Sex-stratified genome-wide association studies including 270,000 individuals show sexual dimorphism in genetic loci for anthropometric traits.

Given the anthropometric differences between men and women and previous evidence of sex-difference in genetic effects, we conducted a genome-wide search for sexually dimorphic associations with height, weight, body mass index, waist circumference, hip circumference, and waist-to-hip-ratio (133,723 individuals) and took forward 348 SNPs into follow-up (additional 137,052 individuals) in a total of 94 studies. Seven loci displayed significant sex-difference (FDR<5%), including four previously established (near GRB14/COBLL1, LYPLAL1/SLC30A10, VEGFA, ADAMTS9) and three novel anthropometric trait loci (near MAP3K1, HSD17B4, PPARG), all of which were genome-wide significant in women (P<5×10(-8)), but not in men. Sex-differences were apparent only for waist phenotypes, not for height, weight, BMI, or hip circumference. Moreover, we found no evidence for genetic effects with opposite directions in men versus women. The PPARG locus is of specific interest due to its role in diabetes genetics and therapy. Our results demonstrate the value of sex-specific GWAS to unravel the sexually dimorphic genetic underpinning of complex traits.

Lipidomic analysis of the liver identifies changes of major and minor lipid species in adiponectin deficient mice.

Adiponectin protects from hepatic fat storage but adiponectin deficient mice (APN-/-) fed a standard chow do not develop liver steatosis. This indicates that other pathways might be activated to compensate for adiponectin deficiency. An unbiased and comprehensive screen was performed to identify hepatic alterations of lipid classes in these mice. APN-/- mice had decreased hepatic cholesteryl esters while active SREBP2 and systemic total cholesterol were not altered. Upregulation of cytochromes for bile acid synthesis suggests enhanced biliary cholesterol excretion. Analysis of 37 individual fatty acid species showed reduced stearate whereas total fatty acids were not altered. Total amount of triglycerides and phospholipids were equally abundant. A selective increase of monounsaturated phosphatidylcholine and phosphatidylethanolamine which positively correlate with hepatic and systemic triglycerides with the latter being elevated in APN-/- mice, was identified. Stearoyl-CoA desaturase 1 (SCD1) is involved in the synthesis of monounsaturated fatty acids and despite higher mRNA expression enzyme activity was not enhanced. Glucosylceramide postulated to contribute to liver damage was decreased. This study demonstrates that adiponectin deficiency is associated with hepatic changes in lipid classes in mice fed a standard chow which may protect from liver steatosis.

Lipid accumulation impairs adiponectin-mediated induction of activin A by increasing TGFbeta in primary human hepatocytes.

Fatty liver is commonly detected in obesity and has been identified as a risk factor for the progression of hepatic fibrosis in a wide range of liver diseases. Transforming growth factor beta (TGFß) and activin A, both members of the TGFß superfamiliy, are central regulators in liver fibrosis and regeneration, and the effect of hepatocyte lipid accumulation on the release of these proteins was studied. Primary human hepatocytes (PHH) were incubated with palmitic acid or oleic acid to increase lipid storage. Whereas activin A and its natural inhibitor follistatin were not affected, TGFß was 2-fold increased. The hepatoprotective adipokine adiponectin dose-dependently induced activin A while lowering follistatin but did not alter TGFß. Activin A was markedly reduced in hepatocyte cell lines compared to PHH and was not induced upon adiponectin incubation demonstrating significant differences of primary and transformed cells. In free fatty acid (FFA)-incubated PHH adiponectin-mediated induction of activin A was impaired. Inhibition of TGFß receptors ALK4/5 and blockage of SMAD3 phosphorylation rescued activin A synthesis in FFA and in TGFß incubated cells suggesting that FFA inhibit adiponectin activity by inducing TGFß. To evaluate whether serum levels of activin A and its antagonist are altered in patients with hepatic steatosis, both proteins were measured in the serum of patients with sonographically diagnosed fatty liver and age- and BMI-matched controls. Systemic adiponectin was significantly reduced in patients with fatty liver but activin A and follistatin were not altered. In summary the current data demonstrate that lipid accumulation in hepatocytes induces TGFß which impairs adiponectin bioactivity, and thereby may contribute to liver injury.

Adipocyte chemerin release is induced by insulin without being translated to higher levels in vivo.

BACKGROUND: Chemerin is an adipokine that regulates insulin sensitivity and insulin secretion. Prolonged hyperinsulinaemia is associated with higher systemic chemerin, and insulin induces adipose tissue chemerin release. These findings led us to hypothesize that systemic chemerin may be associated with post-prandial glucose metabolism and/or may even be induced after oral glucose load. Therefore, the effect of insulin on adipocyte chemerin levels and systemic chemerin in mice was analysed. Further, systemic levels of chemerin after oral glucose load in nondiabetic individuals were studied. DESIGN AND METHODS: Chemerin levels were determined in adipocytes after short-term and long-term treatment with insulin. Effects of acute hyperinsulinaemia were studied in mice. Chemerin was measured during oral glucose tolerance test in 66 healthy, nondiabetic individuals stratified for established body mass index categories. RESULTS: Insulin induces chemerin release from adipocytes within 24 h, while cellular levels are not affected. Short-term hyperinsulinaemia also upregulates adipocyte chemerin in vitro but has no effect on adipose tissue and chemerin serum levels of mice. Systemic chemerin is higher in overweight/obese than normal-weight controls and positively correlates with total cholesterol. Chemerin is not associated with markers of insulin sensitivity like fasting glucose or insulin. Fasting chemerin levels are similar to concentrations measured 1 and 2 h after oral glucose uptake in overweight and obese donors. CONCLUSIONS: Post-prandial hyperinsulinaemia does not contribute to higher chemerin levels in nondiabetic individuals.

Chemerin induces CCL2 and TLR4 in synovial fibroblasts of patients with rheumatoid arthritis and osteoarthritis.

INTRODUCTION: Chemerin stimulates migration of leukocytes to sites of inflammation and also increases inflammatory signaling in chondrocytes suggesting a function of chemerin in joint inflammation. Synovial fibroblasts (SF) are critically involved in synovitis and subsequent cartilage destruction. Here, we analyzed whether synovial fibroblasts express chemerin and its receptor CMKLR1. Further, the role of chemerin in synovial fibroblast chemotaxis, proliferation, insulin response and release of inflammatory proteins was studied. METHODS: Synovial tissue sections were labeled with chemerin antibody and chemerin was measured in synovial fluid by ELISA. Chemerin mRNA and protein as well as CMKLR1 expression were determined in SFs from patients with osteoarthritis (OA) and rheumatoid arthritis (RA). Effects of chemerin on cytokines, chemokines and matrix metalloproteinases (MMP), and on proliferation, migration and insulin signaling were analyzed appropriately. RESULTS: SFs expressed CMKLR1 and chemerin mRNA, and chemerin protein was found in cell supernatants of synovial fibroblasts. Immunohistochemistry detected chemerin in synovial tissue predominantly localized within the lining layer. Chemerin was present in synovial fluids of RA, OA and psoriatic arthritis patients in similar concentrations. Chemerin neither increased IL-6 levels nor MMP-2 or -9 activity in SFs. Also, it did not act as a chemoattractant for these cells. With respect to intracellular signaling, neither basal nor insulin-mediated phosphorylation of Akt was affected. However, chemerin significantly increased TLR4 mRNA and synthesis of CCL2 in SFs while CCL4 and -5 were not altered. Cell proliferation of SFs, however, was modestly reduced by chemerin. CONCLUSIONS: These data show that human SFs express both chemerin and its receptor. As chemerin enhanced expression of TLR4 and induced release of CCL2 in SFs, a role of this protein in innate immune system-associated joint inflammation is proposed.

Adiponectin stimulates release of CCL2, -3, -4 and -5 while the surface abundance of CCR2 and -5 is simultaneously reduced in primary human monocytes.

The adipokine adiponectin is well known to affect the function of immune cells and upregulation of CCL2 by adiponectin in monocytes/macrophages has already been reported. In the current study the effect of adiponectin on CCL2, -3, -4, and -5 and their corresponding receptors CCR1, CCR2, and CCR5 has been analyzed. Adiponectin elevates mRNA and protein of the CC chemokines in primary human monocytes. Simultaneously the surface abundance of CCR2 and CCR5 is reduced while CCR1 is not affected. Downregulation of CCR2 by adiponectin is blocked by a CCR2 antagonist although expression of the CCL2 regulated genes CCR2 and TGF-beta 1 is not altered in the adiponectin-incubated monocytes. CCL2, -3, and -5 concentrations measured in supernatants of monocytes of normal-weight (NW), overweight (OW), and type 2 diabetic (T2D) patients positively correlate with BMI and are increased in obesity and T2D. In contrast CCL4 is similarly abundant in the supernatants of all of these monocytes. The degree of adiponectin-mediated induction of the chemokines CCL3, -4, and -5 negatively correlates with their basal levels and upregulation of CCL3 and CCL5 is significantly impaired in OW and T2D cells. Serum concentrations of these chemokines are almost equal in the three groups and do not correlate with the levels in monocyte supernatants. In conclusion these data demonstrate that adiponectin stimulates release of CCL2 to CCL5 in primary human monocytes, and induction in cells of overweight probands is partly impaired. Adiponectin also lowers surface abundance of CCR2 and CCR5 and downregulation of CCR2 seems to depend on autocrine/paracrine effects of CCL2.

Systemic and hepatic vein galectin-3 are increased in patients with alcoholic liver cirrhosis and negatively correlate with liver function.

Recently we demonstrated higher galectin-3 in portal venous serum (PVS) compared to hepatic venous serum (HVS) in a small cohort of patients with normal liver function suggesting hepatic removal of galectin-3. Here, galectin-3 was measured by ELISA in PVS, HVS and systemic venous blood (SVS) of 33 patients with alcoholic liver cirrhosis and a larger cohort of 11 patients with normal liver function. Galectin-3 was cleared by the healthy but not the cirrhotic liver, and subsequently HVS and SVS galectin-3 levels were significantly increased in the patients with liver cirrhosis compared to controls. In healthy liver galectin-3 was produced by cholangiocytes and synthesis by hepatocytes was only observed in cirrhotic liver. Hepatic venous pressure gradient did not correlate with galectin-3 levels excluding hepatic shunting as the principal cause of higher SVS galectin-3. Galectin-3 was elevated in all blood compartments of patients with CHILD-PUGH stage C compared to patients with CHILD-PUGH stage A, and was higher in patients with ascites than patients without this complication. Galectin-3 was negatively associated with antithrombin-3 whose synthesis is reduced with worse liver function. Galectin-3 positively correlated with urea and creatinine, and PVS galectin-3 showed a negative association with creatinine clearance as an accepted measure of kidney function. To summarize in the current study systemic, portal and hepatic levels of galectin-3 were found to be negatively associated with liver function in patients with alcoholic liver cirrhosis and this may in part be related to impaired hepatic removal and/or increased synthesis in cirrhotic liver.

Impaired hepatic removal of interleukin-6 in patients with liver cirrhosis.

Systemic concentrations of interleukin-6 (IL-6) are elevated in patients with liver cirrhosis, and impaired hepatic uptake of IL-6 was suggested to contribute to higher levels in these patients. To test this hypothesis IL-6 was measured in portal venous serum (PVS), hepatic venous serum (HVS) and systemic venous serum (SVS) of 41 patients with liver cirrhosis and four patients with normal liver function. IL-6 was higher in PVS than HVS of all blood donors and about 43% of portal vein derived IL-6 was extracted by the healthy liver, and 6.3% by the cirrhotic liver demonstrating markedly impaired removal of IL-6 by the latter. Whereas in patients with CHILD-PUGH stage A IL-6 in HVS was almost 25% lower than in PVS, in patients with CHILD-PUGH stage C IL-6 was similarly abundant in the two blood compartments. Ascites is a common complication in cirrhotic patients and was associated with higher IL-6 levels in all blood compartments without significant differences in hepatic excretion. Hepatic venous pressure gradient did not correlate with the degree of hepatic IL-6 removal excluding hepatic shunting as the principal cause of impaired IL-6 uptake. Furthermore, patients with alcoholic liver cirrhosis had higher IL-6 in all blood compartments than patients with cryptogenic liver cirrhosis. Aetiology of liver cirrhosis did not affect hepatic removal rate indicating higher IL-6 synthesis in patients with alcoholic liver cirrhosis. In summary, the current data provide evidence that impaired hepatic removal of IL-6 is explained by hepatic shunting and liver dysfunction in patients with liver cirrhosis partly explaining higher systemic levels.

Elevated free fatty acids and impaired adiponectin bioactivity contribute to reduced SOD2 protein in monocytes of type 2 diabetes patients.

Type 2 diabetes (T2D) is characterized by increased oxidative stress contributing to the development of cardiovascular disease (CVD). Monocytes are critically important in the pathogenesis of CVD and antioxidant enzymes like superoxide dismutase (SOD2) protect these cells from excessive reactive oxygen species (ROS). Adiponectin is an adipocyte-derived protein with atheroprotective function and the effect of adiponectin on monocyte SOD2 was analyzed herein. Adiponectin upregulated SOD2 mRNA and dose- and time-dependently induced SOD2 protein in primary human monocytes. Elevated systemic free fatty acids (FFA) are commonly found in T2D patients and palmitic acid as well as oleic acid reduced monocyte SOD2 protein. Adiponectin mediated upregulation of SOD2, however, was not affected by FFA incubation. SOD2 protein was reduced in T2D monocytes compared to monocytes of age- and body mass index-matched healthy controls. Adiponectin still induced SOD2 in T2D monocytes but efficiency tended to be reduced. In summary this study indicates that elevated systemic free fatty acids and impaired adiponectin activity contribute to reduced SOD2 and most likely increased oxidative stress in T2D monocytes.

MMP-9 activity is increased by adiponectin in primary human hepatocytes but even negatively correlates with serum adiponectin in a rodent model of non-alcoholic steatohepatitis.

Adiponectin protects from inflammation and fibrosis in metabolic liver disease. In the present study we analyzed whether this adipokine may directly affect the activity of matrix metalloproteinases (MMPs), central regulators of fibrinolysis, in hepatocytes. Global gene expression analysis indicated upregulation of MMP-9 and tissue inhibitor of metalloproteinases-1 (TIMP-1) expression in primary human hepatocytes (PHH) in response to stimulation with adiponectin, and these results were confirmed by real-time RT-PCR. Furthermore, gelatin zymography revealed that MMP-9 activity was significantly induced in supernatants of adiponectin stimulated PHHs. In a murine model of hepatic steatosis and in human steatotic liver samples hepatic MMP-9 activity was not significantly altered. However, in two different murine models of non-alcoholic steatohepatitis (NASH) MMP-9 activity was significantly elevated compared to chow fed control mice. Of note, MMP-9 activity did not or even negatively, respectively, correlate with adiponectin serum levels in these models. The current data indicate that in NASH hepatic inflammation and fibrosis but not hepatic steatosis induce liver MMP-9 activity, and this induction seems to be related to the anti-inflammatory activity of adiponectin rather than its effect on hepatocellular MMP-9 expression.

Soluble CD163 is not increased in visceral fat and steatotic liver and is even suppressed by free fatty acids in vitro.

Visceral fat differs from subcutaneous fat by higher local inflammation and increased release of IL-6 and free fatty acids (FFA) which contribute to hepatic steatosis. IL-6 has been shown to upregulate the monocyte/macrophage specific receptor CD163 whose soluble form, sCD163, is increased in inflammatory diseases. Here, it was analyzed whether CD163 and sCD163 are differentially expressed in the human fat depots and fatty liver. CD163 mRNA and protein were similarly expressed in paired samples of human visceral and subcutaneous fat, and comparable levels in portal venous and systemic venous blood of liver-healthy controls indicate that release of sCD163 from visceral adipose tissue was not increased. CD163 was also similarly expressed in steatotic liver when compared to non-steatotic tissues and sCD163 was almost equal in the respective sera. Concentrations of sCD163 were not affected when passing the liver excluding substantial hepatic removal/release of this protein. A high concentration of IL-6 upregulated CD163 protein while physiological doses had no effect. However, sCD163 was not increased by any of the IL-6 doses tested. FFA even modestly decreased CD163 and sCD163. The anti-inflammatory mediators fenofibrate, pioglitazone, and eicosapentaenoic acid (EPA) did not influence sCD163 levels while CD163 was reduced by EPA. These data suggest that in humans neither visceral fat nor fatty liver are major sources of sCD163.

Adiponectin reduces connective tissue growth factor in human hepatocytes which is already induced in non-fibrotic non-alcoholic steatohepatitis.

Connective tissue growth factor (CTGF) is induced in liver fibrosis and enhances the activity of transforming growth factor ß (TGFß). Recently we have shown that the hepatoprotective adipokine adiponectin downregulates CTGF in primary human hepatocytes (PHH). In the current study, the mechanisms mediating suppression of CTGF by adiponectin and the well described downstream effector of adiponectin receptor 2 (AdipoR2), peroxisome proliferator activated receptor α (PPARα), were analyzed in more detail. Adiponectin downregulated CTGF mRNA and protein in primary human hepatocytes (PHH) and suppression was blocked by a PPARα antagonist indicating that AdipoR2 is involved. The PPARα agonists fenofibrate and WY14643 also reduced CTGF protein in these cells. Adiponectin further impaired TGFß-mediated upregulation of CTGF. Phosphorylation of the TGFß downstream effectors SMAD2 and -3 was reduced in PHH incubated with adiponectin or PPARα agonists suggesting that early steps in TGFß signal transduction are impaired. CTGF and TGFß mRNA levels were increased in human non-fibrotic non-alcoholic steatohepatitis (NASH), and here AdipoR2 expression was significantly reduced. Current data show that CTGF and TGFß are already induced in non-fibrotic NASH and this may be partly explained by low adiponectin bioactivity which interferes with TGFß signaling by reducing phosphorylation of SMAD2/3 and by downregulating CTGF.

Systemic chemerin is related to inflammation rather than obesity in type 2 diabetes.

BACKGROUND: The adipokine chemerin modulates the function of innate immune cells and may link obesity and inflammation, and therefore, a possible relation of chemerin to inflammatory proteins in obesity and type 2 diabetes (T2D) was analysed. As visceral fat contributes to systemic inflammation, chemerin was measured in portal venous (PVS), hepatic venous (HVS) and systemic venous (SVS) blood of patients with liver cirrhosis. PATIENTS AND METHODS: Systemic chemerin was determined by ELISA in the serum of normal-weight, overweight and T2D males, in the serum of T2D patients of both sexes, and in PVS, HVS and SVS of patients with liver cirrhosis. RESULTS: Circulating chemerin was similar in T2D and obese individuals but was significantly elevated in both cohorts compared to normal-weight individuals. Chemerin positively correlated with leptin, resistin and C-reactive protein (CRP). In T2D, chemerin was similar in male and female patients and increased in patients with elevated CRP. Chemerin was similar in PVS and SVS, indicating that visceral fat is not a major site of chemerin synthesis. Higher levels of chemerin in HVS demonstrate that chemerin is also released by the liver. CONCLUSIONS: Visceral fat is not a major site of chemerin release, and elevated systemic levels of chemerin in obesity and T2D seem to be associated with inflammation rather than body mass index.

Adiponectin-stimulated CXCL8 release in primary human hepatocytes is regulated by ERK1/ERK2, p38 MAPK, NF-kappaB, and STAT3 signaling pathways.

Adiponectin is believed to exert hepatoprotective effects and induces CXCL8, a chemokine that functions as a survival factor, in vascular cells. In the current study, it is demonstrated that adiponectin also induces CXCL8 expression in primary human hepatocytes but not in hepatocellular carcinoma cell lines. Knock down of the adiponectin receptor (AdipoR) 1 or AdipoR2 by small-interfering RNA indicates that AdipoR1 is involved in adiponectin-stimulated CXCL8 release. Adiponectin activates nuclear factor (NF)-kappaB in primary hepatocytes and pharmacological inhibition of NF-kappaB, the p38 mitogen-activated protein kinase, and extracellular signal-regulated kinase (ERK) 1/ERK2 reduces adiponectin-mediated CXCL8 secretion. Furthermore, adiponectin also activates STAT3 involved in interleukin (IL)-6 and leptin-mediated CXCL8 induction in primary hepatocytes. Inhibition of JAK2 by AG-490 does not abolish adiponectin-stimulated CXCL8, indicating that this kinase is not involved. Pretreatment of primary cells with "STAT3 Inhibitor VI," however, elevates hepatocytic CXCL8 secretion, demonstrating that STAT3 is a negative regulator of CXCL8 in these cells. In accordance with this assumption, IL-6, a well-characterized activator of STAT3, reduces hepatocytic CXCL8. Therefore, adiponectin-stimulated induction of CXCL8 seems to be tightly controlled in primary human hepatocytes, whereas neither NF-kappaB, STAT3, nor CXCL8 are influenced in hepatocytic cell lines. CXCL8 is a survival factor, and its upregulation by adiponectin may contribute to the hepatoprotective effects of this adipokine.

Small-interference RNA-mediated knock-down of aldehyde oxidase 1 in 3T3-L1 cells impairs adipogenesis and adiponectin release.

Aldehyde oxidase 1 (AOX1) is highly abundant in the liver and oxidizes aldehydes thereby generating reactive oxygen species. Enzymes involved in detoxification of aldehydes are expressed in adipocytes and alter adipogenesis, therefore the functional role of AOX1 in adipocytes was analyzed. AOX1 mRNA was higher in visceral compared to subcutaneous human adipose tissue but AOX1 protein was detected in both fat depots. AOX1 expression in adipocytes was confirmed by immunohistochemistry and immunoblot. AOX1 was induced during adipocytic differentiation and was downregulated by fenofibrate in differentiated cells. Knock-down of AOX1 in preadipocytes led to impaired lipid storage and adiponectin release in the differentiated cells. These data indicate that AOX1 is essential for adipogenesis and may link energy and drug metabolism.

Blunted HPA axis reactivity reveals glucocorticoid system dysbalance in a mouse model of high anxiety-related behavior.

Depression and anxiety disorders are often characterized by altered hypothalamic-pituitary-adrenal (HPA) axis re-/activity. However, the presence of a molecular link between dysbalanced neuroendocrine regulation and psychopathologies is not yet fully established. Earlier, we reported that high (HAB), normal (NAB) and low (LAB) anxiety-related behavior mice express divergent anxiety-related and passive/active coping phenotypes. Here, we studied mechanisms that might contribute to the different HPA axis reactivity observed in HAB, NAB and LAB mice and their involvement in the regulation of anxiety-related behavior and passive/active coping style. We found that HAB mice respond with significantly reduced corticosterone (CORT) secretion to an acute stressful stimulus and a blunted response in the Dex/CRH test compared to NAB and LAB mice. At the molecular level, higher expression of the glucocorticoid receptor (GR/Nr3c1) and decreased corticotropin-releasing hormone receptor 1 (CRHR1) expression were observed in the pituitary of HAB mice. We further analyzed whether these stress mediators differed between the HAB, NAB and LAB lines in limbic system-associated brain regions and whether their interplay contributes to the phenotype. Interestingly, not only in the pituitary but also in almost all brain regions investigated, GR expression was significantly higher in HAB mice. In contrast, the amount of CORT in the brain structures analyzed was significantly lower in these animals. The expression of CRHR1 varied in the prefrontal cortex only. Since glucocorticoids regulate both GR and CRHR1, we treated HAB and NAB mice chronically with CORT. After 6 weeks of administration, reduced anxiety- and depression-like behaviors were observed in HAB mice, whereas increased anxiety was found in NABs. In both groups, GR, but not CRHR1, were significantly reduced. Taken together, our study proposes HAB mice as an animal model of simultaneous features of increased anxiety-related and depression-like behaviors with blunted HPA axis reactivity suggesting a dysregulated GR/CORT system as one key mechanism behind their phenotype.

ATP binding cassette transporter A1 (ABCA1) associated proteins: potential drug targets in the metabolic syndrome and atherosclerotic disease?

The metabolic syndrome is defined as a cluster of disorders including visceral obesity, insulin resistance, hypertension, hypertriglyceridemia and low HDL. Patients have an increased risk to develop atherosclerotic disease characterized by excessive macrophage cholesterol deposition in the vascular wall. HDL removes excess cellular cholesterol which is subsequently transported to the liver for biliary excretion and thereby preserves cholesterol homeostasis. Circulating HDL levels are maintained by hepatic ATP-binding cassette transporter A1 (ABCA1) which also transports peripheral cell cholesterol to extracellular lipid acceptors. Lipid export activity of ABCA1 improves pancreatic -cell function and ameliorates insulin release. ABCA1 affects plasma membrane cholesterol distribution and formation of lipid rafts representing signalling platforms for diverse receptors including toll-like receptor 4 (TLR4). Pharmacological activation of ABCA1 pathways presumably progresses metabolic diseases, and current approaches demonstrate beneficial effects of small peptides mimicking ABCA1 ligands which stabilize ABCA1 and enhance lipid efflux similar to its physiological acceptor apolipoprotein A-I. Research of the last decade has resulted in the identification of several ABCA1 binding proteins influencing ABCA1 signalling, stability and activity. In the current review the proteins suggested to form a complex with ABCA1 are summarized and their up to now characterized features towards ABCA1 functions are described.

Adiponectin upregulates hepatocyte CMKLR1 which is reduced in human fatty liver.

Chemokine-like receptor 1 (CMKLR1) ligands chemerin and resolvin E1 are suggested to have a role in non-alcoholic fatty liver disease (NAFLD). Here, expression of CMKLR1 in liver cells and NAFLD was studied. CMKLR1 was detected in primary human hepatocytes (PHH), Kupffer cells, bile-duct cells and hepatic stellate cells. In human and rodent fatty liver and in fibrotic liver of mice fed a methionine-choline deficient diet CMKLR1 was reduced. Hepatocytes are the major cells in the liver and effects of adipokines, cytokines and lipids on CMKLR1 in PHH were analyzed. Increased cellular triglyceride or cholesterol content, lipopolysaccharide, IL-6, TNF and leptin did not influence CMKLR1 levels in PHH whereas profibrotic TGFß tended to reduce CMKLR1. Adiponectin strongly upregulated CMKLR1 mRNA and protein in PHH and hepatic CMKLR1 when injected into wild type mice. Further, CMKLR1 was suppressed in the liver of adiponectin deficient mice. These data indicate that low CMKLR1 in NAFLD may partly result from reduced adiponectin activity.

Connective tissue growth factor level is increased in patients with liver cirrhosis but is not associated with complications or extent of liver injury.

Connective tissue growth factor (CTGF) is a profibrotic protein whose systemic levels are increased in liver cirrhosis. Here, association of CTGF with stages of liver injury and complications of cirrhotic liver disease has been analyzed in patients with different aetiologies of hepatic injury. CTGF is significantly increased in portal venous serum (PVS), hepatic venous serum (HVS) and systemic venous serum (SVS) of 46 patients with liver cirrhosis compared to eight liver-healthy controls. In patients´ blood samples CTGF in HVS is about 6% higher than PVS levels indicating that CTGF produced in the liver is released to the circulation. CTGF is not associated with stages of liver cirrhosis defined by CHILD-PUGH or MELD score nor with secondary complications of portal hypertension (varices, ascites, spontaneous bacterial peritonitis). Transforming growth factor ß (TGFß) induces CTGF synthesis in hepatocytes and a positive association of systemic TGFß1 and SVS and HVS CTGF is found. Three months after placing transjugular intrahepatic portosystemic shunt (TIPS) hepatic venous pressure gradient is reduced whereas CHILD-PUGH score, TGFß1 and CTGF are not altered in serum of 15 patients. Current data show that the cirrhotic liver releases little CTGF but SVS, HVS and PVS CTGF levels are not associated with residual liver function and complications of cirrhosis.

Systemic resistin is increased in type 2 diabetic patients treated with loop diuretics.

Increased serum resistin was found in rodent models of obesity and insulin resistance, whereas contradictory results have been obtained in human studies. In humans, resistin is primarily released by monocytes/macrophages, suggesting that soluble levels may be associated with macrophage activation. Here, systemic and monocyte-released resistin levels were found to be similar in type 2 diabetic (T2D) patients, overweight controls and normal-weight controls. When adjusted for body mass index and age, serum resistin modestly correlated with gamma-glutamyltransferase levels, fasting glucose and interleukin-6. Systemic resistin was marginally increased in T2D patients treated with beta-blockers or urate-lowering drugs and was considerably higher in patients treated with loop diuretics. Monocyte-released resistin was even reduced by the loop diuretic furosemide, excluding the possibility that this drug may directly stimulate resistin synthesis. In summary, the current data indicate that changes accompanying renal dysfunction but not obesity or type 2 diabetes are associated with increased serum resistin.