|Home | About | Journals | Submit | Contact Us | Français|
In this study, we investigated the effects of the major green tea polyphenol, (−)-epigallocatechin-3-gallate (EGCG), on high-fat–induced obesity, symptoms of the metabolic syndrome, and fatty liver in mice. In mice fed a high-fat diet (60% energy as fat), supplementation with dietary EGCG treatment (3.2 g/kg diet) for 16 wk reduced body weight (BW) gain, percent body fat, and visceral fat weight (P < 0.05) compared with mice without EGCG treatment. The BW decrease was associated with increased fecal lipids in the high-fat–fed groups (r2 = 0.521; P < 0.05). EGCG treatment attenuated insulin resistance, plasma cholesterol, and monocyte chemoattractant protein concentrations in high-fat–fed mice (P < 0.05). EGCG treatment also decreased liver weight, liver triglycerides, and plasma alanine aminotransferase concentrations in high-fat–fed mice (P < 0.05). Histological analyses of liver samples revealed decreased lipid accumulation in hepatocytes in mice treated with EGCG compared with high-fat diet-fed mice without EGCG treatment. In another experiment, 3-mo-old high-fat–induced obese mice receiving short-term EGCG treatment (3.2 g/kg diet, 4 wk) had decreased mesenteric fat weight and blood glucose compared with high-fat–fed control mice (P < 0.05). Our results indicate that long-term EGCG treatment attenuated the development of obesity, symptoms associated with the metabolic syndrome, and fatty liver. Short-term EGCG treatment appeared to reverse preexisting high-fat–induced metabolic pathologies in obese mice. These effects may be mediated by decreased lipid absorption, decreased inflammation, and other mechanisms.
The rates of obesity, defined as BMI ≥30 kg/m2, have increased dramatically in the United States in the past 20 y and especially in the last 10 y (1). There is a strong positive association between obesity and type II diabetes, cardiovascular disease, and hypertension (2). These associations describe the metabolic syndrome, a clustering of risk factors including abdominal obesity, insulin resistance, and dyslipidemia. Metabolic syndrome is also often characterized by chronic inflammation and hepatic steatosis (3).
Green tea is consumed worldwide, especially in East Asian countries. Green tea contains caffeine and polyphenolic compounds known as catechins. The most abundant catechin found in green tea is (−)-epigallocatechin-3-gallate (EGCG),7 which has been suggested to be responsible for many of the potential health effects of tea (4,5).
In 1999, Dulloo et al. (6) found that administration of a green tea extract significantly increased energy expenditure and fat oxidation in a group of young males. Since then, several clinical trials have reported the effects of tea preparations on increasing energy expenditure, fat oxidation, weight loss, fat mass, and weight maintenance after weight loss (7–9). Nevertheless, whether these effects are due to catechins or caffeine has yet to be resolved. Several studies in rodent models showed that green tea extract decreased weight gain and body fat gain (10,11). In 2005, it was reported that treatment with TEAVIGO, a green tea extract containing ≥94% EGCG and ≤0.1% caffeine, significantly reduced body weight (BW) and body fat in different strains of mice fed a high-fat diet (12,13).
In addition to its weight loss effects, there are studies that have suggested that tea consumption may alleviate other metabolic abnormalities related to obesity. Several clinical investigations showed that tea treatment reduced fasting blood glucose and improved glucose tolerance in healthy and diabetic subjects (14,15). A recent clinical trial reported that when healthy subjects were given an extract of green, black, and mulberry tea concurrently with a high-starch and -lipid meal, carbohydrate absorption was significantly blunted (16).
There are several studies describing the beneficial effects of tea constituents in animal models of the metabolic syndrome (15,17). One study reported that oral administration of EGCG for 3 wk significantly reduced blood pressure and increased insulin sensitivity in spontaneously hypertensive rats (18). A recent study by Wolfram et al. (19) reported that TEAVIGO treatment significantly improved glucose tolerance and insulin sensitivity in db/db mice.
Hepatic steatosis (or nonalcoholic fatty liver disease) and chronic low-grade inflammation are 2 conditions that are associated with obesity and the metabolic syndrome (3,20). Green tea consumption has been shown to be inversely correlated with liver damage (a consequence of progressive hepatic steatosis) and with markers of inflammation in humans (21,22). In high-fat–fed C57BL/6J and leptin-deficient ob/ob obese mice, green tea treatment reduced liver lipids and markers of liver damage (10,23).
The effects of physiologically relevant doses of EGCG on obesity and high-fat–induced pathologies associated with the metabolic syndrome have not been systematically studied. The goals of this study were to determine the effect of dietary EGCG on weight gain, markers of insulin resistance, hepatic steatosis, and inflammatory markers in a high-fat diet-induced mouse model for obesity and the metabolic syndrome. We studied both the long-term effects of EGCG in growing high-fat–fed mice and the short-term effects of EGCG in 3-mo-old obese mice.
EGCG was a gift from Mitsui Norin. Commercial low-fat diet (LF; 10% energy as fat), high-fat diet (HF; 60% energy as fat), and high-fat plus EGCG (HFE; 60% energy as fat, 3.2 g EGCG/kg diet) diet (Table 1) were prepared by Research Diets. Briefly, each ingredient was weighed (within 0.5% of amount needed) and blended to homogeneity in the appropriate powdered diet. The mixture was then formed into equally sized pellets and placed into a temperature- and humidity-controlled room to remove the moisture from the diet.
All animal experiments were conducted according to protocols approved by the Institutional Animal Care and Use Committee at Rutgers University (Piscataway, NJ). One hundred male C57BL/6J mice (age 5−6 wk) from either the departmental breeding colony or Jackson Laboratories were used in 2 long-term experiments. The first long-term treatment (Expt. 1, n = 38, 6 wk old) was designed to determine the effects of EGCG on BW and body fat in high-fat–fed mice. The 2nd long-term treatment (Expt. 2, n = 56, 5 wk old) was designed to confirm the findings on BW and fat from Expt. 1 and also to determine the effects of EGCG on fecal lipids, hyperglycemia, insulin resistance, markers of inflammation, and fatty liver caused by a high-fat diet. In both studies, mice were fed a LF, HF, or HFE diet for 16 wk. Based on allometric scaling, the presently used dose of 3.2 g/kg diet in mice corresponds to 10 200-mL cups of green tea (containing 2 g tea leaves per cup) per day for an average person requiring 8374 kJ (2000 kcal)/d. BW and food intake (on a per-cage basis) were measured weekly. For fecal lipid measurements from Expt. 2, cage bedding was changed during wk 10 and feces were collected from the cages the following day.
To determine the short-term effects of EGCG on high-fat–fed mice (Expt. 3), 5- to 6-wk-old male C57BL/6J mice were fed either a LF (n = 9) or HF (n = 18) diet for 9 wk, after which, 10 mice from the HF group were randomly picked and switched to the HFE diet. Mice were treated with diets for an additional 4 wk. BW and food intake (on a per cage basis) were measured weekly.
On the day of blood glucose measurement, food was removed and the cage bedding was changed (to minimize coprophagy) 7−8 h before measurement. Blood was collected from the tail vein and glucose concentrations were measured with the OneTouch Basic glucose monitor (LifeScan). Blood glucose (from food-deprived mice) was measured 6 times throughout the course of Expt. 2. For Expt. 3, blood glucose was measured 3 times throughout the course of the study, twice before mice were transferred to the HFE diet (at wk 4 and 7 of study) and once at the end of the study (at wk 13).
For Expt. 1, mice were food deprived for 6 h prior to killing after 16 wk of treatment. Mice were weighed and killed by CO2 overdose. Total percent body fat was measured postmortem by dual-energy X-ray absorptiometry using a PIXImus densitometer (Lunar GE Medical Systems). Visceral adipose tissues (mesenteric, epididymal, and retroperitoneal depots) were harvested, rinsed, and weighed.
For Expt. 2, mice were food deprived and killed as described above. Whole blood was obtained by cardiac puncture. Visceral adipose tissues (mesenteric, epididymal, and retroperitoneal depots) and liver were harvested, rinsed, and weighed. Fatty liver was initially diagnosed by altered coloration (pink color due to lipid accumulation) and later confirmed by histological analyses (criteria described below). One lobe of the liver was fixed in 10% formalin, the rest was snap-frozen on dry ice. Plasma was isolated by centrifugation at 700 × g at room temperature; 15 min, and snap-frozen on dry ice. All frozen samples were stored at −80°C.
For Expt. 3, all mice from the LF, HF, and HFE groups were weighed and killed by CO2 overdose after 13 wk of treatment. Visceral adipose tissue (mesenteric, epididymal, and retroperitoneal depots) were harvested, rinsed, and weighed.
All samples analyzed in food-deprived mice were from Expt. 2. Plasma insulin was determined by ELISA (Millipore). Plasma cholesterol was determined by a Cholesterol E kit (Wako Diagnostics). Plasma monocyte chemoattractant protein 1 (MCP-1) was determined by ELISA (R&D Systems). Plasma tumor necrosis factor-α and interleukin-6 (IL-6) were determined by ELISA (Invitrogen). Plasma alanine aminotransferase (ALT) concentrations were determined spectrophotometrically using the ALT DiscretePak from Catachem.
The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated using the final blood glucose and insulin determinations in food-deprived mice from Expt. 2. This method is widely used to estimate insulin resistance in humans and animal models (24). The following formula was applied (25):
For determination of fecal lipids, feces (0.9−1.3 g) collected during wk 10 of Expt. 2 were weighed, added to 3−4 mL deionized water, and allowed to sit at 4°C overnight. The following day, the feces mixture was homogenized by vortexing. Following vortexing, lipids were extracted with methanol:chloroform (2:1, v:v) using a previously described method (26). The lipophilic layer from the extraction was collected and dried under vacuum. Total lipids were measured gravimetrically.
For determination of hepatic triglycerides in mice from Expt. 2, liver tissue (50−100 mg) was homogenized in 2 mL isopropanol with a Polytron disrupter. The homogenate was centrifuged at 2000 × g; 10 min, and the supernatant was collected. Triglycerides in the supernatant were measured with a Triglyceride H kit (Wako Diagnostics).
Embedded liver tissue blocks from Expt. 2 were cut into 6-micron sections and stained with hematoxylin and eosin. Mounting medium and cover slips were placed on the slides, which were then left to dry overnight. A diagnosis of fatty liver was made based on the presence of macro- or microvesicular fat in >5% of the hepatocytes in a given slide.
Statistical analyses were conducted using GraphPad Prism software. Oneway ANOVA with Tukey's post hoc test was used for statistical analysis of BW gain, body fat, and biochemical measurements. For analysis of BW and blood glucose over the treatment time period, repeated-measures ANOVA with Bonferroni's post hoc test was used. A chi-square test was performed for analysis of fatty liver incidence. Pearson correlation coefficient was used for the correlation analysis between fecal lipids and BW change. All data are presented as means ± SEM. Significance was assigned at P < 0.05.
Daily food intake per cage did not differ among groups in either long-term experiment (data not shown). For Expt. 1 (Fig. 1A) and Expt. 2 (Fig. 1C), HF mice were significantly heavier than LF mice by wk 3−4 of treatment. These trends remained throughout the course of the treatment period. By the end of the 16-wk treatment period, HFE mice weighed significantly less than the HF mice (Fig. 1A,C). BW gain was 33−41% lower in HFE mice (Expt. 1:, 14.02 ± 2.67 g; Expt. 2, 19.21 ± 2.06 g) compared with HF mice (Expt. 1, 23.80 ± 0.11 g; Expt. 2, 30.48 ± 0.90; P < 0.05).
Results from the dual-energy X-ray absorptiometry analysis showed that HF mice exhibited significantly increased percent body fat (49.95 ± 0.98%) compared with LF mice (19.16 ± 1.36%). HFE mice had significantly lower percent body fat (to 44.77 ± 2.06%) compared with HF mice. HF mice had significantly increased total visceral adipose weight compared with LF mice in both Expts. 1 (Fig. 1B) and 2 (Fig. 1D) For Expt. 1, EGCG treatment reduced the weight of the mesenteric adipose depot (19% decrease) and retroperitoneal adipose depot (21% decrease) in HF mice (Fig. 1B). In Expt. 2, EGCG treatment reduced total visceral adipose tissue weight by 37% in HF mice (P < 0.05; Fig. 1D). This was attributable to weight decreases in the mesenteric (48% decrease), epididymal (28% decrease), and retroperitoneal (34% decrease) depots (P < 0.05 for all depots; Fig. 1D).
In Expt. 2, HF mice had higher blood glucose compared with LF mice by wk 3 of treatment (P < 0.05. Fig. 1E). By wk 10, HFE mice had lower blood glucose than HF mice (25% decrease; Fig. 1E) and this trend was significant over the 16-wk treatment period.
HF mice had increased plasma insulin concentrations compared with LF mice (P < 0.05; Table 2) and HFE mice had reduced plasma insulin compared with HF mice (61% reduction; Table 2). HOMA-IR calculations (using the final blood glucose and insulin concentrations) showed that the HF diet increased the HOMA-IR index compared with LF mice (P < 0.05; Table 2), whereas HFE mice exhibited a 76% decrease compared with HF mice (P < 0.05; Table 2).
HF mice had significantly higher plasma cholesterol and MCP-1 concentrations than LF mice in Expt. 2 (Table 2). EGCG treatment attenuated the elevation in cholesterol (by 24%) and MCP-1 (by 69%) in HF mice (Table 2; P < 0.05). Plasma tumor necrosis factor-α and IL-6 were not detectable in mice (using ELISA methods) from any of the treatments (data not shown).
At wk 10 of Expt. 2, fecal lipid concentrations did not differ between the HF (17.73 ± 1.42 mg lipids/g feces) and LF (11.33 ± 1.35 mg lipids/g feces) groups but were greater in HFE mice (43.23 ± 10.77 mg lipids/g feces) compared with either LF or HF mice (P < 0.05). The increase in fecal lipids was inversely correlated with the decrease in BW in mice from high-fat–fed groups (r2 = 0.521; P < 0.05).
HF mice had significantly increased liver size (relative to BW) compared with LF mice in Expt. 2. These mice also had a significantly higher incidence of fatty liver (Table 2; Fig. 2A), higher liver triglycerides (>2-fold of LF concentrations), and elevated plasma ALT concentrations (6-fold of LF concentrations). EGCG treatment reduced incidence of fatty liver (P < 0.05; Table 2; Fig. 2A), liver size (P < 0.05; 22% decrease), liver triglycerides (P < 0.05; 69% decrease), and plasma ALT concentrations (P < 0.05; 67%, decrease) in HF mice (Table 2).
Histopathological analyses showed increased lipid deposition in the livers of HF mice (Fig. 2C) compared with those of LF mice (Fig. 2B). Microvesicular fat was present throughout the liver in HF mice; however, it had a central lobular distribution. The periportal zone was relatively free of accumulated lipids. Livers from HFE (Fig. 2D) mice showed a marked decrease in liver lipid accumulation compared with HF mice.
Daily food intake per cage did not differ among groups in Expt. 3. After 9 wk of a LF or HF diet, HF mice were significantly heavier than the LF mice (P < 0.05; Fig. 3A). EGCG treatment for 4 wk tended to decrease BW gain (Fig. 3A,B).
After 9 wk of LF or HF diet, HF mice had increased total visceral fat weight compared with LF mice (P < 0.05; Fig. 3C). EGCG treatment for 4 wk decreased the weight of the mesenteric depot by 36% compared with HF-fed mice (P < 0.05; Fig. 3C).
By wk 4 of the study, HF mice had elevated blood glucose concentrations compared with LF mice (P < 0.05; Fig. 3D). At wk 13 (after 4 wk of EGCG treatment), HF mice still had higher blood glucose concentrations than LF mice (P < 0.05; respectively, Fig. 3D), but the HFE-treated mice had 22% lower blood glucose than HF mice (P < 0.05; Fig. 3D).
In the present study, 16-wk dietary EGCG treatment significantly decreased BW in mice fed a high-fat diet. This was evident in both long-term experiments (Expts. 1 and 2). The BW gain was larger in Expt. 2 compared with Expt. 1; this may have been due to the fact that the mice in Expt. 2 were younger at the start of treatment.
Previous studies have shown that EGCG treatment (5−10 g EGCG/kg diet) reduced BW in mice fed a high-fat diet (12,13). The presently used dose of 3.2 g EGCG/kg diet in mice, which corresponds to 10 200-mL cups of green tea (containing 2 g tea leaves per cup) per day, may represent a more attainable dose than those previously reported in weight loss studies with animal models. Future studies are needed to determine whether there is a dose-response effect of EGCG on weight gain produced by a high-fat diet.
We found that 16-wk EGCG treatment also significantly decreased total percent body fat and visceral body fat weight. The decrease in visceral body fat by EGCG was apparent in the mesenteric, epididymal, and retroperitoneal depots, with the largest decrease occurring in the mesenteric depot. These decreases in body fat may be due to EGCG inhibiting lipid absorption or increasing fat oxidation. Indeed, we found that long-term EGCG treatment increased fecal lipids with EGCG treatment compared with high-fat–fed control mice, supporting the hypothesis that EGCG decreased lipid absorption.
Although there was no significant change in BW gain or total visceral fat weight in diet-induced obese mice treated with EGCG for 4 wk, there was significant decrease in the weight of the mesenteric depot of visceral fat. Previous studies showed that EGCG significantly reversed total body fat percentage, subcutaneous fat weight, and epididymal fat weight in high-fat–fed mice (12,13). Our study clearly demonstrated the effects of EGCG on the mesenteric and retroperitoneal adipose depots. Previously, weight loss studies in humans showed that visceral adipose tissue is more metabolically active than subcutaneous adipose tissue, suggesting the increased risk for metabolic syndrome by adipose tissue is depot specific (27). This finding is supported by studies that showed visceral fat reduction was more effective at decreasing risk for metabolic syndrome than reduction of the subcutaneous depot in obese humans (28). A recent study in high-fat–fed mice found that mesenteric adipose tissue produced significantly higher levels of MCP-1 compared with the subcutaneous, epididymal, or renal depots in obese mice, which may be related to increased risk for related metabolic conditions (29). Our data, showing significantly lower plasma MCP-1 by EGCG treatment in high-fat–fed mice, suggest that EGCG-mediated decreases in mesenteric and retroperitoneal adipose tissue weight may play a role in the effects of EGCG on high-fat–induced inflammation and the development of metabolic syndrome.
The metabolic syndrome is a grouping of pathological conditions related to obesity and insulin resistance. We found that EGCG significantly decreased blood glucose, insulin, and insulin resistance in high-fat–fed mice. Furthermore, short-term treatment with EGCG reversed the effects of a high-fat diet on blood glucose. A previous study showed that EGCG (10 g TEAVIGO/kg diet) reduced plasma insulin in high-fat–fed C57BL/6J mice (19). However, these measurements were taken in the fed state; an effect of variable food intake on insulin levels in previous studies cannot be ruled out.
The effects of EGCG on insulin resistance in our study are at least partially attributable to the observed decrease in weight gain and body fat. There may, however, be some direct effects of EGCG on improvement of glucose homeostasis. Previous studies showed that green tea supplementation increased muscle glucose transporter protein expression in insulin-resistant rats (17). A study by Koyama et al. (30) found that EGCG treatment (1.5 g/kg diet) for 7 d significantly decreased the expression of the gluconeogenic enzymes phosphoenolpyruvate carboxykinase and glucose-6-phosphatase in the livers of mice. Studies have also shown decreased expression of phosphoenolpyruvate carboxykinase and glucose-6-phosphatase by EGCG in rat hepatoma cells (5,31). These studies suggest that green tea and EGCG improve insulin sensitivity and glucose homeostasis, in part by directly increasing glucose disposal into the muscle and decreasing gluconeogenesis in the liver.
The most interesting finding in our study was that EGCG treatment decreased high-fat–induced hepatic steatosis. This was demonstrated histologically and biochemically. The decreased plasma ALT by EGCG also reflects decreased high-fat–induced hepatocyte injury.
The effect of EGCG on hepatic steatosis may contribute to its effects on insulin resistance. Hepatic steatosis is associated with insulin resistance in humans and studies in rodent models have shown that suppression of hepatic steatosis improved hepatic insulin sensitivity (32). These effects may be mediated through the activation of protein kinase C, which is stimulated by fatty acids and contributes to the phosphorylation of the insulin receptor substrate at Ser-307, resulting in inhibition of normal insulin signaling (32,33). EGCG has been shown to both inhibit protein kinase C and increase insulin receptor substrate-mediated downstream signaling in cell line studies (31).
On the other hand, the effects of EGCG on insulin resistance may partially contribute to the prevention of hepatic steatosis. Insulin resistance causes increased plasma FFA and enhances liver uptake of fatty acids, which are then stored as triglycerides (34). It is possible that EGCG improves insulin sensitivity and consequently reduces the risk for hepatic steatosis. Further studies are needed to determine the causal relationship between insulin resistance and hepatic steatosis, and the role of EGCG in attenuating both of these conditions.
We also observed that EGCG decreased plasma MCP-1 in high-fat–fed mice and this is a new finding, to our knowledge. A previous study found that EGCG treatment reduced MCP-1 expression in endothelial cells exposed to phorbol 12-myristate 13-acetate, a chemical inducer of inflammation (35). A recent study by Kanda et al. (36) showed that MCP-1 is involved in the recruitment of macrophages into the adipose tissue, which leads to chronic inflammation. It is now widely accepted that obesity-induced inflammation plays a key role in the development of metabolic syndrome (37). Kanda et al. (36) reported that MCP-1 contributed to the development of insulin resistance. The results from the present study suggest that reduction of MCP-1 may be one mechanism by which EGCG reduces insulin resistance. On the other hand, the elevated MCP-1 may be a result of the increased body fat caused by the high-fat diet and the reduction of MCP-1 by EGCG treatment may be secondary to the decrease in adipose tissue. Future studies are needed to clearly determine whether EGCG elicits its metabolic effects by attenuation of inflammation.
The present study showed that EGCG significantly reduced plasma total cholesterol that was elevated by a high-fat diet containing cholesterol (Table 2). These results are consistent with previous studies that showed that orally administered EGCG decreased plasma cholesterol and inhibited cholesterol absorption from the intestine of rodents (38). The volume of plasma collected at harvest was a limiting factor in these studies; future studies can determine whether EGCG improves other plasma lipids related to metabolic syndrome, such as plasma triglycerides and HDL-cholesterol.
These studies show that a physiologically relevant dose of dietary EGCG reduces the development of obesity, hyperglycemia, insulin resistance, hypercholesterolemia, and hepatic steatosis in high-fat–fed mice. They also suggest that EGCG treatment may reverse the effects of a high-fat diet on BW and blood glucose. These effects may be related to decreased fat absorption and antiinflammatory effects mediated by EGCG. Other possible mechanisms such as decreased fatty acid synthesis and increased fatty acid oxidation need further investigation. It will also be interesting to determine how EGCG restores glucose homeostasis and attenuates fatty liver disease. Finally, it will be important to clearly determine whether our presently observed effects occur in humans and whether EGCG or green tea consumption can be used as a tool to prevent the development of obesity and its comorbidities.
1Supported by NIH grant CA88961 (to C.S.Y.), American Institute for Cancer Research grant 05A047 (to J.D.L.), NIH grant CA125780 (to J.D.L.), and by facilities supported by NIEHS Center Grant ES05022.
2Author disclosures: M. Bose, J. D. Lambert, J. Ju, K. R. Reuhl, S. A. Shapses, and C. S. Yang, no conflicts of interest.
7Abbreviations used: ALT, alanine aminotransferase; BW, body weight; EGCG, (−)-epigallocatechin-3-gallate; HF, high-fat; HFE, high-fat plus EGCG; HOMA-IR, homeostasis model assessment for insulin resistance; IL-6, interleukin-6; LF, low-fat; MCP-1, monocyte chemoattractant protein 1.