ALOX5AP expression, but not gene haplotypes, is associated with ...

International Journal of Obesity (2006) 30, 447–452 & 2006 Nature Publishing Group All rights reserved 0307-0565/06 $30.00 www.nature.com/ijo

ORIGINAL ARTICLE ALOX5AP expression, but not gene haplotypes, is associated with obesity and insulin resistance ¨ m1, A Sicard3, A Bouloumie´3,4,5, D Langin3,4, M Kaaman1, M Ryde´n1, T Axelsson2, E Nordstro 1 1 P Arner and I Dahlman 1 Department of Medicine, Karolinska Institutet, Stockholm, Sweden; 2Department of Medical Sciences, Uppsala University, Uppsala, Sweden; 3Obesity Research Unit Inserm U586, Toulouse, France; 4Universite´ Paul Sabatier, Toulouse, France and 5 Cardiovascular Physiology Department, J-W Goethe University, Frankfurt am Main, Germany

Objective: Inflammation in adipose tissue may link obesity to insulin resistance and atherosclerosis. Arachidonate 5lipoxygenase activating protein (ALOX5AP) gene is involved in the pathogenesis of atherosclerotic cardiovascular disease (CVD). We investigated ALOX5AP expression in adipose tissue, and association of gene polymorphisms with obesity and insulin resistance. Design: For gene expression analysis in adipose tissue, we studied 12 lean and 36 obese women, eight lean and 13 obese men, and nine women before and 2–4 years after gastric banding surgery. For genetic analysis, we studied 231 nonobese and 350 obese men. Results: The ALOX5AP protein, 5-lipoxygenase activating protein (FLAP), as well as 5-lipoxygenase (5LO) itself, were detected in adipocytes. The mRNA expression of ALOX5AP in subcutaneous adipose tissue was increased in obesity and normalized following weight reduction. High adipose tissue mRNA expression of ALOX5AP is associated with insulin resistance as measured by homeostasis model assessment (HOMAIR). ALOX5AP haplotypes that associate with CVD are not associated with obesity or insulin resistance. Conclusion: ALOX5AP is present in adipose tissue, where its expression is associated with body weight and HOMAIR, and may provide a link between adipose tissue, inflammation and insulin resistance. Investigated ALOX5AP haplotypes are not major primary risk factors for obesity and insulin resistance. International Journal of Obesity (2006) 30, 447–452. doi:10.1038/sj.ijo.0803147; published online 1 November 2005 Keywords: fat cell; atherosclerosis; adipocity; inflammation; leukotrienes

Introduction Obesity, insulin resistance, type 2 diabetes, and atherosclerotic cardiovascular disease (CVD) are closely associated and may have inflammation as a common denominator. Inflammation plays an important role in the development of CVD.1,2 A low-grade inflammation is often present in insulinresistant states such as obesity and predict type 2 diabetes.3,4 The low-grade inflammation of obesity is characterized by macrophage infiltration and increased expression of inflammatory genes in adipose tissue of humans and mice, and this inflammation may play an important role for obesity-related insulin resistance.5,6 Through secretion of proinflammatory

Correspondence: Dr I Dahlman, Department of Medicine M63, Karolinska University Hospital Huddinge, SE-141 86 Stockholm, Sweden. E-mail: [email protected] Received 22 March 2005; revised 19 July 2005; accepted 10 September 2005; published online 1 November 2005

factors into the circulation, adipose tissue inflammation can have systemic effects.7 However, the details of adipose tissue inflammation; the factor/s that recruit macrophages, as well as key molecules and pathways that cause insulin resistance are poorly understood. A preliminary analysis of the profile of gene expression in adipose tissue demonstrated that ALOX5AP mRNA is expressed in human subcutaneous adipose tissue (unpublished). ALOX5AP encodes the protein 5-lipoxygenase activating protein (FLAP), which facilitates the initial step in leukotriene biosynthesis from arachidonic acid catalyzed by 5-lipoxygenase (5LO).8 The most important product of this pathway is leukotriene B4 (LTB4). LTB4 is a ligand for peroxisome proliferator receptor alpha (PPARA), which has important roles in lipid metabolism, as well as peroxisome and mitochondrial beta-oxidation.9,10 Furthermore, LTB4 strongly induces monocyte chemoattractant protein 1 (MCP1) production in human monocytes.11 MCP1 is increased in the circulation of obese subjects, is produced

ALOX5AP and obesity M Kaaman et al

448 by adipose tissue, and decreases glucose uptake in adipocytes.12–14 Both MCP1 and 5LO products recruit leukocytes to sites of inflammation.15 These data indirectly suggest that ALOX5AP also might play a role in obesity. Recently, common haplotypes in the ALOX5AP gene were shown to confer risk of myocardial infarction and stroke, and to be associated with more leukotriene production in neutrophiles.16 In the study by Helgadottir et al.16 ALOX5AP was shown to have a much stronger impact on CVD among men than women. Previously, FLAP has been reported to be primarily expressed in bone marrow-derived cells, and has been implicated in asthma and allergic disease (reviewed by Peters-Golden and Brock 17). Based on the above functions ascribed to 5LO and FLAP, we hypothesize that these genes could have a role in obesity and insulin resistance. We therefore investigated 5LO and FLAP expression in the various cell types of human adipose tissue, studied the influence of obesity on ALOX5AP in this tissue, and explored whether the CVD-associated ALOX5AP haplotypes were associated with obesity and insulin resistance in Swedish men.

Methods Study subjects and clinical examination For genetic analysis we studied 231 nonobese (age 4375 years, body mass index (BMI) 2472 kg/m2, homeostasis model assessment (HOMAIR) 1.771.3, Plasma (Pl)-HDL cholesterol 1.370.4 mmol/l, Pl-triglycerides 1.371.0 mmol/ l) and 350 obese (age 47711 years, BMI 4075 kg/m2, HOMAIR 7.476.6, Pl-HDL cholesterol 1.170.3 mmol/l, Pltriglycerides 2.171.2 mmol/l) men. The obese men included 24 men with hypertension, 13 men with dyslipidemia, and 54 men with type 2 diabetes, who were treated with diet only, sulfonylurea, or metformin. This sample provide 85% power to detect as association with obesity assuming odds ratio ¼ 1.5, and haplotype or allele frequency 10% with additive impact on obesity (http://statgen.iop.kcl.ac.uk/cgibin/powercalc/cc2.cgi). Higher power was obtained for a quantitative trait locus explaining 10% of variation in phenotype. For ALOX5AP mRNA measurement in adipose tissue, subjects from the Stockholm area were consecutively recruited by local advertising to study metabolic regulation in fat cells. It is easier to obtain obese study subjects. We studied 12 lean (BMI 2372 kg/m2, age 3378 years, HOMAIR, 1.470.9, Pl-cholesterol 4.870.7 mmol/l, Pl-HDL cholesterol 1.670.4 mmol/l, Pl-triglycerides 0.870.2 mmol/l) and 36 obese (BMI 4076 kg/m2, age 3575 years, HOMAIR 3.472.2, Pl-cholesterol 5.071.0 mmol/l, Pl-HDL cholesterol 1.170.2 mmol/l, Pl-triglycerides 1.771.3 mmol/l) women, eight lean (BMI 2272 kg/m2, age 2779 years, HOMAIR 1.070.3, Pl-cholesterol 4.270.4 mmol/l, Pl-HDL cholesterol 1.370.4 mmol/l, Pl-triglycerides 1.070.4 mmol/l) and 13 International Journal of Obesity

obese (BMI 3875 kg/m2, age 3777 years, HOMAIR 4.274.0, Pl-cholesterol 5.171.0 mmol/l, Pl-HDL cholesterol 1.070.2 mmol/l, Pl-triglycerides 1.771.2 mmol/l) men, and nine women before (BMI 4275 kg/m2, age 3778 years, HOMAIR 5.273.7) and 2–4 years after gastric banding surgery (BMI 2873 kg/m2, HOMAIR 1.470.5).18 when body weight had reached a steady-state level. For measuring mRNA and protein in different cell types of adipose tissue, subcutaneous adipose tissue was obtained in connection with abdominal liposuction because of cosmetic reasons. All subjects in whom mRNA was quantified were healthy. All individuals came to the laboratory at 0800 after an overnight fast. Height, weight, and waist circumference were determined and a fasting venous blood sample was obtained for measurements of total cholesterol, HDL cholesterol, triglycerides, glucose, Insulin, and DNA extraction as described19,20 Insulin resistance index HOMAIR was calculated as fasting serum insulin (mU/ml) fasting Pl glucose (mmol/l)/22.5.21 In methodological experiments on 60 subjects (BMI 18–50 kg/m2), we found that HOMAIR correlated strongly with direct measurements of insulin sensitivity by the hyperinsulinemic, euglycemic clamp technique (r ¼ 0.83, Po0.001). Needle biopsies of abdominal subcutaneous adipose tissue (1–2 g) were obtained under local anesthesia in the morning after an overnight fast. Adipose tissue was treated with collagenase as described for isolation of adipocytes.22 Isolated fat cells and adipose tissue were immediately frozen in liquid nitrogen for subsequent RNA or protein analysis. Cells from the stroma vascular fraction were separated as described.23 Adipose tissue-derived cells were homogenized in denaturing buffer from RNeasy kit (Qiagen, Courtaboeuf, France) and stored at 801C until RNA preparation. The local committee on ethics approved the project. It was explained in detail to each participant and informed consent was obtained.

RNA extraction and real time PCR Total RNA was extracted from human adipose tissue and reverse transcribed as described.24 Quantitative real-time PCR (RT-qPCR) was performed on an iCycler IQ (Bio-Rad Laboratories, Hercules, CA, USA). ALOX5AP was quantified using TaqMan (Hs00233463_m1, Applied Biosystems, Foster City, CA, USA) with 25 ng of cDNA used in triplicate 25 ml PCR reactions. SYBR Green-based technology was used for the internal control gene b-2 microglobulin (B2MG). The primer pairs for B2MG were designed to span exon–intron boundaries and generate a single amplicon according to dissociation curves and agarose gel electrophoresis. B2MG primers sequences were 50 -TGCTGTCTCCATGTTTGATG TATCT-30 and 50 -TCTCTGCTCCCCACCTCTAAGT-30 . A total of 8 ng of cDNA were used in triplicate 25 ml PCR reactions. In methodological experiments on cDNA from subcutaneous adipose tissue of 200 subjects, CT values for B2MG were not


449 related to BMI or HOMAIR. Expression of ALOX5AP was normalized to the B2MG internal control using the formula 2 (Ct ALOX5AP calibrator – Ct ALOX5AP sample)/2 (Ct B2MG calibrator – Ct B2MG sample), where calibrator is a random sample. Men and women were run on different time-points without controls permitting comparison between groups. For analysis of ALOX5AP mRNA expression in different cell types from adipose tissue, RT-qPCR was performed on a GeneAmp 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) using Taqman technology as described above. 18S ribosomal RNA (Ribosomal RNA Control TaqMan Assay kit, Applied Biosystems, Foster City, CA, USA) was used in these experiments as control to normalize gene expression.

Western blot Total protein was isolated from isolated adipocytes as described.24 Lymphocyte protein was isolated as positive control. Protein (100 mg) was immunoprecipitated overnight at 41C with a polyclonal antibody (Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA). Protein A Sepharose (Pharmacia, Uppsala, Sweden) was added the next day. The samples were boiled at 1001C for 7 min with SDS loading buffer containing b-mercaptoethanol and dithiothreitol (DTT), in order to optimize the reducing conditions. This procedure was made in line with Mandal et al.25 where FLAP was found in dimeric and trimeric form. The samples where run on a 15% SDSpolyacryl gel, blotted onto a PVDF membrane and incubated with a polyclonal FLAP antibody (Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA) or a polyclonal antibody against 5LO kindly donated by Dr Olof Ra˚dmark. Proteins obtained from monocytes and recombinant 5LO (kindly donated by Dr Olof Ra˚dmark at Karolinska Institutet, Stockholm, Sweden) were used as positive controls.

adjusting for age and as indicated for BMI was used for comparison of quantitative phenotypes between groups. w2 test was used to determine whether allele, genotype, or haplotype frequencies differed between affected and nonaffected individuals. Values are means7s.d. unless otherwise indicated. Nominal P-values of 0.05 or less were considered significant.

Results FLAP mRNA and protein are present in adipocytes We demonstrate in fat obtained in connection with abdominal liposuction because of cosmetic reasons that ALOX5AP mRNA is expressed in all cell fractions of adipose tissue; 10-fold higher in the stroma vascular fraction than in isolated adipocytes (Po0.01), and 10-fold higher in infiltrating macrophages than the stroma vascular fraction, Table 1. Furthermore, the ALOX5AP protein, FLAP, is present in human isolated fat cells (Figure 1). Distinct bands were detected at 54, 36, and 18 kDa in isolated fat cells and control cells, that is, lymphocytes, corresponding to published trimers, dimers, and monomers of FLAP.25 In isolated fat cells, we also detected a protein band corresponding to 5-lipogyxenase (Figure 1).

Association of ALOX5AP mRNA expression with obesity and HOMAIR We compared ALOX5AP mRNA expression between obese and lean subjects. Obese subjects expressed significantly Table 1

Expression of ALOX5AP mRNA in human adipose tissue cell types

Adipocytes Stromavascular fraction Preadipocytes Macrophages Endothelial cells

3.6 10671.2 106 3.6 10570.9 105 3.6 10572.4 106 2.7 10471.2 104 2.3 10371.0 103

Genotyping The seven ALOX5AP single nucleotide polymorphisms (SNPs) defining the two haplotypes previously associated with CVD in the Icelandic and British population, respectively, were genotyped by a homogeneous minisequencing assay with fluorescence polarization detection26,27 Primer details and reaction conditions are available from the authors. All assays were in Hardy–Weinberg equilibrium and genotyping accuracy, as determined by randomly retyping 10% of the subjects, was 100%. Haplotypes were inferred using SNPHAP (http://archimedes.well.ox.ac.uk/ pise/snphap-simple.html).

ALOX5AP mRNA level was determined by RT-qPCR and normalized with 18S ribosomal RNA levels. Values are mean7s.e.m. (n ¼ 5–6). Adipocytes versus stromavascular fraction Po0.01, Preadipocytes versus macrophages: P ¼ 0.01, Preadipocytes versus endothelial cells P ¼ 0.009, Macrophages versus endothelial cells P ¼ 0.09.

Statistical analysis Gene expression in different cell types was compared using Mann–Whitney test. Values for HOMAIR and Pl triglycerides were 10log transformed before statistical analysis. ANOVA

Figure 1 Western blot analysis. (a) FLAP protein in isolated adipocytes and lymphocytes (L). Three bands were detected at 54, 36, and 18 kDa corresponding to tri- di- and monomers of FLAP. (b) 5-Lipoxygenase in isolated adipocytes from three subjects. Recombinant protein ( þ ) was used as positive control.

International Journal of Obesity


450 more ALOX5AP in abdominal subcutaneous adipose tissue than lean subjects, for men 8.874.8 versus 4.270.9 arbitrary units (AU) (P ¼ 0.02), and for women 11.476.6 versus 4.873 AU (P ¼ 0.002). In obese individuals undergoing gastric banding for weight reduction, adipose tissue ALOX5AP expression was significantly downregulated from 11.078.0 to 5.572.4 AU (P ¼ 0.05). We next divided the subjects into tertiles based on ALOX5AP mRNA expression. High expression of ALOX5AP mRNA was associated with higher HOMAIR independently of BMI in both women (P-valueo0.01) and men (P-value o0.05), Figure 2.

as compared to nonobese men. Genotyping success rate for the SNP sg13s35 was 96%, for the other SNPs success rate was 499%. No SNP was associated with obesity (Table 2) or investigated quantitative traits including HOMAIR or Pl lipids (results not shown). SHPHAP inferred nine haplotypes with frequency 41% (Table 3). The haplotypes associated with CVD on Iceland and in a UK population were detected with frequencies 12 and 8%, respectively. No haplotype was associated with obesity or investigated quantitative traits (results not shown). Excluding subjects with diabetes, dyslipidemia, and hypertension had no impact on the results.

Analysis of association between ALOX5AP alleles and obesity phenotypes We investigated whether ALOX5AP alleles are associated with obesity and associated metabolic disturbances in men. For this purpose, we genotyped 231 nonobese and 350 obese men for seven common SNPs defining two haplotypes previously shown to contribute to CVD.16 Only men were studied since the association with CVD was much stronger in men than in women. As expected the obese men displayed more extensive insulin resistance syndrome phenotypes including high HOMAIR (7.476.6), low Pl-HDL cholesterol (1.170.3 mmol/l), and high Pl-triglycerides (2.171.2 mmol/l)

Discussion ALOX5AP is the first identified gene that confers substantial population attributable risk for atherogenic CVD.16 We demonstrate that both the ALOX5AP protein FLAP and 5LO are present in fat cells. However, the quantitative comparison of gene expression suggests that ALOX5AP levels are lower in adipocytes than in stroma vascular cells of adipose tissue, such as macrophages and endothelial cells. We show that increased gene expression of ALOX5AP in adipose tissue is associated with an insulin-resistance phenotype, which, in turn, is a risk factor for CVD. First, the expression of ALOX5AP was markedly upregulated in obesity. Second, subjects with high ALOX5AP mRNA level in

Table 3 ALOX5AP haplotypes Haplotype

Allelesa

Frequency (%)

GGTGCAG GGTGAAG AGTGAAG GGAGAAG GAAGAAG GAAGAAA GGAACGG GGAGCAG GGTGAAA

42 12 10 8 8 8 3 3 2

Figure 2 HOMAIR in the tertile with highest (N ¼ 16 women or N ¼ 7 men,

H1 H2b H3 H4 H5c H6 H7 H8 H9b

filled bars) or lowest (N ¼ 16 women or N ¼ 7 men, striped bars) ALOX5AP mRNA expression in adipose tissue using ANOVA with BMI as covariate. Shown are mean7s.d. *P-valueo0.05. **P-valueo0.01.

a SNPs in order sg13s25, sg13s377, sg13s114, sg13s89, sg13s32, sg13s41, sg13s35. The haplotype associated with CVD bIn Iceland and in cThe UK.

Table 2 Position

ALOX5AP genotypes

a

29097553 29101965 29110096 29124441 29130547 29134045 29138117 a

Markerb

Non obese

sg13s25 sg13s377 sg13s114 sg13s89 sg13s32 sg13s41 sg13s35

GG ¼ 183 GG ¼ 163 TT ¼ 112 GG ¼ 216 CC ¼ 64 AA ¼ 207 GG ¼ 186

AG ¼ 45 AG ¼ 58 AT ¼ 92 AG ¼ 15 AC ¼ 108 AG ¼ 22 AG ¼ 44

AA ¼ 1 AA ¼ 8 AA ¼ 26 AA ¼ 0 AA ¼ 59 GG ¼ 1 AA ¼ 1

Position on human chromosome 13 according to Build 34. bAccording to Helgadottir et al.16


w2 P-value

Obese GG ¼ 273 GG ¼ 240 TT ¼ 158 GG ¼ 321 CC ¼ 90 AA ¼ 303 GG ¼ 260

AG ¼ 72 AG ¼ 96 AT ¼ 159 AG ¼ 27 AC ¼ 174 AG ¼ 44 AG ¼ 63

AA ¼ 2 AA ¼ 11 AA ¼ 31 AA ¼ 0 AA ¼ 84 GG ¼ 0 AA ¼ 4

0.92 0.82 0.34 0.57 0.74 0.25 0.62


451 adipose tissue had higher HOMAIR in comparison to those with low mRNA. Although HOMAIR is quite a crude marker of insulin resistance, in separate methodological experiments, it correlates strongly with direct measurements of insulin sensitivity by the hyperinsulinemic, euglycemic clamp technique (r ¼ 0.83). Adipose ALOX5AP is probably a regulator of the insulin resistance phenotype among the obese and not a primary cause of obesity because mRNA levels for ALOX5AP were normalized after weight reduction. FLAP and 5LO protein in adipocytes, which catalyze the initial step in leukotriene synthesis from arachidonic acid, could potentially contribute to recruitment of the inflammatory cells observed in human and experimental obesity, which has been implicated in development of insulin resistance.5,6 Potential local effects of leukotrienes in adipose tissue include macrophage recruitment, nitric oxide synthesase (NOS) production, induction of MCP-1,11 and IL6 production.28 These effectors are implicated both in inducing local insulin resistance in the adipose tissue and could also, by secretion into the circulation or changing the secretory function of adipocytes, contribution to generalized insulin resistance and, ultimately, the development of CVD. Although we propose several mechanisms by which ALOX5AP in adipose tissue could contribute to insulin resistance and CVD, the CVD-associated ALOX5AP haplotypes are not associated with obesity, HOMAIR, or dyslipidemia in men. The CVD-associated haplotypes are observed with similar frequencies in the present and previous investigated populations.16 This support that the haplotype structure is similar in studied populations. Furthermore, investigated obese subjects displayed a high frequency of insulin resistance as measured by HOMAIR. With this in mind, we conclude that the CVD-associated ALOX5AP haplotypes are not major primary causes of obesity and insulin resistance in men. A minor importance of the ALOX5AP haplotypes on these phenotypes cannot be excluded with our sample. Nor can we exclude an impact of other ALOX5AP polymorphisms on obesity and its metabolic complications, or an impact on these phenotypes in women. In summary, FLAP and 5LO protein are detected in adipocytes and other cells in human adipose tissue. ALOX5AP can potentially contribute to recruitment of inflammatory cells, and is upregulated in obese and insulin resistance subjects, thus providing a link between adipose tissue, inflammation, and insulin resistance. The CVD-associated ALOX5AP haplotypes are not major primary risk factors for obesity and insulin resistance.

Acknowledgements The excellent technical assistance of Kerstin Wa˚hle´n, Katarina Hertel, and Britt-Marie Leijonhufvud is greatly acknowledged. This study was supported by grants from

the Swedish Research Council, Swedish Heart and Lung Foundation, Novo Nordic foundation, Swedish Diabetes Association. King Gustaf and Queen Victoria Foundation, Diabetes Research and Wellness Foundation, EFSD, Karolinska Institutet, the K&A Wallenberg foundation via Wallenberg Consortium North (WCN), and the Swedish Society of Medicine.

References 1 Danesh J, Whincup P, Walker M, Lennon L, Thomson A, Appleby P et al. Low grade inflammation and coronary heart disease: prospective study and updated meta-analyses. BMJ 2000; 321: 199–204. 2 Libby P. Inflammation in atherosclerosis. Nature 2002; 420: 868–874. 3 Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. Jama 2001; 286: 327–334. 4 Barzilay JI, Abraham L, Heckbert SR, Cushman M, Kuller LH, Resnick HE et al. The relation of markers of inflammation to the development of glucose disorders in the elderly: the Cardiovascular Health Study. Diabetes 2001; 50: 2384–2389. 5 Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante Jr AW. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 2003; 112: 1796–1808. 6 Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ et al. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest 2003; 112: 1821– 1830. 7 Arner P. The adipocyte in insulin resistance: key molecules and the impact of the thiazolidinediones. Trends Endocrinol Metab 2003; 14: 137–145. 8 Dixon RA, Diehl RE, Opas E, Rands E, Vickers PJ, Evans JF et al. Requirement of a 5-lipoxygenase-activating protein for leukotriene synthesis. Nature 1990; 343: 282–284. 9 Bocher V, Pineda-Torra I, Fruchart JC, Staels B. PPARs: transcription factors controlling lipid and lipoprotein metabolism. Ann NY Acad Sci 2002; 967: 7–18. 10 Mandard S, Muller M, Kersten S. Peroxisome proliferatoractivated receptor alpha target genes. Cell Mol Life Sci 2004; 61: 393–416. 11 Huang L, Zhao A, Wong F, Ayala JM, Struthers M, Ujjainwalla F et al. Leukotriene B4 strongly increases monocyte chemoattractant Protein-1 in human monocytes. Arterioscler Thromb Vasc Biol 2004; 24: 1783–1788. 12 Bruun JM, Lihn AS, Pedersen SB, Richelsen B. MCP-1 release is higher in visceral than subcutaneous human adipose tissue. implication of macrophages resident in the adipose tissue. J Clin Endocrinol Metab 2005; 90: 2282–2289. 13 Christiansen T, Richelsen B, Bruun JM. Monocyte chemoattractant protein-1 is produced in isolated adipocytes, associated with adiposity and reduced after weight loss in morbid obese subjects. Int J Obes Relat Metab Disord 2005; 29: 146–150. 14 Sartipy P, Loskutoff DJ. Monocyte chemoattractant protein 1 in obesity and insulin resistance. Proc Natl Acad Sci USA 2003; 100: 7265–7270. 15 Silva AR, de Assis EF, Caiado LF, Marathe GK, Bozza MT, McIntyre TM et al. Monocyte chemoattractant protein-1 and 5-lipoxygenase products recruit leukocytes in response to platelet-activating factor-like lipids in oxidized low-density lipoprotein. J Immunol 2002; 168: 4112–4120. 16 Helgadottir A, Manolescu A, Thorleifsson G, Gretarsdottir S, Jonsdottir H, Thorsteinsdottir U et al. The gene encoding 5lipoxygenase activating protein confers risk of myocardial infarction and stroke. Nat Genet 2004; 36: 233–239.



452 17 Peters-Golden M, Brock TG. 5-lipoxygenase and FLAP. Prostaglandins Leukot Essent Fatty Acids 2003; 69: 99–109. 18 Lofgren P, Hoffstedt J, Ryden M, Thorne A, Holm C, Wahrenberg H et al. Major gender differences in the lipolytic capacity of abdominal subcutaneous fat cells in obesity observed before and after long-term weight reduction. J Clin Endocrinol Metab 2002; 87: 764–771. 19 Eriksson P, Reynisdottir S, Lonnqvist F, Stemme V, Hamsten A, Arner P. Adipose tissue secretion of plasminogen activator inhibitor-1 in non-obese and obese individuals. Diabetologia 1998; 41: 65–71. 20 Hoffstedt J, Arvidsson E, Sjolin E, Wahlen K, Arner P. Adipose tissue adiponectin production and adiponectin serum concentration in human obesity and insulin resistance. J Clin Endocrinol Metab 2004; 89: 1391–1396. 21 Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB et al. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care 2000; 23: 57–63. 22 Reynisdottir S, Wahrenberg H, Carlstrom K, Rossner S, Arner P. Catecholamine resistance in fat cells of women with upper-body


23

24

25

26

27

28

obesity due to decreased expression of beta 2-adrenoceptors. Diabetologia 1994; 37: 428–435. Curat CA, Miranville A, Sengenes C, Diehl M, Tonus C, Busse R et al. From blood monocytes to adipose tissue-resident macrophages: induction of diapedesis by human mature adipocytes. Diabetes 2004; 53: 1285–1292. Arvidsson E, Blomqvist L, Ryden M. Depot-specific differences in perilipin mRNA but not protein expression in obesity. J Intern Med 2004; 255: 595–601. Mandal AK, Skoch J, Bacskai BJ, Hyman BT, Christmas P, Miller D et al. The membrane organization of leukotriene synthesis. Proc Natl Acad Sci USA 2004; 101: 6587–6592. Syvanen AC, Aalto-Setala K, Harju L, Kontula K, Soderlund H. A primer-guided nucleotide incorporation assay in the genotyping of apolipoprotein E. Genomics 1990; 8: 684–692. Chen X, Levine L, Kwok PY. Fluorescence polarization in homogeneous nucleic acid analysis. Genome Res 1999; 9: 492–498. Andersson P, Serhan CN, Petasis NA, Palmblad J. Interactions between lipoxin A4, the stable analogue 16-phenoxy-lipoxin A4 and leukotriene B4 in cytokine generation by human monocytes. Scand J Immunol 2004; 60: 249–256.