- Email: [email protected]
Metformin improves nonalcoholic fatty liver disease in obese mice
April 2000
102 LONG·TERM ALCOHOL EXPOSURE CHANGES SENSITIVITY OF RAT KUPFFER CELLS TO LIPOPOLYSACCHARIDE. Nobuyuki Enomoto, Hiroshi Kono, Peter Schemmer, Ayako Enomoto, Kenichi Ikejima, Miyoko Hirose, Tsuneo Kitamura, Yoshiyuki Takei, Nobuhiro Sato, David A. Brenner, Ronald G. Thurman, Dept of Gastroenterology, Juntendo Univ, Tokyo, Japan; Dept of Pharmacology, Univ of North Carolina, Chapel Hill, NC; Dept of Medicine, Univ of North Carolina, Chapel Hill, NC; Univ of North Carolina, Chapel Hill, NC. We previously reported that acute ethanol treatment enhances Kupffer cell sensitivity to lipopolysaccharide (LPS) (Gastroenterology 1998; 115: 443), and established a new model of alcoholic liver injury based on the sensitization of Kupffer cells. In this model, 4 weeks after ethanol, CD 14 in Kupffer cells was elevated significantly (Hepatology 1999; 29: 1680). Moreover, PGE 2 produced by activated Kupffer cells participated in the mechanism of ethanol-induced fatty liver (Gastroenterology 1998; 114: AI238). However, it remains obscure if long-term ethanol exposure affects Kupffer cell sensitization to LPS. This study was designed to elucidate the temporal effect of chronic ethanol exposure on Kupffer cell sensitization to LPS. Methods: Rats were given ethanol (5 g/kg body weight) every 24 hours intragastrically for up to 12 weeks, and Kupffer cells were isolated 24 hours after the final ethanol administration and cultured in RPMI 1640 with 10 % PBS. Following addition of LPS (loong/ml) to Kupffer cells, [Ca 2 +]i was measured using a microspectrofluorometer with the fluorescent indicator, fura-2. PGE 2 and TNF-a were measured by ELISA. CDI4 was evaluated by Western analysis. Results: Four weeks after ethanol, CDI4 in Kupffer cells was elevated about 2-fold followed by a gradual decrease returning to a level comparable with controls at 12 weeks. The LPS-induced increase in [CaHl; and TNF-a by Kupffer cells were also increased 2-fold over control values after 4 weeks. Triglyceride content increased with the duration of chronic ethanol treatment for up to 8 weeks. At 8 weeks, PGE 2 produced by Kupffer cells was increased about 3-fold over control value (43 ::': 5 pg/106 cells/4 hours) and triglyceride increased by about 4-fold (Control; 6.0 ::': 0.8, 8 weeks; 26.4 ::': 3.1 mg/g liver) followed by a gradual decrease to basal level. Interestin~ly, after a 12 weeks of ethanol exposure, LPS-induced increase in [Ca 2 ]i and TNF-a production were diminished by about 50% from peak. Liver triglyceride content at 12 weeks was reduced significantly compared to values at 8 weeks (12 weeks; 14.4 ::': 1.3 mg/g liver). Conclusions: Kupffer cells at the early stage of chronic ethanol exposure exhibited sensitization to LPS while decreased sensitivity was observed later. This correlated with triglyceride accumulation in the liver. These data indicate that long-term alcohol exposure changes sensitivity of rat Kupffer cells to LPS.
103 ANTmIOTIC TREATMENT PREVENTS INCREASED KUPFFER CELL INTRACELLULAR SUPEROXIDE INDUCED BY ACUTE ALCOHOL INGESTION. Edward R. Abril, Patricia A. Thome, Hana Holubec, Robert S. McCuskwy, David L. Earnest, Univ of Arizona, Tucson, AZ. We previously reported that acute oral doses of alcohol (a binge) activate liver Kupffer cells (KC). This is accompanied by increased KC intracellular production and extracellular release of superoxide (Oi). Increased KC 0i is thought to contribute to liver injury. Others have shown that toxicant or alcohol (EtOH) induced liver injury in rats can be blunted by pertreatment with galolinium chloride to destroy KC or antibiotics to reduce colonic bacterial production of endotoxin (ET) a stimulant of KC. Since humans chronically ingesting EtOH may have increased blood ET and occasionally binge drink EtOH that could activate KC, we evaluated if treatment with antibiotics could prevent the increase in KC O2 induced by an alcohol binge. Male SO rats (250-330g) wre treated with Ciprofloxacin (IOmg/kg) or Biaxin (40mg/kg) one hr before a binge (no=6/gp) with alcohol (4g/kg) or water q 12 hrs x 5 doses. KC were prepared by centrifugal elutriation after liver perfusion with pronase/collagenase. KC intracellular 0i was quantitated by chemiliuminesence with a luminometer. Results: An alcohol binge significantly increased (p
104 METFORMIN IMPROVES NONALCOHOLIC FATTY LIVER DIS· EASE IN OBESE MICE. Christine Chuckaree, Shi Qi Yang, HuiZhi Lin, Anna Mae Diehl, Johns Hopkins, Baltimore, MD. Background:Type-2 diabetes is prevalent in obesity and both conditions are thought to be major risk factors for nonalcoholic fatty liver diseases (NAFLD). Because the prevalence of diabetes increases with liver disease progression, diabetes may playa role in NAFLD pathogenesis. However, it is also possible that diabetes is a consequence, rather than a cause, of NAFLD. Aim:To evaluate the role of diabetes in the pathogenesis of obesity-related NAFLD by examining the effects of an oral hypoglycemic,
AASLDA919
metformin, in a mouse model of obesity-related NAFLD. Methods: Obese, adult male, ob/ob mice with fatty livers and hyperglycemia were divided into 3 groups. One group (M. n-- l I) received metformin (350 ug/g) mixed in a nutritionally- replete liquid diet for 4 wks. The effects of M on body weight, epididymal fat mass, liver histology, serum liver enzymes (ALT, AP), glucose and triglycerides were compared to 2 untreated control groups, PF (no=8), which was pair-fed the same amount of liquid diet as group M, and AL (no=3), which was allowed liquid diet ad libitum. Results: Initially, the groups had similar body weights, serum glucoses and triglycerides. During the study, diet intake was greater in AL (29::':2 cal/d) than in M and PF (23::':I and 23::':4 cal/d). However, the rate of weight gain (I g/wk) and final body weight (67 g) were no different in PF and AL; and greater in these controls than in M (- O.5g/wk; final weight 60 g), p<.OI. Although caloric intake was similar in M- and PF-mice, adiposity decreased only in M; final mean epididymal fat mass was I.5::':0.1g in M vs. 2.2::':O.2gin PF and I.8::':0.1g in AL (p<.05). Surprisingly, neither M nor food restriction improved hyperglycemia. Final serum glucoses were similar among M (222::':45), PF (203::':26), and AL (233::':60). However,liver enzymes and histology were significantly improved by M. Final ALT was 85::':36 in M vs. 279::':90 in PF and 331 ::':53 in AL (p<.OOl). Final AP in M (67::':19) was also less than in PF (200::':68) and AL (224) (p<.oo5). Moreover, metformin abolished micro- and macro-vesicular steatosis, while steatosis remained 3+ in PF- and 4+ in AL-Iivers. Decreased hepatic steatosis did not correlate with changes in serum triglyceride levels, which remained similar among the groups. Conclusions: In doses that are ineffective at reducing hyperglycemia, metformin significantly improves NAFLD in obese, diabetic mice. Thus, in this model, diabetes does not cause NAFLD. Improvements in NAFLD correlate with decreased central adiposity, suggesting that metformin's beneficial actions may be related to its effects on lipid metabolism.
105 SEX DIFFERENCES OF HEPATIC INJURY AND CYTOKINE PRODUCTION IN CONCANAVALIN A·INDUCED HEPATITIS IN MICE. S. Takamoto, K. Nakamura, Y. Nakade, M. Okada, K. Aso, M. Yoneda, I. Makino, Second Dept of Internal Medicine, Asahikawa Med Coli, Asahikawa, Japan; Dept of Gastroenterology, Dokkyou Univ Sch of Medicine, Dokkyou, Japan; Second Dept of Internal Medicine, Asahikawa Med Coli, Asahikawa, Jamaica. Concanavalin A (Con A) activates T lymphocytes and causes T cellmediated hepatic injury in mice. Tumor necrosis factor a (TNF-a) and interferon y (IFN-y) are key mediators in this experimental model. Since autoimmune diseases are predominant in women, female is considered to have enhanced immune responses and T cell functions. Purpose: To investigate sex differences of liver injury and cytokine production in Con A-induced hepatitis. Methods: Study I: Male and female BALB/c mice (7-8 weeks old) were intravenously given either Con A (10, 20 or 30 mg/kg) or saline vehicle. Animals were killed 2, 8, or 24 hours after the treatment and serum was obtained. Serum ALT levels were determined enzymatically and serum TNF-a and IFN-y concentrations were measured by ELISA. Study 2: Male and female BALB/c mice (3 weeks old) were castrated or ovariectomized. Con A (20 mg/kg) was intravenously administered to animals 4 weeks after the treatment and they were sacrificed 8 hours after Con A injection. Serum ALT, TNF-a, and IFN-y levels were measured. Results: Study I: Serum ALT levels 24 hours after Con A injection were significantly elevated in female than in male (Mean ::': SE, KUIL: saline 60 ::': 20; 10 mg/kg 108 ::': 51; 20 mg/kg 153 ::': 52; 30 mg/kg 817 ::': 487 in male vs saline 36 ::': 7; 10 mg/kg 158 ::': 37; 20 mg/kg 837 ::': 229; 30 mg/kg 2478 :': 534 in female). Serum TNF-a concentrations were significantly higher in female than in male (Mean :': SE, pg/ml: 2 h 586 ::': 146; 8 h 23 ::': 23 in male vs 2 h 1087 ::': 212; 8 h 184 ::': 47 in female). Serum IFN-y concentrations were also significantly higher in female (Mean::': SE, pg/ml: 2 h 536 ::': 71; 8 h 701 :': 102 in male vs 2 h 958 ::': 65; 8 h 1372 :': 118 in female). Study 2: Serum ALT levels were significantly elevated in castrated male mice compared with sham operated animals (Mean:': SE, KUIL: 4149 ::': 1029 in castrated vs 390 ::': 112 in sham operated). Serum TNF-a and IFN-y concentrations were significantly higher in castrated male mice than in sham operated animals (Mean ::': SE, pg/ml: TNF-a 491 :': 68; IFN-y 1250 ::': 84 in castrated vs TNF-a 229 ::': 29; IFN-y 830 ::': 107 in sham operated). In contrast, a significant decrease in serum ALT and TNF-a concentrations was found in ovariectomized mice compared with sham operated animals (Mean:': SE: ALT 907 ::': 406; TNF-a 203 ::': 7 in ovariectomized vs ALT 1980 ::': 678; TNF-a 248 ::': IO in sham operated), but serum IFN-y levels had no difference between ovariectomized and sham operated animals. Conclusion: Con A induces more severe hepatic injury in female mice than in male and sex hormonal effects on cytokine production may playa role in gender-associated difference in Con A-induced hepatitis.
102 LONG·TERM ALCOHOL EXPOSURE CHANGES SENSITIVITY OF RAT KUPFFER CELLS TO LIPOPOLYSACCHARIDE. Nobuyuki Enomoto, Hiroshi Kono, Peter Schemmer, Ayako Enomoto, Kenichi Ikejima, Miyoko Hirose, Tsuneo Kitamura, Yoshiyuki Takei, Nobuhiro Sato, David A. Brenner, Ronald G. Thurman, Dept of Gastroenterology, Juntendo Univ, Tokyo, Japan; Dept of Pharmacology, Univ of North Carolina, Chapel Hill, NC; Dept of Medicine, Univ of North Carolina, Chapel Hill, NC; Univ of North Carolina, Chapel Hill, NC. We previously reported that acute ethanol treatment enhances Kupffer cell sensitivity to lipopolysaccharide (LPS) (Gastroenterology 1998; 115: 443), and established a new model of alcoholic liver injury based on the sensitization of Kupffer cells. In this model, 4 weeks after ethanol, CD 14 in Kupffer cells was elevated significantly (Hepatology 1999; 29: 1680). Moreover, PGE 2 produced by activated Kupffer cells participated in the mechanism of ethanol-induced fatty liver (Gastroenterology 1998; 114: AI238). However, it remains obscure if long-term ethanol exposure affects Kupffer cell sensitization to LPS. This study was designed to elucidate the temporal effect of chronic ethanol exposure on Kupffer cell sensitization to LPS. Methods: Rats were given ethanol (5 g/kg body weight) every 24 hours intragastrically for up to 12 weeks, and Kupffer cells were isolated 24 hours after the final ethanol administration and cultured in RPMI 1640 with 10 % PBS. Following addition of LPS (loong/ml) to Kupffer cells, [Ca 2 +]i was measured using a microspectrofluorometer with the fluorescent indicator, fura-2. PGE 2 and TNF-a were measured by ELISA. CDI4 was evaluated by Western analysis. Results: Four weeks after ethanol, CDI4 in Kupffer cells was elevated about 2-fold followed by a gradual decrease returning to a level comparable with controls at 12 weeks. The LPS-induced increase in [CaHl; and TNF-a by Kupffer cells were also increased 2-fold over control values after 4 weeks. Triglyceride content increased with the duration of chronic ethanol treatment for up to 8 weeks. At 8 weeks, PGE 2 produced by Kupffer cells was increased about 3-fold over control value (43 ::': 5 pg/106 cells/4 hours) and triglyceride increased by about 4-fold (Control; 6.0 ::': 0.8, 8 weeks; 26.4 ::': 3.1 mg/g liver) followed by a gradual decrease to basal level. Interestin~ly, after a 12 weeks of ethanol exposure, LPS-induced increase in [Ca 2 ]i and TNF-a production were diminished by about 50% from peak. Liver triglyceride content at 12 weeks was reduced significantly compared to values at 8 weeks (12 weeks; 14.4 ::': 1.3 mg/g liver). Conclusions: Kupffer cells at the early stage of chronic ethanol exposure exhibited sensitization to LPS while decreased sensitivity was observed later. This correlated with triglyceride accumulation in the liver. These data indicate that long-term alcohol exposure changes sensitivity of rat Kupffer cells to LPS.
103 ANTmIOTIC TREATMENT PREVENTS INCREASED KUPFFER CELL INTRACELLULAR SUPEROXIDE INDUCED BY ACUTE ALCOHOL INGESTION. Edward R. Abril, Patricia A. Thome, Hana Holubec, Robert S. McCuskwy, David L. Earnest, Univ of Arizona, Tucson, AZ. We previously reported that acute oral doses of alcohol (a binge) activate liver Kupffer cells (KC). This is accompanied by increased KC intracellular production and extracellular release of superoxide (Oi). Increased KC 0i is thought to contribute to liver injury. Others have shown that toxicant or alcohol (EtOH) induced liver injury in rats can be blunted by pertreatment with galolinium chloride to destroy KC or antibiotics to reduce colonic bacterial production of endotoxin (ET) a stimulant of KC. Since humans chronically ingesting EtOH may have increased blood ET and occasionally binge drink EtOH that could activate KC, we evaluated if treatment with antibiotics could prevent the increase in KC O2 induced by an alcohol binge. Male SO rats (250-330g) wre treated with Ciprofloxacin (IOmg/kg) or Biaxin (40mg/kg) one hr before a binge (no=6/gp) with alcohol (4g/kg) or water q 12 hrs x 5 doses. KC were prepared by centrifugal elutriation after liver perfusion with pronase/collagenase. KC intracellular 0i was quantitated by chemiliuminesence with a luminometer. Results: An alcohol binge significantly increased (p
104 METFORMIN IMPROVES NONALCOHOLIC FATTY LIVER DIS· EASE IN OBESE MICE. Christine Chuckaree, Shi Qi Yang, HuiZhi Lin, Anna Mae Diehl, Johns Hopkins, Baltimore, MD. Background:Type-2 diabetes is prevalent in obesity and both conditions are thought to be major risk factors for nonalcoholic fatty liver diseases (NAFLD). Because the prevalence of diabetes increases with liver disease progression, diabetes may playa role in NAFLD pathogenesis. However, it is also possible that diabetes is a consequence, rather than a cause, of NAFLD. Aim:To evaluate the role of diabetes in the pathogenesis of obesity-related NAFLD by examining the effects of an oral hypoglycemic,
AASLDA919
metformin, in a mouse model of obesity-related NAFLD. Methods: Obese, adult male, ob/ob mice with fatty livers and hyperglycemia were divided into 3 groups. One group (M. n-- l I) received metformin (350 ug/g) mixed in a nutritionally- replete liquid diet for 4 wks. The effects of M on body weight, epididymal fat mass, liver histology, serum liver enzymes (ALT, AP), glucose and triglycerides were compared to 2 untreated control groups, PF (no=8), which was pair-fed the same amount of liquid diet as group M, and AL (no=3), which was allowed liquid diet ad libitum. Results: Initially, the groups had similar body weights, serum glucoses and triglycerides. During the study, diet intake was greater in AL (29::':2 cal/d) than in M and PF (23::':I and 23::':4 cal/d). However, the rate of weight gain (I g/wk) and final body weight (67 g) were no different in PF and AL; and greater in these controls than in M (- O.5g/wk; final weight 60 g), p<.OI. Although caloric intake was similar in M- and PF-mice, adiposity decreased only in M; final mean epididymal fat mass was I.5::':0.1g in M vs. 2.2::':O.2gin PF and I.8::':0.1g in AL (p<.05). Surprisingly, neither M nor food restriction improved hyperglycemia. Final serum glucoses were similar among M (222::':45), PF (203::':26), and AL (233::':60). However,liver enzymes and histology were significantly improved by M. Final ALT was 85::':36 in M vs. 279::':90 in PF and 331 ::':53 in AL (p<.OOl). Final AP in M (67::':19) was also less than in PF (200::':68) and AL (224) (p<.oo5). Moreover, metformin abolished micro- and macro-vesicular steatosis, while steatosis remained 3+ in PF- and 4+ in AL-Iivers. Decreased hepatic steatosis did not correlate with changes in serum triglyceride levels, which remained similar among the groups. Conclusions: In doses that are ineffective at reducing hyperglycemia, metformin significantly improves NAFLD in obese, diabetic mice. Thus, in this model, diabetes does not cause NAFLD. Improvements in NAFLD correlate with decreased central adiposity, suggesting that metformin's beneficial actions may be related to its effects on lipid metabolism.
105 SEX DIFFERENCES OF HEPATIC INJURY AND CYTOKINE PRODUCTION IN CONCANAVALIN A·INDUCED HEPATITIS IN MICE. S. Takamoto, K. Nakamura, Y. Nakade, M. Okada, K. Aso, M. Yoneda, I. Makino, Second Dept of Internal Medicine, Asahikawa Med Coli, Asahikawa, Japan; Dept of Gastroenterology, Dokkyou Univ Sch of Medicine, Dokkyou, Japan; Second Dept of Internal Medicine, Asahikawa Med Coli, Asahikawa, Jamaica. Concanavalin A (Con A) activates T lymphocytes and causes T cellmediated hepatic injury in mice. Tumor necrosis factor a (TNF-a) and interferon y (IFN-y) are key mediators in this experimental model. Since autoimmune diseases are predominant in women, female is considered to have enhanced immune responses and T cell functions. Purpose: To investigate sex differences of liver injury and cytokine production in Con A-induced hepatitis. Methods: Study I: Male and female BALB/c mice (7-8 weeks old) were intravenously given either Con A (10, 20 or 30 mg/kg) or saline vehicle. Animals were killed 2, 8, or 24 hours after the treatment and serum was obtained. Serum ALT levels were determined enzymatically and serum TNF-a and IFN-y concentrations were measured by ELISA. Study 2: Male and female BALB/c mice (3 weeks old) were castrated or ovariectomized. Con A (20 mg/kg) was intravenously administered to animals 4 weeks after the treatment and they were sacrificed 8 hours after Con A injection. Serum ALT, TNF-a, and IFN-y levels were measured. Results: Study I: Serum ALT levels 24 hours after Con A injection were significantly elevated in female than in male (Mean ::': SE, KUIL: saline 60 ::': 20; 10 mg/kg 108 ::': 51; 20 mg/kg 153 ::': 52; 30 mg/kg 817 ::': 487 in male vs saline 36 ::': 7; 10 mg/kg 158 ::': 37; 20 mg/kg 837 ::': 229; 30 mg/kg 2478 :': 534 in female). Serum TNF-a concentrations were significantly higher in female than in male (Mean :': SE, pg/ml: 2 h 586 ::': 146; 8 h 23 ::': 23 in male vs 2 h 1087 ::': 212; 8 h 184 ::': 47 in female). Serum IFN-y concentrations were also significantly higher in female (Mean::': SE, pg/ml: 2 h 536 ::': 71; 8 h 701 :': 102 in male vs 2 h 958 ::': 65; 8 h 1372 :': 118 in female). Study 2: Serum ALT levels were significantly elevated in castrated male mice compared with sham operated animals (Mean:': SE, KUIL: 4149 ::': 1029 in castrated vs 390 ::': 112 in sham operated). Serum TNF-a and IFN-y concentrations were significantly higher in castrated male mice than in sham operated animals (Mean ::': SE, pg/ml: TNF-a 491 :': 68; IFN-y 1250 ::': 84 in castrated vs TNF-a 229 ::': 29; IFN-y 830 ::': 107 in sham operated). In contrast, a significant decrease in serum ALT and TNF-a concentrations was found in ovariectomized mice compared with sham operated animals (Mean:': SE: ALT 907 ::': 406; TNF-a 203 ::': 7 in ovariectomized vs ALT 1980 ::': 678; TNF-a 248 ::': IO in sham operated), but serum IFN-y levels had no difference between ovariectomized and sham operated animals. Conclusion: Con A induces more severe hepatic injury in female mice than in male and sex hormonal effects on cytokine production may playa role in gender-associated difference in Con A-induced hepatitis.