High Expression of Complement Components in ... - Wiley Online Library

High Expression of Complement Components in Omental Adipose Tissue in Obese Men Britt G. Gabrielsson,* Jenny M. Johansson,* Malin Lo¨nn,* Margareta Jernås,* Torsten Olbers,‡ Markku Peltonen,† Ingrid Larsson,† Lars Lo¨nn,† Lars Sjo¨stro¨m,† Bjo¨rn Carlsson,*† and Lena M.S. Carlsson*

Abstract GABRIELSSON, BRITT G., JENNY M. JOHANSSON, ¨ NN, MARGARETA JERNÅS, TORSTEN MALIN LO OLBERS, MARKKU PELTONEN, INGRID LARSSON, ¨ NN, LARS SJO ¨ STRO ¨ M, BJO ¨ RN CARLSSON, LARS LO AND LENA M.S. CARLSSON. High expression of complement components in omental adipose tissue in obese men. Obes Res. 2003;11:699-708. Objective: Accumulation of visceral fat is recognized as a predictor of obesity-related metabolic disturbances. Factors that are predominantly expressed in this depot could mediate the link between visceral obesity and associated diseases. Research Methods and Procedures: Paired subcutaneous and omental adipose tissue biopsies were obtained from 10 obese men. Gene expression was analyzed by DNA microarrays in triplicate and by real-time polymerase chain reaction. Serum C3 and C4 were analyzed by radial immunodiffusion assays in 91 subjects representing a cross section of the general population. Body composition was measured by computerized tomography. Results: Complement components C2, C3, C4, C7, and Factor B had higher expression in omental compared with subcutaneous adipose tissue (⬃2-, 4-, 17-, 10-, and 7-fold, respectively). In addition, adipsin, which belongs to the alternative pathway, and the classical pathway components C1QB, C1R, and C1S were expressed in both depots. Analysis of tissue distribution showed high expression of C2, C3, and C4 in omental adipose tissue, and only liver had higher expression of these genes. Serum C3 levels corre-

Received for review December 2, 2002. Accepted in final form April 15, 2003. *Research Centre for Endocrinology and Metabolism, †Division for Body Composition and Metabolism, and ‡Department of Internal Medicine, Department of Surgery, Sahlgrenska University Hospital, Go¨teborg University, Go¨teborg, Sweden. Address correspondence to Lena Carlsson M.D., Ph.D., Research Centre for Endocrinology and Metabolism, Gro¨na Stråket 8, Sahlgrenska University Hospital, S-413 45 Go¨teborg, Sweden. E-mail: [email protected] Copyright © 2003 NAASO

lated with both visceral and subcutaneous adipose tissue in both men (r ⫽ 0.65 and p ⬍ 0.001 and r ⫽ 0.52 and p ⬍ 0.001, respectively) and women (r ⫽ 0.34 and p ⫽ 0.023 and r ⫽ 0.49 and p ⬍ 0.001, respectively), whereas C4 levels correlated with only visceral fat in men (r ⫽ 0.36, p ⫽ 0.015) and with both depots in women (visceral: r ⫽ 0.58, p ⬍ 0.001; and subcutaneous: r ⫽ 0.51, p ⬍ 0.001). Discussion: Recent studies show that the metabolic syndrome is associated with chronically elevated levels of several immune markers, some of which may have metabolic effects. The high expression of complement genes in intra-abdominal adipose tissue might suggest that the complement system is involved in the development of visceral adiposity and/or contributes to the metabolic complications associated with increased visceral fat mass. Key words: DNA microarray, real-time polymerase chain reaction, body composition, serum C3, serum C4

Introduction Several studies have shown that central obesity, in particular an increase in visceral fat mass, is tightly linked to glucose intolerance, hyperinsulinemia, hypertriglyceridemia, and other features of the metabolic syndrome (1–3). The mechanisms responsible for the association between an increased visceral fat mass and metabolic disorders are not fully understood. However, removal of relatively small amounts of visceral adipose tissue (VAT)1 results in large reductions in obesity-associated metabolic disturbances, indicating a causal relationship (4,5). It has been suggested that release of free fatty acids from the visceral depot could result in an augmentation of hepatic glucose production and very-low-density lipoprotein secretion. There are also depot differences in the responsiveness to various endocrine factors and intrinsic differences in adipocyte metabolism (2,3). Other possibilities include regional differences in adipose

1 Nonstandard abbreviations: VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; CRP, C-reactive protein; TAT, adipose tissue; PCR, polymerase chain reaction.

OBESITY RESEARCH Vol. 11 No. 6 June 2003

699

Complement and Adipose Tissue, Gabrielsson et al.

Table 1. Characteristics of the patients included in the study Obese men

n Age (years) BMI (kg/m2) VAT area (cm2) SAT area (cm2) Insulin (mU/L) Glucose (mM)

General population (men)

General population (women)

Mean ⴞ SD (range)

Mean ⴞ SD (range)

Mean ⴞ SD (range)

Mean ⴞ SD (range)

Mean ⴞ SD (range)

10 44 ⫾ 8 (33, 56) 41.9 ⫾ 4.5 (34.5, 51.2) 280.0 ⫾ 67.8 (140.5, 403.7) 632.6 ⫾ 183.7 (400.2, 957.6) 25 ⫾ 10 (8, 38) –

23 30 ⫾ 1 (27, 32) 24.0 ⫾ 2.5 (21.1, 28.2) 77.2 ⫾ 34.9 (31.9, 167.7) 174.3 ⫾ 76.8 (73.0, 352.1) –

22 60 ⫾ 1 (58, 62) 25.8 ⫾ 2.8 (21.6, 30.8) 139.7 ⫾ 68.2 (51.6, 225.4) 195.9 ⫾ 77.3 (69.9, 341.3) –

22 30 ⫾ 2 (26, 32) 22.2 ⫾ 2.8 (17.6, 30.9) 42.7 ⫾ 14.1 (17.6, 72.1) 190.1 ⫾ 91.4 (43.4, 466.8) –

24 60 ⫾ 2 (57, 62) 25.4 ⫾ 3.2 (20.8, 30.9) 116.0 ⫾ 46.7 (31.9, 233.9) 288.7 ⫾ 88.3 (124.3, 470.4) –

4.2 ⫾ 0.7 (3.0, 5.7)

4.5 ⫾ 0.6 (3.2, 6.4)

3.9 ⫾ 0.4 (3.2, 5.0)

4.5 ⫾ 0.6 (3.2, 5.3)

Paired adipose tissue biopsies obtained from the obese men were used for expression analysis. Serum C3 and C4 were measured in a cross-section of the general population.

tissue gene expression and protein production that may explain the association between visceral obesity and metabolic disease. Previous studies have shown that intra-abdominal adipose tissue releases more interleukin-6 (6) and plasminogen activator inhibitor-1 (7) than subcutaneous adipose tissue (SAT), whereas secretion of leptin is more pronounced from subcutaneous fat cells than from omental fat cells (8). Studies where depot-related gene expression has been evaluated have so far been limited to established candidate genes for obesity and the metabolic syndrome, for example peroxisome proliferator-activated receptorgamma, leptin, insulin receptor, and plasminogen activator inhibitor-1, but most of these genes vary 2-fold or less in expression levels between depots (9 –13). However, it is possible that as yet unidentified genes, which encode for factors involved in the development of metabolic disorders, are produced in the visceral depot. Using DNA microarray analysis of subcutaneous and omental adipose tissue from obese men, we have previously identified depot-related differences in expression of some fibroblast growth factors (14). The present study shows that genes belonging to the complement system were expressed at high levels in omental adipose tissue. The complement system is a part of the innate immune system, and it has been suggested recently that an elevated chronic immune response could be a link between obesity and metabolic disorders (15). Therefore, high expression of complement 700

OBESITY RESEARCH Vol. 11 No. 6 June 2003

genes in VAT may be involved in the development, or contribute to the metabolic consequences, of visceral obesity.

Research Methods and Procedures Subjects and Samples Adipose tissue biopsies were obtained from 10 men (14) undergoing laparoscopic bariatric surgery for weight reduction. Characteristics of these patients are shown in Table 1. The subcutaneous samples were taken in the upper midline of the abdomen, halfway between the xiphoid process and the umbilicus. The intra-abdominal samples were collected from the major omentum in the upper left quadrant of the abdominal cavity in close proximity to the front wall of the stomach and the spleen. Omental adipose tissue and SAT biopsies were excised at the beginning of the operation, frozen directly in liquid nitrogen, and stored at ⫺80 C. Adipocyte and stroma fractions were prepared from subject 4 after the procedure outlined by Smith et al. (16). For determination of serum C3 and C4, samples from 99 subjects in two age groups (26 to 32 years and 57 to 62 years) representing a cross-section of the general population with respect to body composition were analyzed. Characteristics of these subjects are shown in Table 1. None of the subjects had diabetes or history of cardiovascular disease. Five men and three women with high serum C-reactive protein (CRP) levels (CRP ⬎ 5 mg/L) were excluded from the subsequent

Complement and Adipose Tissue, Gabrielsson et al.

correlation analysis. Body composition was determined using computerized tomography at lumbar 4 level as previously described (14). The Medical Ethics Committee at Go¨ teborg University approved the study. All participants gave written informed consent before participating in the study. Hybridization to and Analysis of Human U95A Microarrays RNA isolations from adipose tissue biopsies from six obese men were performed using the Chomczynski method (17) followed by RNeasy clean-up (Qiagen, Hilden, Germany) before being reverse transcribed into cDNA, as previously described (14). In addition, RNA from adipocytes and stroma fractions from one subject (age ⫽ 37 years, BMI ⫽ 41.0 kg/m2) (14) was prepared. Synthesis of biotinlabeled cRNA, hybridization to DNA microarrays (Human Genome U95A array, Affymetrix, Santa Clara, CA) and detection of hybridized target cRNA were performed according to the Affymetrix Gene Chip Expression Analysis manual. Quality of the cDNA synthesis and in vitro transcription was assessed by hybridization to Test2-arrays (Affymetrix). Target cRNA was prepared from three separate RNA preparations from each patient and depot and hybridized to three Human Genome U95A arrays. RNA samples from the adipocyte and stroma fractions were hybridized to duplicate microarrays. To allow cross comparisons among different samples, the mean target signal on each microarray was scaled to an average intensity of 500. Affymetrix software Micro Array Suite 4.0 and Data Mining Tool 2.0 were used for the analysis of differences in gene expression between subcutaneous and omental adipose tissue. Analysis of microarray data and selection of genes with different expression levels in the two depots were performed as described previously (14).

ferent human tissues. Predeveloped assay reagents for betaactin were obtained from Applied Biosystems and used as reference to normalize the expression levels among the samples. All standards and samples were analyzed in triplicate. Biochemical Analysis Serum samples were taken after overnight fast. Serum insulin was analyzed by radioimmunoassay (Pharmacia Diagnostics AB, Uppsala, Sweden) using the WHO-standard 66/304 (reference ⬍ 20 mU/L). Serum CRP was analyzed using the Tina-Quant CRP assay (reference ⬍ 5 mg/L, Roche Diagnostics, Basel, Switzerland). C3 and C4 were analyzed by radial immunodiffusion assays (reference intervals 0.76 to 1.39 g/L and 0.13 to 0.37 g/L, respectively). The measurements were performed at the laboratories of Clinical Chemistry (insulin and CRP) and Clinical Immunology (C3 and C4) at Sahlgrenska University Hospital, both accredited according to the European norm (EN 17025). Statistical Analysis All values are presented as mean ⫾ SD. Difference in gene expression between depots was analyzed by the Wilcoxon signed rank test. When the relative expression for a gene was less than 2-fold in difference between the depots of a subject, the values were arbitrarily set to one in each depot because of the minimum requirement of 2-fold change of the real-time PCR assay. Relationships between body composition and serum C3 and C4 were analyzed with multiple regression models. All analyses were stratified by sex and adjusted for age. The results are presented as univariate correlation coefficients.

Results Real-Time Polymerase Chain Reaction (PCR) Analysis of Gene Expression Oligonucleotide probes spanning exon-intron boundaries and primers (probe and primer sequences are available on request) for analysis of the complement components C2, C3, C4, and C7 were designed with the Primer Express 1.5 software (Applied Biosystems, Foster City, CA) and were purchased from Applied Biosystems. The probes were labeled at the 5⬘ end with the reporter dye 5-carboxyfluorescein and at the 3⬘ end with the quencher N,N,⬘N⬘-tetramethyl-6-carboxyrhodamine. Reagents (TaqMan Reverse Transcriptase reagents and TaqMan Universal PCR Master mix, Applied Biosystems) and conditions were used according to the manufacturer’s protocol as described previously (14). Paired biopsies from nine obese men were used for real-time PCR analysis. For the study of tissue distribution, MTC Panel I and II cDNAs (Clontech, Palo Alto, CA) were used for screening of complement gene expression in dif-

Using DNA microarray analysis, 28 genes with higher expression levels in omental adipose tissue compared with SAT in the majority of the six subjects were identified. Classification of these genes according to their putative cell/organism function revealed that 25% were immunerelated (Table 2). These included three complement genes, Factor B, C4, and C7 (Table 2), each representing different parts of the complement system: the alternative pathway, the classical pathway, and the terminal complex, respectively. This prompted us to investigate the expression pattern of other factors and components of the complement system. Expression of Factors of the Alternative Pathway of the Complement System Comparison of the expression profiles of the two depots showed that Factor B was detected at higher levels in omental adipose tissue compared with SAT in five of the six OBESITY RESEARCH Vol. 11 No. 6 June 2003

701

Complement and Adipose Tissue, Gabrielsson et al.

Table 2. Genes with higher expression in omental adipose tissue compared with SAT identified by DNA microarray analysis

Gene name

Gene description

GeneBank Subcellular accession no. localization

Cell/organism defense BF Factor B L15702 C4 Complement C4 U24578 C7 Complement C7 J03507 IGHG3 Immunoglobulin heavy constant gamma 3 Y14737 IGKC Immunoglobulin kappa C, light chain M63438 IGL@ Immunoglobulin lambda, light chain M18645 IGL@ Immunoglobulin lambda, light chain X57809 SLPI Secretory leukocyte protease inhibitor X04470 Growth SEMA3B Semaphorin 3B U73167 GAS1 GAS1 TIMP1 Metabolism ALDH1 PTGIS Signal transduction MSLN ANXA8 PTN RARRES1 RARRES1 Structure KRT18 KRT19 LAMB1 VIL2 Transcription NR2F1 NR2F1 WT1 Unknown TM4SF1 TM4SF1

702

Secreted Secreted Secreted Secreted Secreted Secreted Secreted Secreted

Relative gene expression (mean ⴞ SD) 7.7 ⫾ 2.2 16.6 ⫾ 7.4 10.8 ⫾ 2.4 11.4 ⫾ 6.7 3.4 ⫾ 0.6 4.0 ⫾ 1.2 3.9 ⫾ 1.3 8.0 ⫾ 2.2

Membrane, endoplasmic reticulum Membrane Membrane Extracellular matrix

4.8 ⫾ 1.7

4.6 ⫾ 1.8 3.2 ⫾ 1.4

5.4 ⫾ 3.6 4.9 ⫾ 3.0 5.6 ⫾ 2.2

Growth arrest specific 1 Growth arrest specific 1 Tissue inhibitor of metalloprotease 1

L13698 L13698 D11139

Aldehyde dehydrogenase Prostacyclin synthase

K03000 D83402

Cytoplasmic

Mesothelin, CAK1 Annexin 8 Pleiotrophin RAR-responsive TIG1 RAR-responsive TIG1

U40434 X16662 M57399 U27185 AI887421

Membrane Membrane Secreted Membrane Membrane

Keratin 18 Keratin 19 Laminin B1 Ezrin

M26326 Y00503 M61916 X51521

Cytoskeletal Cytoskeletal Cytoskeletal Cytoskeletal

6.9 ⫾ 3.1 6.9 ⫾ 2.2 3.3 ⫾ 0.7 4.2 ⫾ 2.5

COUP-TF COUP-TF Krueppel-like Zn finger protein

HG X16155 X51630

Nuclear Nuclear Nuclear

5.2 ⫾ 2.5 5.2 ⫾ 2.0 4.4 ⫾ 1.5

Tumor antigen L6 Tumor antigen L6 KIAA0954 KIAA1077 Similar to Beta-galactosidase Unknown Unknown

M90657 AI445461 AB023171 AB029000 W28612 AL050214 AL049963

Membrane Membrane Unknown Unknown Unknown Unknown Unknown

3.9 ⫾ 0.7 6.5 ⫾ 1.4 6.5 ⫾ 1.9 6.9 ⫾ 2.2 3.0 ⫾ 1.1 6.4 ⫾ 1.4 4.1 ⫾ 1.4

OBESITY RESEARCH Vol. 11 No. 6 June 2003

8.7 ⫾ 5.6 26.8 ⫾ 18.6 3.7 ⫾ 1.8 9.5 ⫾ 5.5 5.8 ⫾ 2.6

Complement and Adipose Tissue, Gabrielsson et al.

Table 3. Relative expression of components and factors of the complement system in omental adipose tissue compared with SAT Subject Gene The alternative pathway Factor B Adipsin The classical pathway C1QB C1R C1S C2 C4 The terminal complex C5 C7 C8A C8B C8G C9

1

2

3

4

5

6

8.9 ⫾ 1.8 0.7 ⫾ 0.1

8.5 ⫾ 2.7 0.8 ⫾ 0.1

ND 1.4 ⫾ 0.2

10.3 ⫾ 1.4 0.6 ⫾ 0.1

6.4 ⫾ 1.2 0.6 ⫾ 0.3

4.6 ⫾ 1.5 0.5 ⫾ 0.1

2.7 ⫾ 0.7 1.5 ⫾ 0.2 1.1 ⫾ 0.1 3.1 ⫾ 0.2 20.6 ⫾ 8.5

4.2 ⫾ 1.2 1.7 ⫾ 0.3 1.3 ⫾ 0.1 2.0 ⫾ 0.2 26.4 ⫾ 7.1

3.8 1.5 1.5 1.2 7.5

⫾ ⫾ ⫾ ⫾ ⫾

2.4 0.5 0.4 0.2 2.8

5.6 ⫾ 0.2 1.9 ⫾ 0.2 1.5 ⫾ 0.0 3.0 ⫾ 1.4 20.4 ⫾ 4.7

6.2 ⫾ 3.6 2.0 ⫾ 0.2 1.7 ⫾ 0.3 1.8 ⫾ 0.3 15.4 ⫾ 1.7

2.0 ⫾ 0.9 1.6 ⫾ 0.2 1.2 ⫾ 0.3 2.3 ⫾ 0.5 10.3 ⫾ 3.1

ND 13.0 ⫾ 4.3 ND ND ND ND

ND 13.0 ⫾ 3.3 ND ND ND ND

ND 6.1 ⫾ 2.2 ND ND ND ND

ND 10.0 ⫾ 3.0 ND ND ND ND

ND 10.8 ⫾ 3.4 ND ND ND ND

ND 7.1 ⫾ 1.8 ND ND ND ND

Gene expression was analyzed in six obese men by DNA microarrays. Genes that were not detected in either depot in a subject are shown as “ND.” Values represent mean and SD of fold change values from the comparison analysis (see “Research Methods and Procedures”).

subjects (Table 3) and only in the stroma fraction of omental adipose tissue (data not shown). Adipsin/Factor D was expressed at high levels in both subcutaneous and omental adipose tissue (Table 3) and in both the adipocyte and the stroma fractions from both depots (data not shown). Expression of Components of the Classical Pathway C4 was expressed at high levels in all omental adipose tissue samples but not detected in the majority of the SAT samples as analyzed by DNA microarray (Table 3). Furthermore, C4 transcripts were detected only in the omental stroma fraction (data not shown). Other components of the classical pathway, C1R, C1S, C1QB, and C2, were detected in the majority of the adipose tissue samples. C1R, C1S, and C2 were expressed in both cell fractions from both depots, whereas C1QB was detected in only the stroma fractions (data not shown). C1QA, C1QG, and C3 were not represented on the microarray. The classical pathway C3 convertase, C4b2a, is composed of equimolar amounts of C4 and C2. Real-time PCR analysis showed that C2 was expressed in both depots but at higher levels in omental adipose tissue compared with SAT [Figure 1A, C2/beta-actin; 0.6 ⫾ 0.3 and 1.1 ⫾ 0.5 (n ⫽ 9), subcutaneous and omental adipose tissue, respectively, p ⬍ 0.05]. The high expression of C4 in omental adipose tissue

detected by the DNA microarray analysis was confirmed by real-time PCR, which showed on average 17-fold higher C4 expression in omental adipose tissue compared with SAT [Figure 1B, C4/beta-actin; 0.1 ⫾ 0.1 and 1.2 ⫾ 0.6 (n ⫽ 9), subcutaneous and omental adipose tissue, respectively, p ⬍ 0.01]. Because C3 is essential for the activity of either pathway, C3 gene expression in the two adipose tissue depots was investigated using real-time PCR analysis. Figure 1C shows the expression of C3 in paired subcutaneous and omental adipose tissue samples from nine obese men. C3 transcripts were detected in all samples with higher expression levels in omental adipose tissue compared with SAT in all subjects [Figure 1C, C3/beta-actin; 0.4 ⫾ 0.2 and 1.3 ⫾ 0.6 (n ⫽ 9), subcutaneous and omental adipose tissue, respectively, p ⬍ 0.01]. Expression of Terminal Complex Components All components of the terminal complex (Table 3), with the exception of C6, were represented on the microarray, but only C7 expression was detected in adipose tissue. Furthermore, C7 was expressed at higher levels in omental adipose tissue, and this was reflected in higher expression levels in both omental adipocytes and omental stroma fraction compared with corresponding fractions from SAT (data OBESITY RESEARCH Vol. 11 No. 6 June 2003

703

Complement and Adipose Tissue, Gabrielsson et al.

Figure 1: Gene expression of C2 (A), C4 (B), C3 (C), and C7 (D), analyzed by real-time PCR, in subcutaneous (black bars) and omental (white bars) adipose tissue obtained from nine obese men. All four complement components were expressed at higher levels in omental adipose tissue compared with SAT, p ⬍ 0.01. The values were normalized to beta-actin expression in each sample.

not shown). The higher C7 transcript levels in omental fat were confirmed with real-time PCR, resulting in 6-fold (6.3 ⫾ 2.9) higher expression in omental compared with SAT for the nine subjects [Figure 1D, C7/beta-actin; 0.1 ⫾ 0.04 and 0.7 ⫾ 0.2 (n ⫽ 9), subcutaneous and omental adipose tissue, respectively, p ⬍ 0.01]. Tissue Distribution of C2, C3, C4, and C7 Expression Serum C2, C3, and C4 are believed to be derived mainly from the liver (18); therefore, we were also interested in comparing the expression levels of these components in adipose tissue with those in other tissues. In Figure 2, the levels of C2, C3, C4, and C7 transcripts in heart, kidney, leukocytes, liver, lung, skeletal muscle, and small intestine are compared with those in subcutaneous and omental adipose tissue. As expected, liver showed the highest expression of C2, C3, and C4 with ⬃3-, 9-, and 1.5-fold higher levels compared with omental adipose tissue (Figure 2, A–C). C7 has been reported previously to originate from tissues other than the liver (16). Our results showed the highest C7 expression in heart, followed by omental adipose tissue, kidney, small intestine, liver, lung, SAT, and muscle (Figure 2D). 704

OBESITY RESEARCH Vol. 11 No. 6 June 2003

Levels of C3 and C4 in Serum It has been shown previously that fasting serum C3 levels correlate to waist circumference in a group including both men and women (19). To investigate whether gender, age, or different measures of adiposity affected serum C3 levels, serum samples from 45 men and 46 women representing a cross-section of the general population as described in “Research Methods and Procedures” were analyzed. Serum C3 levels in men and women did not differ (1.19 ⫾ 0.25 and 1.22 ⫾ 0.31 g/L, respectively), nor was there any effect by age. Serum C3 levels correlated with visceral, subcutaneous, and total fat areas and BMI in both genders (Table 4) and remained significant after adjustment for age (data not shown). The relatively high expression of C4 in omental adipose tissue compared with both its primary source, the liver, and with SAT led us to investigate the relationship between serum levels of this factor and adiposity. There was no difference in serum C4 levels between men and women (0.25 ⫾ 0.07 and 0.27 ⫾ 0.10 g/L, respectively) and no effect of age. Serum C4 levels correlated with VAT area but not with BMI, subcutaneous, or total adipose tissue (TAT) in men (Table 4). However, the association between serum

Complement and Adipose Tissue, Gabrielsson et al.

Figure 2: C2 (A), C4 (B), C3 (C), and C7 (D) gene expression in heart (H), kidney (K), leukocytes (Leu), liver (Li), lung (Lu), skeletal muscle (M), and small intestine (SmI) compared with expression in SAT (ATsc, black bar) and omental adipose tissue (ATom, white bar) analyzed by real-time PCR. Values for subcutaneous and omental adipose tissue are mean ⫾ SD of respective expression levels shown in Figure 1. Expression was normalized to beta-actin expression in all samples.

Table 4. Serum levels of C3 and C4 related to the areas of VAT, SAT, TAT, and BMI in men (n ⫽ 45) and women (n ⫽ 46) C3 Men VAT r: p: SAT r: p: TAT r: p: BMI r: p:

C4 and VAT was slightly reduced when the analysis was adjusted for age (r ⫽ 0.29, p ⫽ 0.055). In women, both adipose tissue depots and BMI were correlated with serum C4 (Table 4) and remained significant after age adjustment (p ⬍ 0.02 for all correlations with fat areas and p ⬍ 0.05 for correlation with BMI).

C4 Women

Men

Women

0.65 ⬍0.001

0.34 0.023

0.36 0.015

0.58 ⬍0.001

0.52 ⬍0.001

0.49 ⬍0.001

0.15 0.316

0.51 ⬍0.001

0.62 ⬍0.001

0.47 ⬍0.001

0.26 0.083

0.57 ⬍0.001

0.46 0.001

0.56 ⬍0.001

0.07 0.654

0.48 ⬍0.001

Discussion It is well established that adipose tissue produces all factors of the alternative pathway of the complement system (20 –22), and it has been suggested that a fragment of C3 has metabolic effects in human adipose tissue (23,24). However, to our knowledge, this is the first report that has looked systematically at the expression of complement components and factors in adipose tissue. In this study, we show for the first time that genes encoding proteins of the classical pathway were also expressed in human adipose tissue and that there were marked depot differences in the expression of several complement components. Recent studies have shown that several aspects of the metabolic syndrome, which in itself is tightly linked to visceral obesity, are associated with chronically elevated OBESITY RESEARCH Vol. 11 No. 6 June 2003

705

Complement and Adipose Tissue, Gabrielsson et al.

serum levels of acute-phase markers (15,25). In particular, increased levels of the acute-phase proteins CRP and complement C3 can predict risk for future cardiovascular events, and both are negatively correlated with insulin sensitivity in obese subjects (26 –30). Another link between the immune system and adipose tissue is the increase in visceral fat mass that occurs in some diseases associated with chronic inflammation, such as Crohn’s disease, and in human immunodeficiency virus subjects with a natural delayed development of AIDS (31). The complement system is part of the innate immune system, and several of the complement genes are regulated by cytokines, such as interleukin 1, interleukin-6, or tumor necrosis factor alpha or interferon gamma at the transcriptional level (32). Studies in patients undergoing liver transplantation have led to the conclusion that plasma complement proteins are mainly derived from the liver. However, it is likely that plasma C1Q and C7 originate from other tissues (for review, see 18), and the primary source of plasma adipsin/Factor D is suggested to be adipose tissue (18,33). In this study, we show that serum C3 concentrations are positively correlated with both visceral and subcutaneous fat and with BMI in both men and women. Furthermore, the relative expression of C3, related to beta-actin as internal reference gene, in omental and SAT in obese men was ⬃10% and 2.5%, respectively, of that in the liver, which suggests that C3 produced by adipose tissue could contribute to the plasma pool of C3. This is also supported by the findings that in anorectic women, fasting serum levels of C3 are low and normalize with weight gain, whereas in obese women, serum levels of C3 are higher than in controls and are decreased after weight loss (34). In addition, others have shown that fasting serum C3 levels are positively correlated with waist circumference (19) and negatively associated with insulin sensitivity in obese subjects (28,29). In adipose tissue, a proteolytic fragment of C3, acylation-stimulating protein, acts as a paracrine metabolic factor in that it stimulates glucose uptake and triacylglycerol synthesis in human adipocytes (23,24). In addition, both serum C3 levels and acylation-stimulating protein production in vivo by SAT are increased postprandially (19,35). In this study, the C3 mRNA levels were on average 4-fold higher, whereas a previous study reported 2-fold higher expression of C3 in omental adipose tissue compared with SAT (12). In contrast to factors of the alternative pathway, the expression of components of the classical pathway and terminal complex in adipose tissue is less studied. We have previously reported expression of C2 in human adipose tissue (36), and we now show that C1R, C1S, C1QB, C2, C4, and C7 were expressed in all omental adipose tissue samples and also in the majority of the subcutaneous samples. Interestingly, changes in serum levels of the classical pathway components C1Q, C2, and C4 are correlated with change in body weight in both anorectic and obese women 706

OBESITY RESEARCH Vol. 11 No. 6 June 2003

(34). There is no reference in the literature to C4 expression in adipose tissue, although there are several reports of extrahepatic C4 synthesis in monocytes, macrophages, fibroblasts, and epithelial cells (for review, see 18). We found that the difference in C4 expression in subcutaneous and omental adipose tissue was the most striking of all complement genes. This was also reflected in the finding that the relative C4 gene expression in omental adipose tissue was more than one-half of that detected in liver. In the group representing a cross section of the population, serum C4 levels in both genders correlated with visceral fat area, which is in agreement with the detection of high levels of C4 transcripts in omental adipose tissue. This correlation was reduced when adjusted for age, and we also observed a gender difference in that serum C4 also correlated with subcutaneous fat area and BMI in women but not in men. In this context, it has to be recognized that both age and sex also have a strong influence on metabolic control (37). The tissue distribution of C7 expression was different from the other complement genes that were studied. Omental adipose tissue, together with heart and kidney, showed higher expression of C7 compared with liver. C7 was recently reported to be expressed at lower levels in omental adipose tissue from diabetic obese patients compared with nondiabetic obese subjects (38). Adipose tissue consists of a heterogeneous cell population, and little is known about possible differences in cell composition of different depots. Some of the complement genes, and in particular those with striking depot differences in expression (C4 and Factor B), were detected only in the stroma fraction of adipose tissue. It has been shown that mesothelial cells are present in omental but not subcutaneous fat (39), and we have observed larger number of plasma cells in omental adipose tissue compared with SAT (E. Jennische, M. Lo¨ nn, J.M. Johansson, unpublished data). Such intrinsic differences could affect the microenvironment of the adipocytes of the different fat depots by paracrine mechanisms. During recent years, it has become clear that markers of inflammation are associated with type 2 diabetes and cardiovascular disease. This has led to the suggestion that chronic activation of the innate immune system could cause the metabolic syndrome (17,18). The data presented here show that adipose tissue, and in particular the visceral depot, expresses several genes encoding components of the innate immune system. This suggests that adipose tissue itself may be involved in the proposed chronic activation of this relatively nonspecific defense system.

Acknowledgments The Swedish Medical Research Council (11295, 11502, 13141, and 13507), the Swedish Society for Medical Research, IngaBritt and Arne Lundberg Forskningsstiftelse, and The National Board of Health and Welfare supported this work.

Complement and Adipose Tissue, Gabrielsson et al.

References 1. Kissebah AH. Intra-abdominal fat: is it a major factor in developing diabetes and coronary artery disease? Diabetes Res Clin Pract. 1996;30(suppl):25–30. 2. Bjo¨rntorp P. Metabolic implications of body fat distribution. Diabetes Care. 1991;14:1132– 43. 3. Montague CT, O’Rahilly S. The perils of portliness: causes and consequences of visceral adiposity. Diabetes. 2000;49: 883– 8. 4. Gabriely I, Ma XH, Yang XM, et al. Removal of visceral fat prevents insulin resistance and glucose intolerance of aging. Diabetes.2002;51:2951– 8. 5. Thorne A, Lonnqvist F, Apelman J, Hellers G, Arner P. A pilot study of long-term effects of a novel obesity treatment: omentectomy in connection with adjustable gastric banding. Int J Obes Relat Metab Disord. 2002;26:193–9. 6. Fried SK, Bunkin DA, Greenberg AS. Omental and subcutaneous adipose tissues of obese subjects release interleukin-6: depot difference and regulation by glucocorticoid. J Clin Endocrinol Metab. 1998;83:847–50. 7. Alessi MC, Peiretti F, Morange P, Henry M, Nalbone G, Juhan-Vague I. Production of plasminogen activator inhibitor 1 by human adipose tissue: possible link between visceral fat accumulation and vascular disease. Diabetes. 1997;46: 860 –7. 8. Van Harmelen V, Reynisdottir S, Eriksson P, et al. Leptin secretion from subcutaneous and visceral adipose tissue in women. Diabetes. 1998;47:913–917. 9. Lefebvre AM, Laville M, Vega N, et al. Depot-specific differences in adipose tissue gene expression in lean and obese subjects. Diabetes. 1998;47:98 –103. 10. Imbeault P, Vidal H, Tremblay A, et al. Age-related differences in messenger ribonucleic acid expression of key proteins involved in adipose cell differentiation and metabolism. J Clin Endocrinol Metab. 2001;86:828 –33. 11. Montague CT, Prins JB, Sanders L et al. Depot-related gene expression in human subcutaneous and omental adipocytes. Diabetes. 1998;47:1384 –91. 12. Dusserre E, Moulin P, Vidal H. Differences in mRNA expression of the proteins secreted by the adipocytes in human subcutaneous and visceral adipose tissues. Biochim Biophys Acta. 2000;1500:88 –96. 13. Eriksson P, Van Harmelen V, Hoffstedt J, et al. Regional variation in plasminogen activator inhibitor-1 expression in adipose tissue from obese individuals. Thromb Haemost. 2000;83:545– 8. 14. Gabrielsson BG, Johansson JM, Jennische E, et al. Depotspecific expression of fibroblast growth factors in human adipose tissue. Obes Res. 2002;10:608 –16. 15. Yudkin JS, Stehouwer CDA, Emeis JJ, Coppack SW. C-reactive protein in healthy subjects: associations with obesity, insulin resistance, and endothelial dysfunction: a potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol. 1999;19:972– 8. 16. Smith U, Sjo¨stro¨m L, Bjo¨rntorp P. Comparison of two methods for determining human adipose cell size. J Lipid Res. 1972;13:822– 4.

17. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987;162:156 –9. 18. Morgan BP, Gasque P. Extrahepatic complement biosynthesis: where, when and why? Clin Exp Immunol. 1997;107:1–7. 19. Halkes CJM, van Dijk H, de Jaegere PPT, et al. Postprandial increase of complement component 3 in normolipidemic patients with coronary artery disease: effects of expandeddose simvastatin. Arterioscler Thromb Vasc Biol. 2001;21: 1526 –30. 20. Choy LN, Rosen BS, Spiegelman BM. Adipsin and an endogenous pathway of complement from adipose cells. J Biol Chem. 1992;267:12736 – 41. 21. Cianflone K, Maslowska M. Differentiation-induced production of ASP in human adipocytes. Eur J Clin Invest. 1995;25: 817–25. 22. Peake PW, O’Grady S, Pussell BA, Charlesworth JA. Detection and quantification of the control proteins of the alternative pathway of complement in 3T3–L1 adipocytes. Eur J Clin Invest. 1997;27:922–7. 23. Cianflone K, Roncari DA, Maslowska M, Baldo A, Forden J, Sniderman AD. Adipsin/acylation stimulating protein system in human adipocytes: regulation of triacylglycerol synthesis. Biochemistry. 1994;33:9489 –95. 24. Maslowska M, Sniderman AD, Germinario R, Cianflone K. ASP stimulates glucose transport in cultured human adipocytes. Int J Obes Relat Metab Disord. 1997;21:261– 6. 25. Pickup JC, Crook MA. Is type II diabetes mellitus a disease of the innate immune system? Diabetologia. 1998;41:1241– 8. 26. Muscari A, Massarelli G, Bastagli L, et al. Relationship of serum C3 to fasting insulin, risk factors and previous ischaemic events in middle-aged men. Eur Heart J. 2000;21: 1081–90. 27. Lemieux I, Pascot A, Prud’homme D, et al. Elevated C-reactive protein: another component of the atherothrombotic profile of abdominal obesity. Arterioscler Thromb Vasc Biol 2001;21:961–967. 28. Weyer C, Tataranni PA, Pratley RE. Insulin action and insulinemia are closely related to the fasting complement C3, but not acylation stimulating protein concentration. Diabetes Care. 2000;23:779 – 85. 29. Koistinen HA, Koivisto VA, Ebeling P. Serum complement protein C3 concentration is elevated in insulin resistance in obese men. Eur J Int Med. 2000;11:21– 6. 30. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR. Comparison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events. N Engl J Med. 2002;347:1557– 65. 31. Pond CM. Long-term changes in adipose tissue in human disease. Proc Nutr Soc. 2001;60:365–74. 32. Volanakis JE. Transcriptional regulation of complement genes. Ann Rev Immunol. 1995;13:277–305. 33. White RT, Damm D, Hancock N, et al. Human adipsin is identical to complement factor D and is expressed at high levels in adipose tissue. J Biol Chem. 1992;267:9210 –3. 34. Pomeroy C, Mitchell J, Eckert E, Raymond N, Crosby R, Dalmasso AP. Effect of body weight and caloric restriction on OBESITY RESEARCH Vol. 11 No. 6 June 2003

707

Complement and Adipose Tissue, Gabrielsson et al.

serum complement proteins, including Factor D/adipsin: studies in anorexia nervosa and obesity. Clin Exp Immunol. 1997; 108:507–15. 35. Saleh J, Summers LK, Cianflone K, Fielding BA, Sniderman AD, Frayn KN. Coordinated release of acylation stimulating protein (ASP) and triacylglycerol clearance by human adipose tissue in vivo in the postprandial period. J Lipid Res. 1998;39:884 –91. 36. Gabrielsson BG, Carlsson B, Carlsson LMS. Partial genome scale analysis of gene expression in human adipose tissue using DNA array. Obes Res. 2000;8:374 –384.

708

OBESITY RESEARCH Vol. 11 No. 6 June 2003

37. Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults: findings from the third National Health and Nutrition Examination Survey. JAMA. 2002;287:356–9. 38. Corominola H, Conner LJ, Beavers LS, et al. Identification of novel genes differentially expressed in omental fat of obese subjects and obese type 2 diabetic patients. Diabetes. 2001; 50:2822–30. 39. van Hinsbergh VWH, Kooistra T, Scheffer MA, van Bockel JH, van Muijen GNP. Characterization and fibrinolytic properties of human omental tissue mesothelial cells: comparison with endothelial cells. Blood. 1990;75:1490 –7.