| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine (P.A., B.M.S., E.D., J.H., M.R., I.D.), Karolinska Institutet, Huddinge, Stockholm, S-14186, Sweden; and Division of Surgery (E.N.), Department of Clinical Sciences, Karolinska Institutet, Danderyd Hospital Stockholm, S-18288, Sweden
Address all correspondence and requests for reprints to: Peter Arner, M.D., Ph.D., Professor, Karolinska Institutet, Karolinska University Hospital-Huddinge, Department of Medicine, M61, SE-141 86 Stockholm, Sweden. E-mail: peter.arner{at}ki.se.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of the study was to elucidate the role of adipose STAMP2 expression in human obesity and insulin resistance.
Design: The design was to quantify STAMP2 in human abdominal sc and omental white adipose tissue (WAT), isolated adipocytes, and stroma and in vitro differentiated preadipocytes and relate levels of STAMP2 in sc WAT to clinical and adipocyte phenotypes involved in insulin resistance.
Participants: Nonobese and obese women and men (n = 236) recruited from an obesity clinic or through local advertisement.
Main Outcome Measurement: Clinical measures included body mass index, body fat, total adiponectin, and homeostasis model assessment as measure of overall insulin resistance. In adipocytes we determined cell size, sensitivity of lipolysis and lipogenesis to insulin, adiponectin secretion, and inflammatory gene expression.
Results: STAMP2 levels in sc and visceral WAT and adipocytes were increased in obesity (P = 0.0008–0.05) but not influenced by weight loss. Increased WAT STAMP2 levels associated with a high amount of body fat (P = 0.04), high homeostasis model assessment (P = 0.01), and large adipocytes (P = 0.02). Subjects with high STAMP2 levels displayed reduced sensitivity of adipocyte lipogenesis (P = 0.04) and lipolysis (P = 0.03) to insulin but had normal adiponectin levels. WAT STAMP2 levels correlated with expression of the macrophage marker CD68 (P = 0.0006).
Conclusion: Human WAT STAMP2 associates with obesity and insulin resistance independently of adiponectin, but the role of STAMP2 in obesity and its complications seems different from that in mice.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Recently six transmembrane protein of prostate 2 (STAMP2; also called TIARP) was reported to be a counterregulator of inflammation and insulin resistance in mice (2). In mice, STAMP2 expression in WAT is induced by obesity and inflammatory cytokines. STAMP2–/– mice display elevated expression of proinflammatory mediators in WAT and impairment of insulin-stimulated glucose transport in adipocytes. The expression of adiponectin, a regulator of insulin sensitivity, is reduced. STAMP2–/– mice accumulate sc WAT and display an insulin-resistant phenotype with elevated plasma glucose, insulin, and lipids. Previously STAMP2 mRNA has been shown to be induced in 3T3-L1 preadipocytes during differentiation and by treatment with TNF
(TNFA) (3). Together these results suggest that STAMP2 expression may be induced as a protective antiinflammatory factor and that enhancing STAMP2 signaling potentially could be a cure for human insulin resistance without the side effect of weight gain.
With the aim of elucidating the role of STAMP2 in human obesity and insulin resistance, we here quantified the expression of STAMP2 in human WAT and isolated adipocytes and related the expression to clinical and adipocyte phenotypes associated with insulin resistance and inflammation.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All subjects were recruited by local advertisement or from an outpatient clinic for treatment of obesity for the purpose of studying genes regulating obesity and fat cell function. Obesity was defined as body mass index (BMI) of 30 kg/m2 or greater. sc WAT STAMP2 and other phenotypes were investigated in three cohorts. Cohort 1 comprised 15 nonobese [aged 40 ± 9 yr, BMI 23 ± 1 kg/m2, homeostasis model assessment (HOMA) 1.2 + 0.5] and 81 obese (aged 38 ± 9 yr, BMI 39 ± 5 kg/m2, HOMA 3.5 + 2.0) women investigated for body composition, overall insulin resistance, and adipocyte lipolysis and lipogenesis phenotypes. Cohort 2 comprised seven nonobese (five women and two men, aged 33 ± 10 yr, BMI 23 ± 2 kg/m2) and seven obese (six women and one man, aged 48 ± 12 yr, BMI 34 ± 6 kg/m2) subjects for whom mRNA from both intact sc WAT pieces and isolated adipocytes were available. Cohort 3 comprised 13 obese subjects (10 women and three men, aged 39 ± 9 yr, BMI 40 ± 1 kg/m2) investigated before and 2–4 yr after intense antiobesity therapy with antiobesity surgery or behavioral modification when they had reached a nonobese weight-stable state; these subjects have been described before (4).
Visceral, i.e. omental, WAT was investigated in cohort 4, which comprised 24 nonobese (22 women and two men, aged 43 ± 11 yr, BMI 23 ± 2 kg/m2) and 69 obese (39 women and 30 men, aged 39 ± 10 yr, BMI 42 ± 6 kg/m2). Cohort 5 comprised 20 obese subjects (18 women and two men, aged 40 ± 10 yr, BMI 45 ± 4 kg/m2) for whom paired samples of sc and omental WAT samples were available. In cohorts 4 and 5, the nonobese subjects were operated for uncomplicated gallstone disease and the obese with antiobesity surgery. Subcutaneous adipose tissue from the surgical incision and omental adipose tissue were obtained at the beginning of surgery. These patients had been fasting overnight, and only saline was given as iv infusion until adipose tissue was removed.
Pancreas, liver, and muscle samples were obtained from normal tissue in patients without weight loss undergoing elective surgery for malignant disorders.
The study was approved by the ethics committee in Stockholm, and informed consent was obtained from all participants.
Clinical evaluation
Participants were investigated at 0800 h after an overnight fast. Biopsies of the sc abdominal WAT (0.5–2 g) were obtained by needle aspiration under local anesthesia. WAT samples were brought to the laboratory in saline, and one part was used immediately for adipocyte experiments and another was frozen in liquid nitrogen. Venous blood samples were taken for measurements of insulin and glucose, and HOMA was calculated from these values (5). A high HOMA reflects insulin resistance (5). In addition, total plasma adiponectin was measured as described (6). Body fat and lean body mass were quantified by bioimpedance with the Bodystat equipment (Bodystat, Douglas, UK).
Adipocyte experiments
One piece of WAT (300 mg) was incubated exactly as described and release of total adiponectin to medium was determined as described (6).
From WAT samples we isolated the fat cell fraction according to the collagenase procedure as described (7). The fat cells were either frozen in liquid nitrogen for later mRNA isolation or immediately used for lipolysis and lipogenesis investigations conducted as described (4). Mean fat cell weight and volume were determined. Briefly, in the lipolysis experiments, diluted cell suspensions [2% (vol/vol)] were incubated in duplicate for 2 h at 37 C with air as the gas phase in Krebs-Ringer phosphate buffer (pH 7.4) supplemented with glucose (8.6 mmol/liter), ascorbic acid (0.1 mg/ml), and BSA (20 mg/ml) without (basal) or with increasing concentrations (10–16 to 10–4 mol/liter) of either noradrenaline or insulin, and glycerol release (lipolysis index) was determined. In the lipolysis experiments with insulin, the standard medium was supplemented with 8-bromoadenosine-cAMP (10–3 mol/liter) and adenosine deaminase (1 mU/liter). In the lipogenesis experiments, the standard medium was supplemented with a low concentration of unlabeled and tritiated glucose (10–6 mol/liter) and insulin (0, 10–16 to 10–6 mol/liter), and radioactive incorporation into total lipids was determined. At these low glucose concentrations, lipogenesis closely mirrors glucose transport. Results were expressed as amount of glycerol release or amount of glucose incorporated into lipids per 2 h and 107 fat cells. Hormone responsiveness was calculated as lipolysis or lipogenesis at maximum effective hormone concentration. In both lipolysis and lipogenesis experiments, half-maximum effective hormone concentration (EC50) was determined and turned into a pD2 (–log10 EC50), which reflects hormone sensitivity.
In some experiments we saved the nonadipose cells harvested from the stromal fraction of adipose tissue that was obtained after collagenase treatment. This fraction was either frozen for subsequent mRNA preparation or used for in vitro differentiation of preadipocytes under serum-free conditions exactly as described (7). The preadipocytes were seeded out in 12-well plates at a density of 50,000 cells/cm2 and incubated for 48 h with TNFA (100 ng/ml; Sigma-Aldrich, St. Louis, MO) before RNA isolation. Control cells were incubated with medium alone. Previous studies have shown that the maximum effects of TNFA on intracellular signaling and lipolysis in this type of fat cell preparation are obtained with 100 ng/ml of the cytokine (8).
RNA preparation and quantification
Adipose tissue pieces (300 mg), 200 µl isolated adipocytes, or stromal cells were kept at –70 C for subsequent RNA extraction using the RNeasy minikit (QIAGEN, Hilden, Germany). RNA samples were treated with RNase-free DNase (QIAGEN). The RNA concentration was determined using a spectrophotometer. High-quality RNA was confirmed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA).
One microgram of RNA was reverse transcribed using the Omniscript reverse transcription kit (QIAGEN) and random hexamer primers. Real-time quantitative was performed in an iCycler IQ (Bio-Rad Laboratories, Hercules, CA) using the SYBR Green-based technology. Primer pairs were designed to span exon-intron boundaries and to be specific for the gene to be amplified according to BLAST searches. The following primers were used for specific mRNA quantification. STAMP2: 5'-CAGAGTACCTTGCTCATTTGGT-3' and 5'-TGTCATTTCCACACACAAACAC-3', and 18S: 5'-CACATGGCCTCCAAGGAGTAAG-3' and 5'-CCAGCAGTGAGGGTCTCTCT-3'. Dissociation curve analyses and agarose gel electrophoresis were used to validate that a single amplicon was amplified. CD68 (Hs00154355_m1), TNFA (Hs00174128_m1), and IL-6 (Hs00174131_m1) were amplified with Taqman assays (Applied Biosystems, Foster City, CA) according to the manufacturer’s instruction. A direct comparative method was used for data analysis (Applied Biosystems).
Protein expression of STAMP2
Protein analyses were performed on samples from two cohorts. One group consisted of 10 obese and nine nonobese women who were carefully matched for age. The other group consisted of nine obese subjects for whom paired samples from sc and omental WAT was available. These subjects overlapped with the cohorts used for mRNA measurements that were described above.
For protein isolation, approximately 300 mg of frozen WAT were crushed and lysed in protein lysis buffer [1% Triton X-100, Tris-HCL (pH 7.6), and 150 mmol/liter NaCl, 4 C], supplemented with protease inhibitors (1 mmol/liter phenylmethylsulfonyl fluoride) and Complete (Roche Molecular Biochemicals, Mannheim, Germany), and homogenized using a microtome. The homogenate was centrifuged at 14,000 rpm for 30 min, and the infranatant was collected and saved. Protein content was assayed using BCA protein assay reagent kit (Pierce, Rockford, IL) as described above. One hundred micrograms of total protein were loaded on polyacrylamide gels and separated by standard 12% SDS-PAGE. Gels were transferred to polyvinylidine fluoride membranes (GE Healthcare, Little Chaffore, UK). Blots were blocked for 1 h at room temperature in Tris-buffered saline with 0.1% Tween 20 and 5% nonfat dried milk. This was followed by an overnight incubation at 4 C in the presence of antibodies directed against STAMP2 (catalog no. 11944–1-AP; Proteintech Group Inc., Chicago, IL). This antibody is developed against human STAMP2 and the specificity has previously been tested on human liver samples. Antibodies against the control protein β-actin were from Sigma-Aldrich. Secondary -rabbit antibodies conjugated to horseradish peroxidase were from Sigma-Aldrich. Antigen-antibody complexes were detected by chemiluminescence using a kit of reagents form Pierce (Supersignal; Rockford, IL), and specific bands were detected using a Chemidoc XRS system (Bio-Rad, Munich, Germany). Images were analyzed using the Quantity One Software supplied by the manufacturer (Bio-Rad). Expression levels were expressed in relative units for subsequent statistical analysis. The human STAMP2 antibody detected a single band at 50–55 kDa, which is the predicted size of human STAMP2 (see, for instance, http://www.genecards.org/cgi-bin/carddisp.pl?gene=STEAP4). β-Actin expression did not differ between groups.
Statistical analysis
Unpaired or paired t test (two sided) was used to compare mRNA and protein levels between two groups. For analysis of the relationship between STAMP mRNA levels and quantitative variables, cohort 1 was divided into tertiles based on STAMP2 levels. Next, analysis of covariance with age and BMI as covariates were used to evaluate differences in quantitative phenotypes between the tertiles. Analyses with or without the covariates age and BMI gave in each analysis almost identical P values. HOMA was log10 transformed before analysis to become normally distributed. Values are mean ± SD. Correlation was evaluated by z test.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
WAT STAMP2 mRNA levels were increased in human obesity (Fig. 1A
). Increased STAMP2 levels in obese compared with nonobese subjects were observed in abdominal sc WAT (cohort 1, P = 0.0008) and isolated adipocytes (cohort 2, P = 0.05) as well as omental WAT (cohort 4, P = 0.0004) (Fig. 1A). The STAMP2 mRNA levels were about 33% increased in the obese subjects. By contrast, adipose tissue STAMP2 mRNA levels were not affected by marked long-term weight loss to a nonobese weight-stable state [STAMP2 fold change after vs. before 0.98 + 0.54 arbitrary units (AU), P = ns, figure not shown]. Furthermore, there was no difference in STAMP2 between sexes, and age did not influence STAMP2 expression (results not shown).
|
STAMP2 expression is enriched in adipose tissue and detected in adipocytes
STAMP2 mRNA levels were almost 2-fold higher in sc compared with omental WAT of the same subject (P = 0.000008) (Fig. 1B
). In the sc WAT, higher STAMP2 mRNA levels were detected in adipose tissue pieces, compared with isolated adipocytes (P = 10–7) (Fig. 1C). We did not have isolated stroma cells from these samples. However, adipose stroma cells were isolated from four other subjects, in which STAMP2 was quantified to 2.3 ± 1.4 AU. In studies not shown in the figure, STAMP2 levels decreased during in vitro differentiation of preadipocytes (n = 11), measuring d 4, 10.1 + 3.9 AU, d 8, 3.7 + 1.7 AU (d 4 vs. d 8, P = 510–5), and d 12, 3.3 + 0.9 AU (d 4 vs. d 12, P = 0.0003).
STAMP2 was detected in one of three human pancreas samples but in none of three muscle samples and none of two liver samples (values not shown).
STAMP2 protein levels reflect mRNA expression
STAMP2 protein was detected by Western blot using a polyclonal antibody (Fig. 2). One clear band was detected in each sample. Densitometric scanning showed a significantly increased STAMP2 expression in obese subjects (Fig. 2A, upper panel, P = 0.032). β-Actin was used as a loading control, but there was no significant difference in expression levels between obese and nonobese subjects (Fig. 2A, lower panel). STAMP2 expression was also assessed in paired samples of sc and omental WAT from nine obese subjects (Fig. 2B). Paralleling the findings at the mRNA level, we observed a significantly lower STAMP2 expression in omental WAT (Fig. 2B upper panel) but no difference in β-actin expression (Fig. 2B, lower panel).
|
We next investigated the relationship between adiposity and STAMP2 mRNA levels in abdominal sc WAT in more detail. For this purpose we used cohort 1, comprising 96 women with a wide variation in BMI. Increased STAMP2 levels associated with a high amount of body fat (P = 0.04) (Fig. 3A) but not lean body mass (Fig. 3B). There was also a difference in overall in vivo insulin resistance, measured as HOMA, between tertiles with different STAMP2 levels (P = 0.01) (Fig. 3C), which was independent of BMI and age. Low STAMP2 levels associated with low HOMA-levels. The tertile with middle STAMP2 levels displayed marginally higher HOMA-levels than the tertile with the highest STAMP2 levels, but this difference was not significant in post hoc analysis. Finally, high STAMP2 levels associated with large adipocytes (P = 0.02) (Fig. 3D).
|
What could be the function of STAMP2 in human abdominal sc WAT? To get closer to the role STAMP2 may have in WAT, we next investigated how mRNA levels associated with various adipocyte phenotypes (Fig. 4). High STAMP2 levels in sc abdominal adipose tissue associated with adipocyte insulin resistance. Subjects with high STAMP2 levels displayed reduced sensitivity, i.e. pD2, of lipogenesis (P = 0.04) and lipolysis (P = 0.03) to insulin (Fig. 4, A and B). However, there was no association between STAMP2 and maximum stimulation of lipogenesis, or maximum inhibition of lipolysis, by insulin (values not shown). Nor was there any association between STAMP2 levels and basal lipolysis or lipogenesis or catecholamine-induced lipolysis (values not shown). Adipose STAMP2 levels were unrelated to adipose tissue secretion of total adiponectin (Fig. 4C) as well as to circulating concentration of total adiponectin (Fig. 4D).
|
). We first confirmed that obesity associates with a proinflammatory state in WAT, with obese subjects displaying increased mRNA levels of the macrophage cell surface antigen CD68 and TNFA in sc WAT (Fig. 5). There was also a trend toward increased levels of IL-6 mRNA in obesity, although this increase was not significantly possible due to a large interindividual variation in IL-6 levels. Adipose tissue STAMP2 strongly correlated with mRNA levels of CD68 (P = 0.0006) but not with levels of TNFA or IL-6 (Table 1). Inflammatory gene expression was not dependent on age (values not shown).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Adipose STAMP2 expression is clearly increased in obesity. This was observed at the mRNA and protein levels, in different adipose tissue regions (sc and visceral; the former region displaying higher STAMP2 levels than the latter), and in different tissue compartments (intact tissue and isolated fat cells). By contrast, STAMP2 levels were not affected by weight loss. The resistance to slimming was observed in subjects undergoing marked weight reduction from the obese to the lean stage. The findings with weight loss were obtained with rather few subjects. Therefore, we cannot exclude some minor effects on STAMP2, which could not be detected because of lack of power.
In addition, we confirmed results from a mouse fat cell line that TNFA induces STAMP2 expression in adipocytes. Further examining the pattern of STAMP2 expression, we observed that STAMP2 levels were enriched in adipose tissue, compared with other investigated organs. Furthermore, STAMP2 was expressed at higher levels in the intact adipose tissue than in isolated fat cells. A likely explanation for this is that other cells than fat cells are the major contributor to total STAMP2 levels in the adipose tissue. Unfortunately, we did not investigate all the various cell types in the adipose tissue. Clearly, however, STAMP2 is present in normal fat cells because it was easily detected in isolated fat cells. Although STAMP2 was down-regulated during preadipocyte differentiation, STAMP2 mRNA was clearly detected in the differentiated adipocytes.
In mice it has been hypothesized that WAT STAMP2 is increased in obesity to counteract the dangerous metabolic consequences of obesity. This belief is based on the finding that STAMP2–/– mouse display an insulin-resistant phenotype. Our results, i.e. positive association between human adipose tissue STAMP2 levels and adipocyte and systemic insulin resistance independently of age and BMI, do not directly support a protective role of STAMP2 in human obesity.
By contrast, the positive associations between adipose tissue STAMP2 levels, obesity, and insulin resistance are more in favor with the hypothesis that human STAMP2 contributes to insulin resistance. A primary causative role for STAMP2 in development of obesity is to some extent supported by the finding that STAMP2 levels are not normalized by weight loss. Furthermore, the observed adipose tissue-enriched expression of STAMP2 suggests that adipose tissue is an important site of action. However, the targets of STAMP2 in adipose tissue are unknown. According to our results, adiponectin is unlikely to mediate the association between human STAMP2 levels and insulin resistance. It should be noted that this assumption is based on measurements of mRNA and total protein secretion of adiponectin.
In accordance with the original results in mice (2), we propose that STAMP2 could be one regulator of inflammatory-induced insulin resistance in human adipose tissue. Recent studies have highlighted the importance on local inflammation in adipose tissue for the development of insulin resistance (9). In the present study, the proinflammatory cytokine TNFA clearly induces STAMP2 expression during human adipocyte differentiation. We observed no correlation between STAMP2 and TNFA mRNA levels in WAT but a strong positive correlation with expression levels of the macrophage marker CD68. This suggests that other, as-yet-unidentified inflammatory signals than TNFA are of importance for WAT STAMP2 action in vivo. We are aware that not only macrophages but also fat cells in adipose tissue express CD68 (10). However, we still think measuring CD68 levels provide relevant information because it has been shown that CD68 levels in adipose tissue correlate with number of macrophages and levels of different inflammatory markers (12).
We have previously shown that type 2 diabetes is associated with down-regulated expression of mitochondria electron transport chain genes in WAT and speculates that this is one pathway underlying induced insulin resistance (11). Interestingly, STAMP2 has been suggested to be involved in the regulation of the electron transport chain (2). We therefore speculate that obesity associated inflammatory changes in WAT and increased STAMP2 could result in electron transport chain dysfunction and hereby cause insulin resistance.
In summary, human adipose tissue STAMP2 levels associate with obesity and overall, as well as adipocyte, insulin resistance. However, the role of STAMP in obesity and its complications is probably minor in humans.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to declare.
First Published Online April 1, 2008
Abbreviations: AU, Arbitrary unit; BMI, body mass index; HOMA, homeostasis model assessment; STAMP2, six transmembrane protein of prostate 2; TNFA, TNF; WAT, white adipose tissue.
Received January 28, 2008.
Accepted March 24, 2008.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
-induced adipose-related protein (TIARP), a cell-surface protein that is highly induced by tumor necrosis factor- and adipose conversion. J Biol Chem 276:33938–33946-mediated lipolysis in human fat cells. J Biol Chem 277:1085–1091. Diabetes 55:1792–1799
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
