plasma apoB-100 and B-48 in C57BL/6 mice fed high-fat diets [14]. Recently, Chao and Huang observed that bitter melon extract activates PPARa: an important transcription factor involved in lipid and glucose homeostasis, and upregulates the expression of the acyl CoA oxidase gene in rat hepatoma cells H4IIEC3 [15]. Considering these previous observations in the context of our present results, the mechanism responsible for the observed reduction in the concentration of hepatic TAG by dietary MF is likely to be reduced synthesis of TAG and enhanced fatty acid oxidation. In these studies, we also measured mRNA expression of carnitine palmitoylacyltransferase I, which is the rate-limiting enzyme for hepatic mitochondrial fatty acid oxidation. We found, for the first time, that expression of CPT I is enhanced following feeding of MF, particularly in combination with a low fat diet. These results suggest that the diminished concentration of hepatic TAG is directly related to fatty acid oxidation, since an inverse relationship between hepatic TAG content and fatty acid oxidation has been shown [21, 22, 28]. Thus, the mechanism(s) responsible for the observed reduction in the concentration of hepatic TAG may, in part, be attributed to the concomitant elevation of CPT I expression. At present, there has been no direct demonstration that increased enzyme activity accompanies this increase in expression; further studies are needed to determine enzyme activity in animals fed diets with and without MF. The dietary MF-dependent reduction of hepatic total cholesterol was not accompanied by reduced cholesterol synthesis, as estimated by incorporation of [1(2)-14C] acetate into both free and esterified cholesterol Only the reduction in the concentration of cholesterol ester is attributed to decreased synthesis. As shown in Table 4, dietary MF increased fecal excretion of neutral steroids, but
123
Lipids (2012) 47:495–503
not of acidic steroids, in rats fed low- and high-fat diets, suggesting that MF may influence cholesterol metabolism at the initial stage by inhibiting the absorption of endogenous and exogenous cholesterol. Bitter melon has been reported to contain saponin as an active component [29, 30]; saponin is a known inhibitor of cholesterol absorption in the intestine [31, 32], and thus increases fecal excretion of neutral steroids. Thus, the MF-dependent reduction in the concentration of liver cholesterol may be a consequence of increased fecal excretion of neutral steroids, rather than of altered cholesterogenesis, although further studies are required to elucidate the mechanism. In summary, bitter melon has active components which may be able to ameliorate the severity of lipid disorders such as hyperlipidemia and fatty liver. The mechanism responsible for the action of bitter melon extracts such as MF is probably the reduced synthesis of liver TAG with a concomitant elevation of fatty acid oxidation. On the other hand, the MF-dependent reduction of liver cholesterol is due to an increase in fecal excretion of neutral sterols. Further studies aimed at characterization of the active component in MF and the precise mechanism(s) responsible for the observed reductions in the liver TAG and cholesterol should be carried out, and this work is in progress in our laboratory. Acknowledgments G. V. K. Senanayake would like to acknowledge and thank the Japanese Government and the University of Miyazaki for the provision of a Monbusho scholarship grant. The authors wish to thank Mitsuru Maruyama for her excellent technical assistance. We also thank Dr. Melissa Clark in the department of Comparative Biosciences, College of Veterinary Medicine, University of Illinois, Urbana-Champaign, IL, USA, for her valuable comments and for revision of the English in the manuscript. Conflict of interest interest.
The authors declare that there is no conflicts of
References 1. Meir P, Yaniv Z (1985) An in vitro study on the effect of Momordica charantia on glucose uptake and glucose metabolism in rats. Planta Med 51:12–16 2. Raman A, Lau C (1996) Anti-diabetic properties and phytochemistry of Momordica charantia L. (Cucurbitaceae). Phytomed 2:349–362 3. Platel K, Srinivasan K (1997) Plant foods in the management of diabetes mellitus: vegetables as potential hypoglycemic agents. Nahrung 41:68–74 4. Sitasawad SL, Shewade Y, Bhonde R (2000) Role of bittergourd fruit juice in STZ-induced diabetic state in vivo and in vitro. J Ethnopharmacol 73:71–79 5. McCarty MF (2004) Does bitter melon contain an activator of AMP-activated kinase? Med Hypotheses 63:340–343 6. Grover JK, Yadav SP (2004) Pharmacological actions and potential uses of Momordica charantia: a review. J Ethnopharmacol 93:123–132
Lipids (2012) 47:495–503 7. Krawinkel MB, Keding GB (2006) Bitter gourd (Momordica charantia): a dietary approach to hyperglycemia. Nutr Rev 64:331–337 8. Yin JY, Zhang H, Ye J (2008) Traditional Chinese medicine in treatment of metabolic syndrome. Endo Metab Immune Disord Drug Targ 8:99–111 9. Ahmed I, Lakhani MS, Gillett M, John A, Raza H (2001) Hypotriglyceridemic and hypocholesterolemic effects of anti-diabetic Momordica charantia (karela) fruit extract in streptozotocininduced diabetic rats. Diabetes Res Clin Pract 51:155–161 10. Jayasooriya AP, Sakono M, Yukizaki C, Kawano M, Yamamoto K, Fukuda N (2000) Effects of Momordica charantia powder on serum glucose levels and various lipid parameters in rats fed with cholesterol-free and cholesterol-enriched diets. J Ethnopharmacol 72:331–336 11. Senanayake GVK, Maruyama M, Shibuya K, Sakono M, Fukuda N, Morishita T, Yukizaki C, Kawano M, Ohta H (2004) The effect of bitter melon (Momordica charantia) on serum and liver triglyceride levels in rats. J Ethnopharmacol 91:257–262 12. Senanayake GVK, Maruyama M, Sakono M, Fukuda N, Morishita T, Yukizaki C, Kawano M, Ohta H (2004) The effects of bitter melon (Momordica charantia) extracts on serum and liver lipid parameters in hamsters fed cholesterol-free and cholesterolenriched diets. J Nutr Sic Vitaminol 50:253–257 13. Nerurkar PV, Pearson L, Efird JT, Adeli K, Therault AG, Nerurkar VR (2005) Microsomal triglyceride transfer protein gene expression and apo B secretion are inhibited by bitter melon in HepG2 cells. J Nutr 135:702–705 14. Nerurkar PV, Lee YK, Motosue M, Adeli K, Nerurkar VR (2008) Momordica charantia (bitter melon) reduces plasma apolipoprotein B-100 and increases hepatic insulin receptor substrate and phosphoinositide-3 kinase interactions. Br J Nutr 100:751–759 15. Chao C-Y, Huang C-J (2003) Bitter gourd (Momordica charantia) extract activates peroxisome proliferator-activated receptors and upregulates the expression of the acyl CoA oxidase gene in H4IIEC3 hepatoma cells. J Biomed Sci 10:782–791 16. Oishi Y, Sakamoto T, Udagawa H, Taniguchi H, KobayashiHattori K, Ozawa Y, Takita T (2007) Inhibition of increases in blood glucose and serum neutral fat by Momordica charantia saponin fraction. Biosci Biotechnol Biochem 71:735–740 17. American Institute of Nutrition (1977) Report of the American Institute of Nutrition ad hoc committee on standards for nutritional studies. J Nutr 107:1340–1348 18. Seglen PO (1973) Preparation of rat liver cells. III. Enzymatic requirements for tissue dispersion. Exp Cell Res 82:391–398 19. Folch J, Lees M, Sloane-Stanley GH (1957) A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226:497–509
503 20. Fukuda N, Azain MJ, Ontko JA (1982) Altered hepatic metabolism of free fatty acids underlying hypersecretion of very low density lipoproteins in the genetically obese Zucker rat. J Biol Chem 257:14066–14072 21. Fukuda N, Ontko JA (1984) Interactions between fatty acid synthesis, oxidation, and esterification in the reduction of triglyceride-rich lipoproteins by the liver. J Lipid Res 25:831–842 22. Fukuda N, Machida M, Oku H, Chinen I (1989) Hypocholesterolemic activity of bagasse alkaline extract in rats fed a high cholesterol diet. Agric Biol Chem 53:3097–3098 23. Chen Q, Chan LLY, Li ETS (2003) Bitter melon (Momordica charantia) reduces adiposity, lower serum insulin and normalizes glucose tolerance in rats fed a high fat diet. J Nutr 133:1088–1093 24. Shih C-C, Lin C-H, Wu J-B (2009) Momordica charantia extract on insulin resistance and the skeletal muscle GLUT4 protein in fructose-fed rats. J Ethnopharmacol 123:82–90 25. Fuangchan A, Sonthisombat P, Seubnukarn T et al (2011) Hypoglycemic effect of bitter melon compared with metformin in newly diagnosed type 2 diabetes patients. J Ethnopharmacol 134:422–428 26. Dietschy JM, McGarry JD (1974) Limitations of acetate as a substrate for measuring cholesterol synthesis in liver. J Biol Chem 249:52–58 27. Fears R, Morgan B (1976) Studies on the response of cholesterol biogenesis to feeding in rats. Evidence against the existence of diurnal rhythms. Biochem J 158:53–60 28. Fukuda N, Hidaka T, Toda T, Sugano M (1990) Altered hepatic metabolism of free fatty acid in rats fed a threonine-imbalanced diet. J Nutr Sci Vitaminol 36:467–474 29. Ng TG, Wong CM, Li WW, Teubg HW (1986) A steryl glycoside fraction from Momordica charantia seeds with an inhibitory action on lipid metabolism in vitro. Biochem Cell Biol 64:766–771 30. Matsuda H, Shimoda H, Uemura T, Yoshikawa M (1999) Preventive effect of sesquiterpenes from bay leaf on blood ethanol elevation in ethanol-loaded rat: structure requirement and suppression of gastric emptying. Bioorg Med Chem Lett 20:2647–2652 31. Malinaow MR, McLaughlin P, Papworth L, Stafford C, Kohler GO, Livingston AL, Cheeke PR (1977) Effect of alfalfa saponins on intestinal cholesterol absorption in rats. Am J Clin Nutr 30:2061–2067 32. Francis G, Kerem Z, Makkar HPS, Becker K (2002) The biological action of saponins in animal systems: a review. Br J Nutr 88:587–605
123
