Free fatty acids and IL-6 induce adipocyte galectin-3 which is increased in white and brown adipose tissues of obese mice.

Galectin-3 regulates immune cell function and clearance of advanced glycation end products. Galectin-3 is increased in serum of obese humans and mice and most studies suggest that this protein protects from inflammation in metabolic diseases. Current data show that galectin-3 is markedly elevated in the liver, subcutaneous and intra-abdominal fat depots of mice fed a high fat diet and ob/ob mice. Galectin-3 is also increased in brown adipose tissues of these animals and immunohistochemistry confirms higher levels in adipocytes. Raised galectin-3 in obese white adipocytes has been described in the literature and regulation of adipocyte galectin-3 by metabolites with a role in obesity has been analyzed. Galectin-3 is expressed in 3T3-L1 fibroblasts and human preadipocytes and is modestly induced in mature adipocytes. In 3T3-L1 adipocytes galectin-3 is localized in the cytoplasm and is also detected in cell supernatants. Glucose does not alter soluble galectin-3. Lipopolysaccharide has no effect while TNF reduces and IL-6 raises this lectin in cell supernatants. Palmitate and oleate modestly elevate soluble galectin-3. Differentiation of 3T3-L1 cells in the presence of 100 µM and 200 µM linoleate induces soluble galectin-3 and cellular levels are upregulated by the higher concentration. Current data suggest that free fatty acids and IL-6 increase galectin-3 in adipocytes and thereby may contribute to higher levels in obesity.

Manganese superoxide dismutase knock-down in 3T3-L1 preadipocytes impairs subsequent adipogenesis.

Adipogenesis is associated with the upregulation of the antioxidative enzyme manganese superoxide dismutase (MnSOD) suggesting a vital function of this enzyme in adipocyte maturation. In the current work, MnSOD was knocked-down with small-interference RNA in preadipocytes to study its role in adipocyte differentiation. In mature adipocytes differentiated from these cells, proteins characteristic for mature adipocytes, which are strongly induced in late adipogenesis like adiponectin and fatty acid-binding protein 4, are markedly reduced. Triglycerides begin to accumulate after about 6 days of the induction of adipogenesis, and are strongly diminished in cells with low MnSOD. Proteins upregulated early during differentiation, like fatty acid synthase and cytochrome C oxidase-4, are not altered. Cell viability, insulin-mediated phosphorylation of Akt, antioxidative capacity (AOC), superoxide levels, and heme oxygenase 1 with the latter being induced upon oxidative stress are not affected. L-Buthionine-(S,R)-sulfoximine (BSO) depletes glutathione and modestly lowers AOC of mature adipocytes. Addition of BSO to 3T3-L1 cells 3 days after the initiation of differentiation impairs triglyceride accumulation and expression of proteins induced in late adipogenesis. Of note, proteins that increased early during adipogenesis are also diminished, suggesting that BSO causes de-differentiation of these cells. Preadipocyte proliferation is not considerably affected by low MnSOD and BSO. These data suggest that glutathione and MnSOD are essential for adipogenesis.

Free fatty acids, lipopolysaccharide and IL-1α induce adipocyte manganese superoxide dismutase which is increased in visceral adipose tissues of obese rodents.

Excess fat storage in adipocytes is associated with increased generation of reactive oxygen species (ROS) and impaired activity of antioxidant mechanisms. Manganese superoxide dismutase (MnSOD) is a mitochondrial enzyme involved in detoxification of ROS, and objective of the current study is to analyze expression and regulation of MnSOD in obesity. MnSOD is increased in visceral but not subcutaneous fat depots of rodents kept on high fat diets (HFD) and ob/ob mice. MnSOD is elevated in visceral adipocytes of fat fed mice and exposure of differentiating 3T3-L1 cells to lipopolysaccharide, IL-1α, saturated, monounsaturated and polyunsaturated free fatty acids (FFA) upregulates its level. FFA do not alter cytochrome oxidase 4 arguing against overall induction of mitochondrial enzymes. Upregulation of MnSOD in fat loaded cells is not mediated by IL-6, TNF or sterol regulatory element binding protein 2 which are induced in these cells. MnSOD is similarly abundant in perirenal fat of Zucker diabetic rats and non-diabetic animals with similar body weight and glucose has no effect on MnSOD in 3T3-L1 cells. To evaluate whether MnSOD affects adipocyte fat storage, MnSOD was knocked-down in adipocytes for the last three days of differentiation and in mature adipocytes. Knock-down of MnSOD does neither alter lipid storage nor viability of these cells. Heme oxygenase-1 which is induced upon oxidative stress is not altered while antioxidative capacity of the cells is modestly reduced. Current data show that inflammation and excess triglyceride storage raise adipocyte MnSOD which is induced in epididymal adipocytes in obesity.

LDL but not HDL increases adiponectin release of primary human adipocytes.

Adipocytes in obesity have inappropriately low cholesterol while adiponectin release is reduced. Cholesterol shortage may contribute to low adiponectin and 3T3-L1 cells treated with lovastatin have diminished adiponectin in cell supernatants. LDL and HDL deliver cholesterol to adipocytes. LDL but not HDL increases adiponectin in cell supernatants of primary human adipocytes. The effect of LDL is not blocked by receptor associated protein suggesting that members of the LDL-receptor family are not involved. To evaluate whether these in vitro observations translate into changes in systemic adiponectin, adiponectin was measured in serum of three patients before, immediately after and 3d after LDL-apheresis. Whereas circulating lipoproteins are reduced immediately after apheresis adiponectin is not changed. Therefore, acute lowering of lipoproteins does not affect systemic adiponectin also excluding that plenty of adiponectin is bound to lipoprotein particles. Accordingly, levels of adiponectin in purified lipoproteins are quite low. Familial hypobetalipoproteinemia (FHBL) is a rare disorder associated with low plasma LDL. Serum adiponectin is, however, similar compared to healthy controls. Thus, neither LDL nor HDL directly contributes to circulating adiponectin concentrations.

Manganese superoxide dismutase is reduced in the liver of male but not female humans and rodents with non-alcoholic fatty liver disease.

Non-alcoholic fatty liver disease (NAFLD) is among the most common liver diseases. Oxidative stress is one of the pathogenic mechanisms contributing to the progression of simple fatty liver to non-alcoholic steatohepatitis (NASH). Manganese superoxide dismutase (MnSOD) is a mitochondrial antioxidative enzyme and here its expression in rodent and human NAFLD has been analyzed. MnSOD is found reduced in the liver of male mice fed a high fat diet and male ob/ob mice. Female mice fed an atherogenic diet to induce NASH have MnSOD protein levels comparable to controls. In a cohort of 30 controls, 41 patients with fatty liver and 39 NASH patients, MnSOD mRNA is significantly lower in the steatotic and NASH liver. When analyzed in both genders separately reduction of MnSOD expression is only found in males. Here, MnSOD mRNA negatively correlates with steatosis grade but not with extent of fibrosis or inflammation. MnSOD is, however, not reduced in primary human hepatocytes (PHH) treated with palmitate or oleate to increase cellular triglycerides. Lipopolysaccharide, TNF, IL-6, TGFß and leptin which are all raised in NAFLD do not affect MnSOD in PHH. Adiponectin which attenuates oxidative stress partly by increasing MnSOD in macrophages does not induce MnSOD in PHH. In summary, current data show that hepatic MnSOD is reduced in male but not female humans and rodents with NAFLD.

Chemerin is highly expressed in hepatocytes and is induced in non-alcoholic steatohepatitis liver.

Chemerin is a recently described adipokine whose adipose tissue and serum levels are increased in obesity. Chemerin is expressed in the liver, and here, expression of chemerin has been studied in liver cells and in non-alcoholic fatty liver disease (NAFLD) which is more often found in obesity. Chemerin is shown to be highly expressed in primary human hepatocytes (PHH) whereas hepatic stellate cells (HSC) produce only low levels of this protein. In mice fed a high fat diet hepatic chemerin mRNA but not protein is increased. Chemerin protein is comparably expressed in the liver of control animals and ob/ob mice. Rodents fed a Paigen diet or methionine-choline deficient diet (MCD) develop non-alcoholic steatohepatitis (NASH), and liver chemerin protein tends to be higher in the first and is significantly increased in the latter. Of note, MCD fed mice have similar serum chemerin levels as the respective control animals despite lower body weight. In human fatty liver and NASH liver chemerin mRNA also tends to be induced. Cytokines like TNF and adipokines with an established role in NASH do not considerably affect PHH chemerin protein. The antidiabetic drug metformin reduces cellular and soluble chemerin in PHH as has already been described in adipose tissue. In conclusion current data show that primary human hepatocytes are a major source of hepatic chemerin and increased liver chemerin in NASH may even contribute to systemic levels.

Adiponectin receptor 1 C-terminus interacts with PDZ-domain proteins such as syntrophins.

Adiponectin receptor 1 (AdipoR1) is one of the two signaling receptors of adiponectin with multiple beneficial effects in metabolic diseases. AdipoR1 C-terminal peptide is concordant with the consensus sequence of class I PSD-95, disc large, ZO-1 (PDZ) proteins, and screening of a liver yeast two hybrid library identified binding to ß2-syntrophin (SNTB2). Hybridization of a PDZ-domain array with AdipoR1 C-terminal peptide shows association with PDZ-domains of further proteins including ß1- and α-syntrophin (SNTA). Interaction of PDZ proteins and C-terminal peptides requires a free carboxy terminus next to the PDZ-binding region and is blocked by carboxy terminal added tags. N-terminal tagged AdipoR1 is more highly expressed than C-terminal tagged receptor suggesting that the free carboxy terminus may form a complex with PDZ proteins to regulate cellular AdipoR1 levels. The C- and N-terminal tagged AdipoR1 proteins are mainly localized in the cytoplasma. N-terminal but not C-terminal tagged AdipoR1 colocalizes with syntrophins in adiponectin incubated Huh7 cells. Adiponectin induced hepatic phosphorylation of AMPK and p38 MAPK which are targets of AdipoR1 is, however, not blocked in SNTA and SNTB2 deficient mice. Further, AdipoR1 protein is similarly abundant in the liver of knock-out and wild type mice when kept on a standard chow or a high fat diet. In summary these data suggest that AdipoR1 protein levels are regulated by so far uncharacterized class I PDZ proteins which are distinct from SNTA and SNTB2.

Portal vein omentin is increased in patients with liver cirrhosis but is not associated with complications of portal hypertension.

BACKGROUND: Omentin is a visceral fat-derived adipokine associated with endothelium-dependent vasodilation. Impaired endothelial function is a major cause of portal hypertension in liver cirrhosis. The aim was to assess associations of omentin with systemic markers of endothelial function, namely arginine and asymmetric dimethylarginine (ADMA) and complications of portal hypertension in liver cirrhosis. MATERIALS AND METHODS: Systemic omentin was measured by ELISA in portal venous serum (PVS), systemic venous serum (SVS) and hepatic venous serum (HVS) of 40 patients with liver cirrhosis and 10 liver-healthy controls. ADMA and arginine were determined in SVS of the patients by ELISA. RESULTS: Omentin is elevated in PVS and tends to be increased in SVS and HVS of patients with liver cirrhosis compared with controls. Omentin is principally expressed in visceral fat, and PVS omentin tends to be higher than SVS levels. Lower HVS than PVS omentin suggests that omentin may be partly removed from the circulation by the liver. Omentin in serum is not associated with stages of liver cirrhosis defined by CHILD-POUGH or MELD score and is not affected in patients with ascites. HVS omentin tends to be reduced in patients with large varices compared with patients without/with small varices. Arginine/ADMA ratio is reduced in patients with massive ascites but is not associated with variceal size. Further, Arginine/ADMA ratio does not correlate with omentin. CONCLUSION: Current data show that PVS omentin is increased in liver cirrhosis but is not associated with complications of portal hypertension.