SK
Sander Kersten
Author with expertise in Lipid Metabolism and Cardiovascular Disease
Achievements
Cited Author
Open Access Advocate
Key Stats
Upvotes received:
0
Publications:
24
(83% Open Access)
Cited by:
8,536
h-index:
89
/
i10-index:
205
Reputation
Biology
< 1%
Chemistry
< 1%
Economics
< 1%
Show more
How is this calculated?
Publications
0

Peroxisome proliferator–activated receptor α mediates the adaptive response to fasting

Sander Kersten et al.Jun 1, 1999
Prolonged deprivation of food induces dramatic changes in mammalian metabolism, including the release of large amounts of fatty acids from the adipose tissue, followed by their oxidation in the liver. The nuclear receptor known as peroxisome proliferator–activated receptor α (PPARα) was found to play a role in regulating mitochondrial and peroxisomal fatty acid oxidation, suggesting that PPARα may be involved in the transcriptional response to fasting. To investigate this possibility, PPARα-null mice were subjected to a high fat diet or to fasting, and their responses were compared with those of wild-type mice. PPARα-null mice chronically fed a high fat diet showed a massive accumulation of lipid in their livers. A similar phenotype was noted in PPARα-null mice fasted for 24 hours, who also displayed severe hypoglycemia, hypoketonemia, hypothermia, and elevated plasma free fatty acid levels, indicating a dramatic inhibition of fatty acid uptake and oxidation. It is shown that to accommodate the increased requirement for hepatic fatty acid oxidation, PPARα mRNA is induced during fasting in wild-type mice. The data indicate that PPARα plays a pivotal role in the management of energy stores during fasting. By modulating gene expression, PPARα stimulates hepatic fatty acid oxidation to supply substrates that can be metabolized by other tissues.
0

Calorie Restriction-like Effects of 30 Days of Resveratrol Supplementation on Energy Metabolism and Metabolic Profile in Obese Humans

Silvie Timmers et al.Nov 1, 2011

Summary

 Resveratrol is a natural compound that affects energy metabolism and mitochondrial function and serves as a calorie restriction mimetic, at least in animal models of obesity. Here, we treated 11 healthy, obese men with placebo and 150 mg/day resveratrol (resVida) in a randomized double-blind crossover study for 30 days. Resveratrol significantly reduced sleeping and resting metabolic rate. In muscle, resveratrol activated AMPK, increased SIRT1 and PGC-1α protein levels, increased citrate synthase activity without change in mitochondrial content, and improved muscle mitochondrial respiration on a fatty acid-derived substrate. Furthermore, resveratrol elevated intramyocellular lipid levels and decreased intrahepatic lipid content, circulating glucose, triglycerides, alanine-aminotransferase, and inflammation markers. Systolic blood pressure dropped and HOMA index improved after resveratrol. In the postprandial state, adipose tissue lipolysis and plasma fatty acid and glycerol decreased. In conclusion, we demonstrate that 30 days of resveratrol supplementation induces metabolic changes in obese humans, mimicking the effects of calorie restriction.
0

The Inflammasome-Mediated Caspase-1 Activation Controls Adipocyte Differentiation and Insulin Sensitivity

Rinke Stienstra et al.Dec 1, 2010

Summary

 Obesity-induced inflammation originating from expanding adipose tissue interferes with insulin sensitivity. Important metabolic effects have been recently attributed to IL-1β and IL-18, two members of the IL-1 family of cytokines. Processing of IL-1β and IL-18 requires cleavage by caspase-1, a cysteine protease regulated by a protein complex called the inflammasome. We demonstrate that the inflammasome/caspase-1 governs adipocyte differentiation and insulin sensitivity. Caspase-1 is upregulated during adipocyte differentiation and directs adipocytes toward a more insulin-resistant phenotype. Treatment of differentiating adipocytes with recombinant IL-1β and IL-18, or blocking their effects by inhibitors, reveals that the effects of caspase-1 on adipocyte differentiation are largely conveyed by IL-1β. Caspase-1 and IL-1β activity in adipose tissue is increased both in diet-induced and genetically induced obese animal models. Conversely, mice deficient in caspase-1 are more insulin sensitive as compared to wild-type animals. In addition, differentiation of preadipocytes isolated from caspase-1−/− or NLRP3−/− mice resulted in more metabolically active fat cells. In vivo, treatment of obese mice with a caspase-1 inhibitor significantly increases their insulin sensitivity. Indirect calorimetry analysis revealed higher fat oxidation rates in caspase-1−/− animals. In conclusion, the inflammasome is an important regulator of adipocyte function and insulin sensitivity, and caspase-1 inhibition may represent a novel therapeutic target in clinical conditions associated with obesity and insulin resistance.
0

Characterization of the Fasting-induced Adipose Factor FIAF, a Novel Peroxisome Proliferator-activated Receptor Target Gene

Sander Kersten et al.Sep 1, 2000
Fasting is associated with significant changes in nutrient metabolism, many of which are governed by transcription factors that regulate the expression of rate-limiting enzymes. One factor that plays an important role in the metabolic response to fasting is the peroxisome proliferator-activated receptor α (PPARα). To gain more insight into the role of PPARα during fasting, and into the regulation of metabolism during fasting in general, a search for unknown PPARα target genes was performed. Using subtractive hybridization (SABRE) comparing liver mRNA from wild-type and PPARα null mice, we isolated a novel PPARα target gene, encoding the secreted protein FIAF (for fasting induced adipose factor), that belongs to the family of fibrinogen/angiopoietin-like proteins. FIAF is predominantly expressed in adipose tissue and is strongly up-regulated by fasting in white adipose tissue and liver. Moreover, FIAF mRNA is decreased in white adipose tissue of PPARγ +/− mice. FIAF protein can be detected in various tissues and in blood plasma, suggesting that FIAF has an endocrine function. Its plasma abundance is increased by fasting and decreased by chronic high fat feeding. The data suggest that FIAF represents a novel endocrine signal involved in the regulation of metabolism, especially under fasting conditions.
0

Triglyceride-rich lipoproteins and their remnants: metabolic insights, role in atherosclerotic cardiovascular disease, and emerging therapeutic strategies—a consensus statement from the European Atherosclerosis Society

Henry Ginsberg et al.Jul 30, 2021
Recent advances in human genetics, together with a large body of epidemiologic, preclinical, and clinical trial results, provide strong support for a causal association between triglycerides (TG), TG-rich lipoproteins (TRL), and TRL remnants, and increased risk of myocardial infarction, ischaemic stroke, and aortic valve stenosis. These data also indicate that TRL and their remnants may contribute significantly to residual cardiovascular risk in patients on optimized low-density lipoprotein (LDL)-lowering therapy. This statement critically appraises current understanding of the structure, function, and metabolism of TRL, and their pathophysiological role in atherosclerotic cardiovascular disease (ASCVD). Key points are (i) a working definition of normo- and hypertriglyceridaemic states and their relation to risk of ASCVD, (ii) a conceptual framework for the generation of remnants due to dysregulation of TRL production, lipolysis, and remodelling, as well as clearance of remnant lipoproteins from the circulation, (iii) the pleiotropic proatherogenic actions of TRL and remnants at the arterial wall, (iv) challenges in defining, quantitating, and assessing the atherogenic properties of remnant particles, and (v) exploration of the relative atherogenicity of TRL and remnants compared to LDL. Assessment of these issues provides a foundation for evaluating approaches to effectively reduce levels of TRL and remnants by targeting either production, lipolysis, or hepatic clearance, or a combination of these mechanisms. This consensus statement updates current understanding in an integrated manner, thereby providing a platform for new therapeutic paradigms targeting TRL and their remnants, with the aim of reducing the risk of ASCVD.
0

The Fasting-induced Adipose Factor/Angiopoietin-like Protein 4 Is Physically Associated with Lipoproteins and Governs Plasma Lipid Levels and Adiposity

Stéphane Mandard et al.Nov 6, 2005
Proteins secreted from adipose tissue are increasingly recognized to play an important role in the regulation of glucose metabolism. However, much less is known about their effect on lipid metabolism. The fasting-induced adipose factor (FIAF/angiopoietin-like protein 4/peroxisome proliferator-activated receptor γ angiopoietin-related protein) was previously identified as a target of hypolipidemic fibrate drugs and insulin-sensitizing thiazolidinediones. Using transgenic mice that mildly overexpress FIAF in peripheral tissues we show that FIAF is an extremely powerful regulator of lipid metabolism and adiposity. FIAF overexpression caused a 50% reduction in adipose tissue weight, partly by stimulating fatty acid oxidation and uncoupling in fat. In addition, FIAF overexpression increased plasma levels of triglycerides, free fatty acids, glycerol, total cholesterol, and high density lipoprotein (HDL)-cholesterol. Functional tests indicated that FIAF overexpression severely impaired plasma triglyceride clearance but had no effect on very low density lipoprotein production. The effects of FIAF overexpression were amplified by a high fat diet, resulting in markedly elevated plasma and liver triglycerides, plasma free fatty acids, and plasma glycerol levels, and impaired glucose tolerance in FIAF transgenic mice fed a high fat diet. Remarkably, in mice the full-length form of FIAF was physically associated with HDL, whereas truncated FIAF was associated with low density lipoprotein. In human both full-length and truncated FIAF were associated with HDL. The composite data suggest that via physical association with plasma lipoproteins, FIAF acts as a powerful signal from fat and other tissues to prevent fat storage and stimulate fat mobilization. Our data indicate that disturbances in FIAF signaling might be involved in dyslipidemia. Proteins secreted from adipose tissue are increasingly recognized to play an important role in the regulation of glucose metabolism. However, much less is known about their effect on lipid metabolism. The fasting-induced adipose factor (FIAF/angiopoietin-like protein 4/peroxisome proliferator-activated receptor γ angiopoietin-related protein) was previously identified as a target of hypolipidemic fibrate drugs and insulin-sensitizing thiazolidinediones. Using transgenic mice that mildly overexpress FIAF in peripheral tissues we show that FIAF is an extremely powerful regulator of lipid metabolism and adiposity. FIAF overexpression caused a 50% reduction in adipose tissue weight, partly by stimulating fatty acid oxidation and uncoupling in fat. In addition, FIAF overexpression increased plasma levels of triglycerides, free fatty acids, glycerol, total cholesterol, and high density lipoprotein (HDL)-cholesterol. Functional tests indicated that FIAF overexpression severely impaired plasma triglyceride clearance but had no effect on very low density lipoprotein production. The effects of FIAF overexpression were amplified by a high fat diet, resulting in markedly elevated plasma and liver triglycerides, plasma free fatty acids, and plasma glycerol levels, and impaired glucose tolerance in FIAF transgenic mice fed a high fat diet. Remarkably, in mice the full-length form of FIAF was physically associated with HDL, whereas truncated FIAF was associated with low density lipoprotein. In human both full-length and truncated FIAF were associated with HDL. The composite data suggest that via physical association with plasma lipoproteins, FIAF acts as a powerful signal from fat and other tissues to prevent fat storage and stimulate fat mobilization. Our data indicate that disturbances in FIAF signaling might be involved in dyslipidemia. The fasting-induced adipose factor/angiopoietin-like protein 4 is physically associated with lipoproteins and governs plasma lipid levels and adiposity. VOLUME 281 (2006) PAGES 934-944Journal of Biological ChemistryVol. 281Issue 30PreviewPAGE 937: Full-Text PDF Open Access Obesity and associated diabetes mellitus type 2 have become major health concerns throughout the world. These ailments are intimately linked as excess weight greatly increases the likelihood of developing diabetes. Obesity also increases the risk for other clinical abnormalities such as hypertension and dyslipidemia. Although the positive association between obesity and several co-morbidities has been well established at the epidemiological level, the mechanisms behind these associations are much less clear. Much attention has been given to the role of plasma free fatty acids (FFAs), 3The abbreviations used are: FFAfree fatty acidFIAFfasting-induced adipose factorPPARperoxisome proliferator-activated receptorTGtriglycerideFIAF-TgFIAF transgenicVLDLvery low density lipoproteinHDLhigh density lipoproteinFPLCfast protein liquid chromatographyLPLlipoprotein lipaseWATwhite adipose tissueBATbrown adipose tissueHSLhormone-sensitive lipaseQ-PCRquantitative real time PCRHFDhigh fat dietLFDlow fat dietASPacylation-stimulating protein. which are elevated in the obese and able to disrupt cellular metabolism in several organs. However, it also has become evident that a variety of hormonal factors produced by adipose tissue can greatly affect organ functioning, especially at the metabolic and immunological levels (1Ahima R.S. Flier J.S. Trends Endocrinol. Metab. 2000; 11: 327-332Abstract Full Text Full Text PDF PubMed Scopus (1206) Google Scholar). Indeed, these so-called adipocytokines or adipokines are now known to affect diverse biological processes, ranging from energy intake, insulin sensitivity, and hepatic glucose production, to reproductive and immunological function (2Zhang Y. Proenca R. Maffei M. Barone M. Leopold L. Friedman J.M. Nature. 1994; 372: 425-432Crossref PubMed Scopus (11807) Google Scholar, 3Berg A.H. Combs T.P. Du X. Brownlee M. Scherer P.E. Nat. Med. 2001; 7: 947-953Crossref PubMed Scopus (2218) Google Scholar, 4Steppan C.M. Bailey S.T. Bhat S. Brown E.J. Banerjee R.R. Wright C.M. Patel H.R. Ahima R.S. Lazar M.A. Nature. 2001; 409: 307-312Crossref PubMed Scopus (4004) Google Scholar, 5Hotamisligil G.S. Shargill N.S. Spiegelman B.M. Science. 1993; 259: 87-91Crossref PubMed Scopus (6193) Google Scholar). Whereas the effects of adipocytokines on glucose homeostasis are becoming increasingly transparent, much less is known about how they regulate plasma lipid metabolism. free fatty acid fasting-induced adipose factor peroxisome proliferator-activated receptor triglyceride FIAF transgenic very low density lipoprotein high density lipoprotein fast protein liquid chromatography lipoprotein lipase white adipose tissue brown adipose tissue hormone-sensitive lipase quantitative real time PCR high fat diet low fat diet acylation-stimulating protein. A relatively poorly characterized adipocytokine that may be involved in regulation of plasma lipid metabolism is the fasting-induced adipose factor (FIAF), also known as PPARγ angiopoietin-related protein, angiopoietin-like protein 4, or hepatic fibrinogen/angiopoietin-related protein (6Kersten S. Mandard S. Tan N.S. Escher P. Metzger D. Chambon P. Gonzalez F.J. Desvergne B. Wahli W. J. Biol. Chem. 2000; 275: 28488-28493Abstract Full Text Full Text PDF PubMed Scopus (467) Google Scholar, 8Yoon J.C. Chickering T.W. Rosen E.D. Dussault B. Qin Y. Soukas A. Friedman J.M. Holmes W.E. Spiegelman B.M. Mol. Cell. Biol. 2000; 20: 5343-5349Crossref PubMed Scopus (339) Google Scholar, 9Kim I. Kim H.G. Kim H. Kim H.H. Park S.K. Uhm C.S. Lee Z.H. Koh G.Y. Biochem. J. 2000; 346: 603-610Crossref PubMed Scopus (232) Google Scholar, 10Yoshida K. Shimizugawa T. Ono M. Furukawa H. J. Lipid Res. 2002; 43: 1770-1772Abstract Full Text Full Text PDF PubMed Scopus (309) Google Scholar). FIAF is a glycoprotein of ∼50 kDa that belongs to the family of fibrinogen/angiopoietin-like proteins. In mice FIAF is most highly expressed in white and brown adipose tissue, and to a much lesser extent in other tissues such as heart, skeletal muscle, and liver (6Kersten S. Mandard S. Tan N.S. Escher P. Metzger D. Chambon P. Gonzalez F.J. Desvergne B. Wahli W. J. Biol. Chem. 2000; 275: 28488-28493Abstract Full Text Full Text PDF PubMed Scopus (467) Google Scholar, 8Yoon J.C. Chickering T.W. Rosen E.D. Dussault B. Qin Y. Soukas A. Friedman J.M. Holmes W.E. Spiegelman B.M. Mol. Cell. Biol. 2000; 20: 5343-5349Crossref PubMed Scopus (339) Google Scholar). The most compelling data to date link FIAF with regulation of lipid metabolism. FIAF was first identified as a target gene of the nuclear receptors PPARα and PPARγ, which govern lipid metabolism in liver and white adipose tissue, respectively (6Kersten S. Mandard S. Tan N.S. Escher P. Metzger D. Chambon P. Gonzalez F.J. Desvergne B. Wahli W. J. Biol. Chem. 2000; 275: 28488-28493Abstract Full Text Full Text PDF PubMed Scopus (467) Google Scholar, 8Yoon J.C. Chickering T.W. Rosen E.D. Dussault B. Qin Y. Soukas A. Friedman J.M. Holmes W.E. Spiegelman B.M. Mol. Cell. Biol. 2000; 20: 5343-5349Crossref PubMed Scopus (339) Google Scholar). Subsequent studies employing adenoviral-mediated overexpression of FIAF or injection of recombinant FIAF have shown that FIAF potently elevates plasma triglyceride (TG) levels (10Yoshida K. Shimizugawa T. Ono M. Furukawa H. J. Lipid Res. 2002; 43: 1770-1772Abstract Full Text Full Text PDF PubMed Scopus (309) Google Scholar, 11Ge H. Yang G. Yu X. Pourbahrami T. Li C. J. Lipid Res. 2004; 45: 2071-2079Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar, 12Xu A. Lam M.C. Chan K.W. Wang Y. Zhang J. Hoo R.L. Xu J.Y. Chen B. Chow W.S. Tso A.W. Lam K.S. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 6086-6091Crossref PubMed Scopus (255) Google Scholar). This is possibly achieved by inhibition of lipoprotein lipase, the enzyme that is rate-limiting for plasma TG hydrolysis (10Yoshida K. Shimizugawa T. Ono M. Furukawa H. J. Lipid Res. 2002; 43: 1770-1772Abstract Full Text Full Text PDF PubMed Scopus (309) Google Scholar, 11Ge H. Yang G. Yu X. Pourbahrami T. Li C. J. Lipid Res. 2004; 45: 2071-2079Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar, 13Yu X. Burgess S.C. Ge H. Wong K.K. Nassem R.H. Garry D.J. Sherry A.D. Malloy C.R. Berger J.P. Li C. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 1767-1772Crossref PubMed Scopus (87) Google Scholar). According to a recent report, FIAF may also potently lower plasma glucose levels. In wild-type C57/B6 as well as in db/db mice, adenoviral-mediated overexpression of FIAF was found to improve hyperglycemia, hyperinsulinemia, and glucose tolerance (12Xu A. Lam M.C. Chan K.W. Wang Y. Zhang J. Hoo R.L. Xu J.Y. Chen B. Chow W.S. Tso A.W. Lam K.S. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 6086-6091Crossref PubMed Scopus (255) Google Scholar). In addition to lipid metabolism, FIAF has also been associated with angiogenesis. Expression of FIAF is up-regulated during hypoxia in both endothelial cells and cardiomyocytes, which probably occurs via the hypoxia-inducible factor 1α (HIF-1α) (14Belanger A.J. Lu H. Date T. Liu L.X. Vincent K.A. Akita G.Y. Cheng S.H. Gregory R.J. Jiang C. J. Mol. Cell. Cardiol. 2002; 34: 765-774Abstract Full Text PDF PubMed Scopus (117) Google Scholar, 15Le Jan S. Amy C. Cazes A. Monnot C. Lamande N. Favier J. Philippe J. Sibony M. Gasc J.M. Corvol P. Germain S. Am. J. Pathol. 2003; 162: 1521-1528Abstract Full Text Full Text PDF PubMed Scopus (271) Google Scholar). Furthermore, FIAF is expressed in certain tumors, especially in the hypoxic area surrounding necrotic cells, and has been suggested as a marker for conventional renal cell carcinoma. In the chicken chorioallantoic membrane assay, FIAF is able to induce a strong pro-angiogenic response (15Le Jan S. Amy C. Cazes A. Monnot C. Lamande N. Favier J. Philippe J. Sibony M. Gasc J.M. Corvol P. Germain S. Am. J. Pathol. 2003; 162: 1521-1528Abstract Full Text Full Text PDF PubMed Scopus (271) Google Scholar). In contrast, others have ascribed a potent anti-angiogenic function to FIAF (16Ito Y. Oike Y. Yasunaga K. Hamada K. Miyata K. Matsumoto S. Sugano S. Tanihara H. Masuho Y. Suda T. Cancer Res. 2003; 63: 6651-6657PubMed Google Scholar). Thus, the role of FIAF in angiogenesis remains ambiguous. To determine the physiological role of FIAF, we studied the effect of FIAF overexpression in a transgenic mouse model. Our data indicate that FIAF, which is physically associated with plasma lipoproteins, is an important determinant of plasma TG concentration and clearance at physiological levels of expression. In addition, it stimulates adipose tissue lipolysis. In contrast to a recent study (12Xu A. Lam M.C. Chan K.W. Wang Y. Zhang J. Hoo R.L. Xu J.Y. Chen B. Chow W.S. Tso A.W. Lam K.S. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 6086-6091Crossref PubMed Scopus (255) Google Scholar), in our model FIAF overexpression was associated with deterioration of glucose tolerance. Finally, we observed that the effects of FIAF overexpression were clearly amplified by feeding a high fat diet. Generation of FIAF Transgenic (FIAF-Tg) Mice—The complete murine FIAF gene was amplified by PCR (High fidelity Taq polymerase, Roche Applied Science) from genomic DNA of mouse ES cells (Sv129) using primers CGGGCTCCAGATCTTCTTCTGCACCAG and GTCAGAGGCGGCATTGGACCCCCTTGAAGTA and subcloned into the pGEM-Teasy vector (Promega, Leiden, the Netherlands). The proper sequence of the exons was verified by DNA sequencing. The mFIAF gene was subsequently cut out with NotI and placed behind the murine aP2 (adipocyte fatty acid-binding protein) promoter within pBS-SKII+ (kind gift of Dr. Bruce Spiegelman). The promoter plus the mFIAF gene were excised with ClaI and SacII and gel purified before being injected into fertilized oocytes (strain FVB-NIco, Eurogentec Transgenic Production Service, Seraing, Belgium). Genotyping was performed on genomic DNA from mouse ears by Q-PCR (Sybr Green) using primers GCCCCCATTGGTCACTCCTACAG and CGGCTCAGACTTAGACTTGCTC, which anneal within the aP2 promoter and the mFIAF gene, respectively, and control primers GCTGCTGGAGAATGAGTTGAATGC and CTCGGCTGAGTTTGAAGTTGAAGATG for the apoB gene. Animal Experiments—All mice were on a pure-bred FVB-NIco background. Male animals were kept in normal cages with food and water ad libitum. Mice in the fed state were sacrificed at the beginning of the light cycle. Mice in the fasted state were deprived of food for 6 h starting at the beginning of the light cycle. At the time of sacrifice animals were between 2 and 4 months of age. For the diet intervention, 2-month-old male mice were fed with a low or high fat diet for 10 weeks. The respective diets provided either 10 or 45% energy percent in the form of lard fat (D12450B or D12451, Research Diets, New Brunswick, NJ). Blood was collected via orbital puncture into EDTA tubes. Tissues were excised, weighed, and immediately frozen in liquid nitrogen. Intragastric Lipid Loading Test—Mice fasted for 16 h were administered 350 μl of olive oil (Bertolli, Extra Virgin) by intragastric gavage. Blood was collected by tail bleeding every 2 h for plasma TG measurement. VLDL Production Test—Mice fasted for 16 h were injected via the tail vein with 500 mg/kg bodyweight Triton WR1339 (Tyloxapol) under general anesthesia. Blood was collected by tail bleeding at several time points during 2.5 h for plasma TG measurement. Intraperitoneal Glucose Tolerance Test—After a 5-h fast mice were injected intraperitoneally with glucose (2 g/kg bodyweight on chow, 1 g/kg bodyweight on low fat/high fat diet). Blood was collected by tail bleeding after 0, 20, 40, 60, 90, and 150 min, and glucose was measured using Accucheck compact (Roche Diagnostics, Almere, the Netherlands). Insulin Tolerance Test—After a 5-h fast mice were injected intraperitoneally with insulin (0.75 unit/kg bodyweight). Blood was collected by tail bleeding after 0, 20, 40, and 60 min, and glucose was measured using Accucheck compact. Plasma Metabolites—Plasma was obtained from blood by centrifugation for 10 min at 10,000 × g. The plasma glucose concentration was determined using a kit from Elitech (Sopachem, Wageningen, the Netherlands). Plasma and tissue triglycerides, plasma glycerol, and plasma total and HDL-cholesterol concentration were determined using kits from Instruchemie (Delfzijl, the Netherlands). Plasma free fatty acids were determined using a kit from WAKO Chemicals (Sopachem, Wageningen, the Netherlands). Lipoprotein Profiling—Lipoproteins were separated using fast protein liquid chromatography (FPLC). 0.2 ml of pooled mouse plasma or human plasma was injected onto a Superose 6B 10/30 column (Amersham Biosciences) and eluted at a constant flow of 0.5 ml/min with PBS (pH 7.4). The effluent was collected in 0.5-ml fractions, and triglyceride and cholesterol levels were determined. Plasma LPL and Hepatic Lipase Level Assay—Plasma lipoprotein lipase (LPL) and hepatic lipase levels were determined in post-heparin plasma as described before (17Berbee J.F. van der Hoogt C.C. Sundararaman D. Havekes L.M. Rensen P.C. J. Lipid Res. 2005; 46: 297-306Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). Q-PCR—Total RNA was extracted from tissues with TRIzol reagent (Invitrogen). 1 μg of total RNA was reverse-transcribed with iScript (Bio-Rad). cDNA was PCR-amplified with Platinum Taq DNA polymerase (Invitrogen) on a Bio-Rad iCycler apparatus. Primers were designed to generate a PCR amplification product of 100-150 bp. Specificity of the amplification was verified by melt curve analysis and evaluation of efficiency of PCR amplification. Sequences of primers used are available on request. Immunoblot—Immunoblotting on FPLC fractions, plasma, and tissues was carried out as described previously (6Kersten S. Mandard S. Tan N.S. Escher P. Metzger D. Chambon P. Gonzalez F.J. Desvergne B. Wahli W. J. Biol. Chem. 2000; 275: 28488-28493Abstract Full Text Full Text PDF PubMed Scopus (467) Google Scholar, 7Mandard S. Zandbergen F. Tan N.S. Escher P. Patsouris D. Koenig W. Kleemann R. Bakker A. Veenman F. Wahli W. Muller M. Kersten S. J. Biol. Chem. 2004; 279: 34411-34420Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar). FIAF antibodies were directed against peptide epitopes within the N-terminal region of the mFIAF and hFIAF proteins. Microarray—RNA was prepared from adipose tissue of 10 wild-type and 10 FIAF-Tg mice using TRIzol and subsequently pooled per group. Pooled RNA was further purified using Qiagen RNeasy columns and the quality verified by laboratory on a chip analysis (Bioanalyzer 2100, Agilent). 1 μg of RNA was used for one cycle cRNA synthesis (Affymetrix, Santa Clara, CA). Hybridization, washing, and scanning of Affymetrix GeneChip mouse genome 430 2.0 arrays were carried out according to standard Affymetrix protocols. Fluorometric data were processed by Affymetrix GeneChip Operating software, and the gene chips were globally scaled to all the probe sets with an identical target intensity value. Further analysis was performed by Data Mining Tool (Affymetrix). Ethical Considerations—The animal experiments were approved by the animal experimentation committee of Wageningen University. All human experiments were approved by the medical ethics committee of Wageningen University. FIAF-Tg Mice Have Reduced Fat Mass—To investigate the role of FIAF in mammalian metabolism we generated transgenic mice that express mouse FIAF under the influence of the aP2 promoter (Fig. 1A), aiming at adipose tissue-specific overexpression (18Ross S.R. Graves R.A. Greenstein A. Platt K.A. Shyu H.L. Mellovitz B. Spiegelman B.M. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 9590-9594Crossref PubMed Scopus (184) Google Scholar). Two independent transgenic lines were obtained. Results presented here are from one transgenic line that showed up-regulation of mFIAF mRNA to levels similar to those achieved after fasting. FIAF mRNA was modestly up-regulated in white and brown adipose tissue and, unexpectedly, in skeletal muscle and heart, yet not in liver (Fig. 1, B and C). Activity of the transgene in skeletal muscle and heart is probably due to the design of the targeting vector, which contains the intact mouse FIAF gene. Previously, we showed that mFIAF gene expression is responsive to PPARs via a PPAR-responsive element within intron 3 (7Mandard S. Zandbergen F. Tan N.S. Escher P. Patsouris D. Koenig W. Kleemann R. Bakker A. Veenman F. Wahli W. Muller M. Kersten S. J. Biol. Chem. 2004; 279: 34411-34420Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar). For reasons that are not clear this did not cause FIAF overexpression in liver. The difference in FIAF mRNA between wild-type and FIAF-Tg mice was translated at the protein level, as observed by immunoblot on adipose tissue using purified anti-mFIAF antibody (Fig. 1D). A modest increase in FIAF-S2 and FIAF-S1 proteins, which are the predominant and most easily detectable (truncated) forms of mFIAF in blood plasma (7Mandard S. Zandbergen F. Tan N.S. Escher P. Patsouris D. Koenig W. Kleemann R. Bakker A. Veenman F. Wahli W. Muller M. Kersten S. J. Biol. Chem. 2004; 279: 34411-34420Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar), was observed in plasma of FIAF-Tg mice (Fig. 1E and data not shown). Strikingly, FIAF-Tg mice weighed significantly less than their wild-type littermates (Fig. 2A). This was mostly due to a decrease in white adipose tissue (WAT) weight, which was ∼50% lower in FIAF-Tg mice (Fig. 2A). In contrast, liver weight as well as the weight of numerous other organs was unaltered in the FIAF-Tg mice, whereas weight of brown adipose tissue (BAT) was modestly decreased. Reduced WAT weight was due to a decrease in adipocyte size (Fig. 2B). Reduced adiposity was not the result of diminished food intake, which was identical between the two sets of mice (Fig. 2C). These data indicate that FIAF overexpression results in loss of body fat. FIAF-Tg Mice Display Hypertriglyceridemia and Hypercholesterolemia—To get a better understanding of the potential cause of the reduced adiposity, plasma lipid parameters were assessed in both fed and 6-h-fasted mice. A marked increase in plasma TG levels was observed in the FIAF-Tg mice, which was independent of their feeding status (Fig. 3A). In addition, plasma FFAs, glycerol, total cholesterol, and HDL-cholesterol concentrations were significantly elevated in the FIAF-Tg mice in the fasted state, whereas glucose was unchanged (Fig. 3, B-F). Thus, FIAF overexpression has profound effects on plasma lipid concentrations. Profiling of lipoproteins using FPLC analysis revealed that the increase in plasma TG in the 6-h-fasted animals was attributable to the VLDL fractions (Fig. 4A), whereas the increase in cholesterol was found in both VLDL and HDL fractions (Fig. 4B). Immunoblotting demonstrated a marked increase in apoB48 and apoB100 protein content in the VLDL fractions in FIAF-Tg mice, indicating that the number of VLDL particles in plasma is increased in these mice (Fig. 4C). It should be emphasized that in mice apoB48 accounts for approximately two-thirds of total apoB production in liver. Similarly, we found markedly increased apoAI and apoAII protein content in the HDL fractions in FIAF-Tg mice, pointing toward a pronounced increase in the number of HDL particles (Fig. 4D).FIGURE 4Fasting plasma VLDL and HDL concentrations are elevated in FIAF-Tg mice. Pooled plasma of 10 wild-type mice (gray squares) and 10 FIAF-Tg mice (black squares) fasted for 6 h was used for lipoprotein profiling by FPLC. Fractions were assayed for triglycerides (A), and cholesterol (B). Elution of protein markers is indicated. C, immunoblot of plasma lipoprotein FPLC fractions representing VLDL using antibody against apoB. D, immunoblot of plasma lipoprotein FPLC fractions representing HDL using antibody against apoAI and apoAII.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Elevated plasma TG levels in 6-h-fasted mice can either be due to increased VLDL production or impaired clearance. To discriminate between these two possibilities, a VLDL production test and an oral lipid-loading test were performed. Despite greatly elevated 16-h fasting plasma TG levels in FIAF-Tg mice, no difference in VLDL production, determined by the slope of increase of plasma TG after injection with Triton WR1339, was observed between wild-type and FIAF-Tg mice (Fig. 5A). In contrast, an oral lipid loading test revealed dramatic differences between the two sets of mice. Whereas in wild-type mice intragastric loading of olive oil caused a moderate and transient increase in plasma TG, in FIAF-Tg mice plasma TG went up dramatically, reaching levels of almost 35 mm after 8 h (Fig. 5B). Post-prandial plasma FFAs were also significantly increased in FIAF-Tg mice (Fig. 5C). These data show unambiguously that clearance of TG-rich apoB-containing lipoproteins is severely inhibited by FIAF overexpression. Impaired clearance of plasma TG can be caused by decreased presence of lipoprotein lipase (LPL) and/or inhibition of LPL activity. To investigate whether total LPL content was altered, total hepatic lipase and LPL levels were determined in post-heparin plasma. Plasma levels of these lipases were very similar between wild-type and FIAF-Tg mice (Fig. 5, D and E). In addition, mRNA expression and tissue protein level of LPL in adipose tissue or skeletal muscle were not affected by FIAF overexpression (Fig. 5F and data not shown). This suggests that FIAF may block in vivo LPL activity, rather than reduce total LPL level. This notion is backed up by evidence showing inhibition of LPL activity by recombinant FIAF (10Yoshida K. Shimizugawa T. Ono M. Furukawa H. J. Lipid Res. 2002; 43: 1770-1772Abstract Full Text Full Text PDF PubMed Scopus (309) Google Scholar). Together, these data indicate that FIAF overexpression markedly impairs plasma TG clearance and accordingly raises plasma TG levels, most likely via inhibition of LPL activity. FIAF Is Physically Associated with HDL—Several proteins that are physically associated with plasma lipoproteins, including apoCs and apoAV, are known to influence LPL activity. Because FIAF overexpression increases fasting plasma VLDL and HDL levels, we hypothesized that FIAF may be physically associated with VLDL or HDL. To determine whether this is the case, immunoblotting was performed on the corresponding mouse plasma FPLC fractions using a well characterized anti-mFIAF antibody (6Kersten S. Mandard S. Tan N.S. Escher P. Metzger D. Chambon P. Gonzalez F.J. Desvergne B. Wahli W. J. Biol. Chem. 2000; 275: 28488-28493Abstract Full Text Full Text PDF PubMed Scopus (467) Google Scholar). Remarkably, it was observed that full-length FIAF was present in the HDL fractions, but not in the VLDL fractions (Fig. 6A). A small band, corresponding to truncated FIAF-S2 (7Mandard S. Zandbergen F. Tan N.S. Escher P. Patsouris D. Koenig W. Kleemann R. Bakker A. Veenman F. Wahli W. Muller M. Kersten S. J. Biol. Chem. 2004; 279: 34411-34420Abstract Full Text Full Text PDF PubMed Scopus (222) Google Scholar), was visible in the first HDL fraction that overlaps with LDL. Further analysis clearly demonstrated that truncated FIAF is present in the LDL fractions. Both full-length and truncated FIAF were more abundant in the HDL and LDL fractions, respectively, of FIAF-Tg mice (Fig. 6B). Remarkably, in human plasma full-length FIAF, truncated FIAF and FIAF multimers, the latter visualized by omitting dithiothreitol from the loading buffer, were associated exclusively with HDL (Fig. 6, C and D). Together, the data indicate that in mice the full-length form of FIAF is physically associated with HDL, whereas truncated FIAF is associated with LDL. In human both full-length and truncated FIAF are associated with HDL. The latter observation suggests that plasma FIAF and HDL levels may be correlated. Indeed, in a group of 16 healthy young subjects we found a significant positive correlation (r = 0.57, p = 0.01) between plasma levels of FIAF-S2 and HDL (Fig. 6E). In contrast, no significant correlation was found for plasma TG or LDL (data not shown). Gene Expression Changes in WAT of FIAF-Tg Mice Indicate Enhanced Lipolysis and Oxidative Metabolism—The parallel increase in plasma FFAs and glycerol in FIAF-Tg mice cannot be explained by inhibition of plasma TG catabolism but rather is indicative of enhanced adipose tissue lipolysis. Lipolysis is catalyzed by hormone-sensitive lipase (HSL), which is tightly regulated by numerous external stimuli, including insulin and β-adrenergic activity. mRNA expression of HSL was not altered in the FIAF-Tg mice, nor that of other genes that could affect plasma FFA and glycerol concentrations, including monoglyceride lipase, perilipin, PPARγ, and glycerol kinase (Fig. 7). Interestingly, expression of adipose triglyceride lipase (ATGL)/desnutrin, a recently discovered adipose tissue lipase that works in conjunction with HSL (19Villena J.A. Roy S. Sarkadi-Nagy E. Kim K.H. Sul H.S. J. Biol. Chem. 2004; 279: 47066-47075Abstract Full Text Full Text PDF PubMed Scopus (510) Google Scholar, 20Zimmermann R. Strauss J.G. Haemmerle G. Schoiswohl G. Birner-Gruenberger R. Riederer M. Lass A. Neuberger G. Eisenhaber F. Hermetter A. Zechner R. Science. 2004; 306: 1383-1386Crossref PubMed Scopus (1533) Google Scholar), was 50% increased in the FIAF-Tg mice. Accordingly, the elevated plasma FFA and glycerol levels are likely caused by enhanced lipolysis possibly via up-regulation of ATGL/desnutrin. After hydrolysis of plasma TG by LPL, the fatty acids that enter the adipocyte can either be reconverted into triglycerides and stored as such, or can be oxidized for
0

Kupffer cells promote hepatic steatosis via interleukin-1β-dependent suppression of peroxisome proliferator-activated receptor α activity

Rinke Stienstra et al.Sep 29, 2009
Kupffer cells have been implicated in the pathogenesis of various liver diseases. However, their involvement in metabolic disorders of the liver, including fatty liver disease, remains unclear. The present study sought to determine the impact of Kupffer cells on hepatic triglyceride storage and to explore the possible mechanisms involved. To that end, C57Bl/6 mice rendered obese and steatotic by chronic high-fat feeding were treated for 1 week with clodronate liposomes, which cause depletion of Kupffer cells. Loss of expression of marker genes Cd68, F4/80, and Clec4f, and loss of Cd68 immunostaining verified almost complete removal of Kupffer cells from the liver. Also, expression of complement components C1, the chemokine (C-C motif) ligand 6 (Ccl6), and cytokines interleukin-15 (IL-15) and IL-1β were markedly reduced. Importantly, Kupffer cell depletion significantly decreased liver triglyceride and glucosylceramide levels concurrent with increased expression of genes involved in fatty acid oxidation including peroxisome proliferator-activated receptor alpha (PPARα), carnitine palmitoyltransferase 1A (Cpt1α), and fatty acid transport protein 2 (Fatp2). Treatment of mice with IL-1β decreased expression of PPARα and its target genes, which was confirmed in primary hepatocytes. Consistent with these data, IL-1β suppressed human and mouse PPARα promoter activity. Suppression of PPARα promoter activity was recapitulated by overexpression of nuclear factor κB (NF-κB) subunit p50 and p65, and was abolished upon deletion of putative NF-κB binding sites. Finally, IL-1β and NF-κB interfered with the ability of PPARα to activate gene transcription. Conclusion: Our data point toward important cross-talk between Kupffer cells and hepatocytes in the regulation of hepatic triglyceride storage. The effect of Kupffer cells on liver triglycerides are at least partially mediated by IL-1β, which suppresses PPARα expression and activity. (HEPATOLOGY 2010.)
Load More