The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Department of Clinical Chemistry, Kobe Pharmaceutical University, Motoyamakita, Higashinada, Kobe, 658-8558, Japan. E-mail: firstname.lastname@example.org
Objective: In an attempt to clarify the conflicting data on resistin mRNA expression and protein analysis by western blotting in adipose tissue and serum, we developed a sensitive enzyme-linked immunosorbent assay (ELISA) for direct measurement of mouse resistin.
Research Methods and Procedures: We developed polyclonal antibodies directed to the N (21 to 40) and C (79 to 91) termini of mouse resistin. Then, affinity-purified anti-C-terminal resistin immunoglobin G (IgG) was biotinylated. ELISA was based on the sandwiching of antigen between antibody IgG coated on polystyrene plates and biotinylated antibody IgG. The bound biotinylated antibody was quantified with streptavidin-linked horseradish peroxidase.
Results: New ELISA can measure a concentration as low as 0.5 ng/mL of recombinant mouse resistin and is sensitive and specific enough to measure resistin protein in various adipose tissues and in sera. In normal mice, decreases in resistin concentrations in both white adipose tissue and serum were age dependent during 6 to 24 weeks of development. Resistin concentrations were significantly higher in omental adipose tissue in comparison with perirenal and abdominal adipose tissues and were 2- to 5-fold higher in females than males during the growth period. ob/ob mice had significantly lower resistin concentrations than the control mice in both sera and the white adipose tissues, particularly in the omental fat. The treatment by testosterone, but not progesterone or β-estradiol, in cultured adipocytes reduces resistin protein levels in a dose-dependent manner.
Discussion: New sensitive ELISA for mouse resistin clarified that the resistin concentrations in normal mice were markedly elevated in the omental adipose depots as compared with the perirenal and abdominal adipocyte depots and significantly elevated compared with adipose tissues in genetically obese mice.
Obesity predisposes a variety of illnesses, from hypertension and coronary heart disease to type 2 diabetes. Adipocytes are active in maintaining energy balance in the body (1, 2, 3) and are important in the processes of satiety, bone function, and reproduction. Most of their functions are carried out through proteins secreted by adipocytes capable of acting locally or at distant sites (4). Recent studies have shown that various proteins derived from adipose tissues that have endocrine functions contribute to insulin resistance, the hallmark of type 2 diabetes (5). More recently, a novel adipose-derived protein, named resistin, was identified by Steppan et al. (6). They speculated that it constitutes an important link between obesity and insulin resistance in type 2 diabetes. They found increased serum resistin concentrations in genetically and diet-induced animal models of obesity, whereas other researchers have reported a decrease in resistin mRNA levels in these animals (7, 8). Direct measurement of resistin protein, therefore, should resolve these conflicting results. It is useful to determine regional adipose differences in resistin concentrations because resistin is an adipocyte-secreted molecule with potential links to fat accumulation and insulin resistance.
We developed a sensitive enzyme-linked immunosorbent assay (ELISA)1 for mouse resistin and examined age-dependent changes in resistin concentrations in various mouse adipose tissues and sera of normal mice and in the concentrations in adipose tissues and sera of ob/ob mice.
Research Methods and Procedures
All animals were treated in accordance with the Animal Welfare Guideline of Utano National Hospital. ddY mice 6, 10, 15, and 24 weeks of age and C57BL/6n mice 8 weeks of age were purchased from SEAC Yoshitomi (Fukuoka, Japan). The number of animals for each age group was eight. C57BL/6J-ob (ob/ob) and C57BL/KsJ-db (db/db) mice at 8 weeks of age were a gift from Shionogi & Co., Ltd. (Osaka, Japan). The number of animals for each group (C57BL/6n, ob/ob and db/db mice) was four. After all animals were fasted for 24 hours, blood samples were taken, and their white adipose tissues (perirenal, abdominal, and omental adipose tissues) were removed and stored at −80 °C until used.
Preparation of Tissue Extracts and Serum Samples
Mouse sera were mixed with an equal volume of 4% sodium dodecylsulfate, then denatured with heat for 10 minutes at 100 °C. Mouse adipose tissues were weighed then homogenized in a Polytron homogenizer in 4 to 8 volume/weight of 25 mM phosphate-buffered saline (pH 7.4) containing 0.15 M NaCl and 1 mM phenylmethylsulfonyl fluoride. The homogenates were centrifuged at 15, 000 rpm for 20 minutes, after which the supernatants were analyzed.
Synthetic Peptides and Antibody Production
Synthetic peptides NH2-SSMPLCPIDEAIDKKIKQDF-COOH (N-terminal resistin residue, 21 to 40) and NH2-GSACGSWDIREEK-COOH (C-terminal resistin residue, 79 to 91) were both purified by high-performance liquid chromatography (purity > 85%) (Biologica Co., Ltd, Nagoya, Japan). The synthetic peptides were coupled to keyhole limpet hemocyanin with N-(ε-maleimidocaproyloxy) succinimide (Sigma Chemical Co., St. Louis, MO) to develop polyclonal antibodies for these resistins. The carrier-conjugated peptide was emulsified with Freund's complete adjuvant (Difco Laboratories, Detroit, MI) then injected subcutaneously (0.8 mg/injection) to rabbits six times at 10-day intervals. Blood samples were collected 10 days after the last injection. The specific antibodies in the sera were affinity purified in a resistin peptide-coupled CNBr-activated Sepharose column.
Preparation of Biotinylated Antiresistin Antibody Conjugate
Affinity-purified anti-C terminal resistin immunoglobin G (IgG) was biotinylated with 5-(N-succinimidyl-oxycarbonyl) pentyl-d-biotinamide (Dojindo, Kumamoto, Japan).
ELISA for Resistin
The two-site ELISA for resistin was performed as follows. Microtiter plates (Costar, Cambridge, MA) were coated with 100 μL of affinity-purified N-terminal resistin IgG (3.0 μg/mL) diluted with 10 mM carbonate buffer (pH 9.3) for 2 hours. After two washes of the plate, non-specific binding sites in each well were blocked for 1 hour with 200 μL of 10 mM carbonate buffer containing 0.5% bovine serum albumin (BSA). A standard solution (0 to 50 ng/mL of recombinant mouse resistin; American Research Products, Inc., Belmont, MA) and samples diluted with buffer [50 mM Tris-HCl buffer (pH 7.0) and containing 200 mM NaCl, 10 mM CaCl2, 1% BSA, and 0.1% Triton X-100] were added to the wells, after which the plate was incubated for 2 hours. After four washes with BSA-free dilution buffer, 100 μL of biotinylated anti-C terminal resistin IgG conjugate (5 ng/mL) was added, and the plate was incubated for 1 hour. After another four washes, the plate was incubated for 1 hour with 100 μL of streptavidin-linked horseradish peroxidase (Amdex, Jyllinge, Denmark) diluted 10, 000-fold. After four final washes, the plate was kept for 20 minutes to react with 100 μL of a substrate solution of 3′, 3′, 5′, 5′-tetramethylbenzidine and H2O2 (Kirkegaard & Perry Laboratories Inc., Gaithersburg, MD). The reaction was stopped by the addition of 100 μL of 1 M phosphoric acid, and then the absorbance at 450 nm was read with an ELISA reader (Labsystems, Vantaa, Finland). All washes and incubations were done at room temperature with gentle shaking.
3T3-L1 fibroblasts (Japan Health Sciences Foundation, Osaka, Japan) initially were maintained at 37 °C in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS) (Gibco BRL, Life Technologies, Inc., Rockville, MD) under a 95% air/5% CO2 atmosphere. Differentiation was induced 2 days after confluence was reached by incubating the cells for 48 hours in DMEM supplemented with 0.5 mM 3-isobutyl-1-methylxanthine, 4 μg/mL dexamethasone, and 10% FCS. The cells then were maintained for an additional 4 to 6 days in DMEM supplemented with 10% FCS, the medium being changed every other day. By this protocol, >90% of the cells expressed the adipocyte phenotype.
All values are means ± standard deviation (SD). Differences were considered statistically significant at p < 0.05. Comparisons among groups were made by one-way ANOVA with post hoc Bonferroni test followed by the unpaired, non-parametric Student's t test.
Calibration Curve of the Sandwich-Type ELISA for Mouse Resistin
Polyclonal antibodies directed to the N (21 to 40) and C (79 to 91) termini of mouse resistin were used for the sandwich-type ELISA for mouse resistin. These antibodies detected a 12-kDa band that comigrated with the recombinant resistin (data not shown). Resistin detection was inhibited by preincubation of the antibody with the peptide against which it was raised. The calibration curves for mouse recombinant resistin show a quantitative relationship for 0.5 to 50 ng/mL (Figure 1). The accuracy of the assay was determined by recovery experiments on two samples with 5.0 and 25.0 ng/mL resistin concentrations. When two different amounts of recombinant mouse resistin (5.0 or 10 ng/mL) were added to the samples, the average recovery was 95.0 ± 5.7%. The respective within- and between-day coefficients of variation were 1.8% to 2.1% (n = 10) and 7.4% to 8.3% (n = 10). This new assay can measure a concentration as low as 0.5 ng/mL and is sensitive and specific enough to measure resistin protein in various adipose tissues and in sera. No significant cross-reactivity to human recombinant resistin was found at 50 ng/mL (data not shown). The detectable cross-reactivity was with recombinant mouse resistin-like molecule-α (1.1%) and -β (0.7%) (PeproTech Inc., Rocky Hill, NJ).
Age-Related Resistin Changes in Normal Mouse Adipose Tissues
We examined age-related changes in the resistin contents of visceral adipose depots (omental, abdominal, and perirenal fats) during postnatal development of male and female ddY mice at 6, 10, 15, and 24 weeks of age (Figure 2). The resistin concentration of each tissue was calculated from the gram wet weight of the tissue. In the normal male mice, resistin concentrations were highest at week 6 in the perirenal (990 ng/g), abdominal (500 ng/g), and omental (1020 ng/g) fat, respectively decreasing rapidly to 41%, 33%, and 73% by week 10, after which they slowly decreased or remained unchanged up to 24 weeks of age. In the normal female mice, concentrations also were highest at week 6 in the perirenal (940 ng/g), abdominal (∼1090 ng/g), and omental (3770 ng/g) fat, decreasing gradually up to 24 weeks of age. Resistin concentrations in the perirenal, abdominal, and omental fat, respectively, decreased to 28%, 36%, and 54% in the males and 40%, 49%, and 49% in females up to 24 weeks of age. Resistin concentrations in the perirenal, abdominal, and omental fat of the female mice were ∼1.6-, 2.8-, and 3.4-fold those in the males. Resistin was markedly increased in the omental fat (males, 590–1020 ng/g; females, 1820–3770 ng/g) throughout development. In the female mice, it was elevated ∼3.5-fold in the omental fat as compared with the perirenal and abdominal fat.
Age-Related Changes in Serum Resistin Concentrations in Normal Mice
Resistin concentrations were high at week 6 in the ddY mice (males, 7.7 ng/mL; females, 23.5 ng/mL), then slowly decreased up to 24 weeks of age (Figure 3). Concentration in the females throughout development were ∼3-fold those in the males.
Resistin Concentrations in Adipose Tissues of ob/ob Mice
Resistin concentrations in the adipose tissues of ob/ob mice (C57BL/6J-ob) at 8 weeks of age, in whom obesity is an inherited trait, were markedly decreased as compared with the age-matched control mice (C57BL/6n) (Figure 4, A, males; B, females). Relative respective decreases in resistin for perirenal, abdominal, and omental fat in the male ob/ob mice were 30%, 40%, and 18% and in the females 24%, 23%, and 8%, in comparison with the controls. Resistin in the omental fat was severely decreased. Unlike in the normal mice, there was no difference between male and female ob/ob mice, or in the three adipose tissues.
Resistin Concentrations in Sera of ob/ob and db/db Mice
Serum resistin concentrations in both ob/ob (C57BL/6j-ob) and db/db (C57BL/KsJ-db) mice at 8 weeks of age were markedly decreased as compared with the age-matched control mice (C57BL/6n) in both male and female mice (Figure 5).
Effect of Testosterone, Progesterone, or β-Estradiol on the Secretion of Resistin by Cultured 3T3-L1 Adipocytes
3T3-L1 cells were incubated with various concentrations of testosterone, progesterone, or β-estradiol for 24 hours, after which the resistin concentration in each culture medium was measured directly by our ELISA. The effects of testosterone, progesterone, or β-estradiol on the expression of resistin protein are shown as percentages of the control value. 3T3-L1 cells were exposed to 0.01, 0.1, and 1 μM testosterone or progesterone and 0.01, 0.03, 0.1, 0.3, 1, 3, and 10 μM β-estradiol for 24 hours. Testosterone (0.01 and 0.1 μM) significantly (p < 0.05) decreased them by 33% and 42% and 1 μM of testosterone by 48% (p < 0.01) as compared with the control values (Figure 6A). Progesterone did not affect the resistin levels at those concentrations (Figure 6B). β-Estradiol significantly decreased them at 0.01, 0.03, and 1 μM and did not affect them at 0.1, 0.3, 3, and 10 μM (Figure 6C).
Obesity is considered the most common and most important factor in the development of insulin resistance. Studies suggest that visceral obesity (of which omental fat is a surrogate) carries a high risk of complications (9, 10). Visceral fat retains markedly more resistin mRNA than subcutaneous fat (11). Surgical removal of visceral fat has been shown to improve the insulin effect on hepatic glucose production in animal models of obesity (11, 12). Other reports, however, show that abdominal subcutaneous adipose tissue may be more important (13, 14). Increased resistin expression in abdominal fat may explain the increased risk for type 2 diabetes associated with central obesity (14). McTernan et al. (15) reported that resistin protein expression levels were similar in both human abdominal subcutaneous and omental adipose tissue depots and that expression in abdominal fat depots was increased in comparison with thigh and breast tissue depots. Thus, there are conflicting data with regard to resistin expression in adipose tissue, as in serum, in those studies confirmed mainly at the mRNA level or by protein analysis using western blotting (16, 17). We developed a sensitive sandwich-type ELISA for mouse resistin that measures a concentration as low as 0.5 ng/mL. Resistin was readily assessed in mouse adipose tissues and serum. We consider it very important to assess age-related changes in the resistin concentrations of the various adipose tissues of normal mice and, therefore, examined concentrations in white adipose tissue depots, including abdominal, omental, and perirenal fat, during postnatal development at 6, 10, 15, and 24 weeks of age. Resistin concentrations in the adipose tissues decreased with age during postnatal development and were ∼3.5-fold higher in the omental than the perirenal and abdominal fat throughout development. These findings show that the resistin concentrations in normal mice are markedly elevated in the omental adipose depots as compared with the perirenal and abdominal adipocyte depots. Resistin concentrations in the adipose tissues and sera were significantly higher in the female mice throughout development. We measured resistin protein in the media from 3T3-L1 adipocytes after testosterone, progesterone, or β-estradiol treatment. Testosterone treatment in cultured adipocytes reduces resistin protein levels in dose-dependent manner. However, progesterone did not suppress it, and β-estradiol did not suppress resistin expression in a dose-dependent manner. Testosterone may function to suppress resistin production and secretion. These in vitro findings may be related to the more markedly decreased resistin concentration in male than female mice. However, there are conflicting data using male prolactin transgenic and castrated mice in which resistin mRNA expression levels are increased in vivo by testosterone (18).
Steppan (6) reported elevated serum resistin in both diet-induced and genetically obese mice and that the adipocyte resistin gene and protein expression, both down-regulated by fasting, rapidly increased on refeeding. Other studies (7, 8), however, marked down-regulation of adipocyte resistin gene expression in both high-fat diet-induced obesity mice and a genetic model of obesity that included ob/ob, db/db, tub/tub, and KKAy mice. Kim reported very low resistin mRNA expression in adipose tissue of fasting or diabetic animals that increased markedly on feeding or insulin administration (19). In obese rats, resistin was down-regulated in visceral adipose tissue (20). To address these differences, we measured resistin concentrations in the white adipose tissues of ob/ob mice and compared them with those of age-matched control mice (C57BL/6n). Resistin concentrations in the omental, abdominal, and perirenal fat were all markedly decreased in ob/ob mice in comparison with the controls. Suppression was most dramatic in the omental fat (82% reduction in males, 92% reduction in females relative to the controls). No significant differences were found between the male and female groups nor between the adipose tissues of ob/ob mice. C57BL/6n mice showed the 2-fold high level of resistin in omental adipose as compared with ddY mice. The differences of resistin levels between two normal mouse models might be because of the strain differences. Because the resistin concentration in adipose tissue is expressed as nanograms per gram wet weight of adipose tissue, and ob/ob mice have a greater mass of adipose tissue and may have an altered secretion of resistin, we measured circulating levels of resistin. Serum resistin concentrations in ob/ob mice (C57BL/6j-ob) at 8 weeks of age were markedly decreased as compared with the age-matched control mice (C57BL/6n) in both male and female mice. In addition, we also found significantly low concentrations in sera of the mouse model, db/db (C57BL/KsJ-db) mice. Our results by ELISA are contrary to a report by Rajala et al. (21) demonstrating the elevated levels of serum resistin in ob/ob mice; in addition, there were no significant differences between males and females in wild-type C57BL/6J mice. Their serum resistin values are approximately twice those reported by us. These might be based on differences between two assay systems, namely RIA by a competitive assay of which absolute values are influenced by the affinity of an antibody used, and sandwich-type ELISA is performed by putting resistin into two antibodies with different specificities. Because recent reports have implicated a role for resistin in glucose homeostasis (22, 23), it would be extremely interesting to determine whether correlations exist between resistin and glucose levels during the pathogenesis of type 2 diabetes in the ob/ob mouse model. Therefore, further studies are needed to analyze the relation to the circulating levels of resistin during the development of obesity and its associated diabetes in the ob/ob mouse as compared with its littermates using our newly developed immunoassay. The ability to precisely and accurately measure circulation levels of resistin is an invaluable tool for understanding the regulation of mouse resistin.
Adipocytes secrete a variety of mediators, including tumor necrosis factor (TNF)-α, leptin, free fatty acids, and adiponectin, all of which affect the body's ability to respond to insulin and metabolize glucose. Serum TNF-α levels are elevated in obese rodent models (24, 25) and may contribute to obesity-induced insulin resistance (26, 27, 28, 29, 30). Some evidence suggests that thiazolidinediones improves insulin sensitivity by suppressing the production of TNF-α in enlarged adipocytes (31). TNF-α strongly suppresses resistin expression (17). Our findings are consistent with those of other investigators (7, 8) showing that adipose tissue resistin expression is severely suppressed in obesity. Decreased resistin expression in the obese model suggests that resistin is produced in enlarged adipocytes but that concurrently elevated TNF-α strongly suppresses resistin expression.
Using our sensitive ELISA system to determine resistin protein concentrations, we plan further studies to investigate the mode of regulation and biological functions of resistin and whether it is an effector of insulin resistance in obesity.
This work was supported, in part, by a grant-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology, Japan and by the Kobe Pharmaceutical University Collaboration Fund and the Science Research Promotion Fund of the Japan Private School Promotion Foundation.