| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Cell Physiology and Metabolism (C.A., P.C.R., C.T.-C., F.R.-J., P.M.), University Medical Center; and Division of Endocrinology, Diabetology, and Nutrition (C.A., P.C.R., C.T.-C., F.R.-J.), Department of Internal Medicine, Faculty of Medicine, University of Geneva, 1211 Geneva 4, Switzerland
Address all correspondence and requests for reprints to: Dr. Patrick Muzzin, Department of Medical Biochemistry, University Medical Center, University of Geneva, 1 rue Michel Servet, 1211 Geneva 4, Switzerland. E-mail: patrick.muzzin{at}medecine.unige.ch.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent in vivo and in vitro studies indicated that resistin influences glucose metabolism. In rats, resistin infusion markedly increased glucose production by the liver and thereby decreased hepatic insulin action (5). Furthermore, Pravenec et al. (6) reported that transgenic rats overexpressing resistin showed both impaired skeletal muscle glucose metabolism and glucose intolerance. In L6 rat muscle cells, resistin was reported to inhibit insulin-stimulated glucose uptake without altering insulin receptor signaling, presumably by decreasing the intrinsic activity of cell surface glucose transporters (7). Finally, in humans a recent longitudinal analysis revealed that serum resistin levels are positively correlated with changes in body mass index, body fat, visceral fat area, mean glucose, and insulin in obese individuals (8).
However, and in contrast with the above-mentioned observations, other data have cast some doubt on the hypothesis linking resistin with insulin resistance by showing that resistin mRNA expression was paradoxically decreased in white adipose tissue in several obesity models, such as ob/ob, db/db, tub/tub, KKAy, and diet-induced obese mice (9) and that resistin serum levels were decreased in C57, db/db, and KKAy mice fed a high-fat diet (10).
Thus, the regulation of resistin expression in rodent models of obesity is controversial, and its potential role as a mediator between adipose tissue and insulin resistance remains to be clarified. In the present study we decided to evaluate white adipose tissue resistin expression in a model of insulin resistance before and during normalization of the insulin resistance state. For this purpose, we used ob/ob mice treated with leptin. It is well known that leptin replacement in ob/ob mice inhibits food intake, reduces body weight, stimulates energy expenditure, and decreases hyperglycemia and hyperinsulinemia (11, 12). Our reasoning was that as leptin replacement in ob/ob mice is able to attenuate the insulin resistance state present in these mice, and it has been proposed that resistin may contribute to decrease insulin sensitivity, leptin may ameliorate the insulin sensitivity of ob/ob mice by decreasing resistin gene expression in white adipose tissue. If leptin proved efficient in decreasing resistin gene expression in white adipose tissue concomitantly with its insulin sensitivity-ameliorating effect, we asked what is the site of leptin action? Is it central or peripheral?
As a complementary approach to assess the impact of the dynamic changes in body weight and glucose metabolism on adipose tissue resistin expression, C57BL/6J mice were fed a high-fat diet, which has been shown to induce weight gain and favor the development of insulin resistance in this strain of rodents. After a short time of high-fat feeding, we measured adipose resistin expression and looked for potential correlations between resistin gene expression and various parameters of insulin resistance.
To summarize, the precise aims of this study were 1) to compare resistin expression between ob/ob and control mice, 2) to determine whether leptin treatment is able to down-regulate resistin gene expression in ob/ob mice, 3) to investigate whether the possible leptin effects are centrally or peripherally mediated, and 4) to check for correlations between resistin expression and various parameters of insulin resistance in rodent models of obesity.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals
Two- to 3-month old C57BL/6J wild-type, ob/ob, and db/db male mice (Elevage Janvier, France) or male Wistar rats (Charles River Laboratory, Lexington, MA) were housed individually and kept on a 12-h light, 12-h dark cycle in a temperature-controlled room at 24 C. They were allowed ad libitum access to water and a standard laboratory chow unless otherwise stated. All procedures used were approved by the Office Vétérinaire Fédéral et Cantonal (Geneva, Switzerland).
High-fat diet
The high-fat diet is composed of pork fat (2154, Kliba, Kaiseraugst, Switzerland). It has an energy density of 4.31 kcal/kg, with 40% of energy derived from fat. C57BL/6J wild-type mice were fed a standard chow diet until 6 wk of age, then were fed the high-fat diet for 2, 4, or 6 d.
Leptin injections
Mice were divided into three groups: one received daily ip injections of leptin at a dose of 10 µg/g body weight for 4 d, and the other two were injected daily with PBS and either fed ad libitum or pair-fed with the leptin-treated group. Food intake and body weight were measured daily during 1 wk before the injection period and during the 4-d experimental period. The animals were killed by cervical dislocation, and the epididymal white adipose tissue was rapidly dissected and immediately frozen in liquid nitrogen.
Chronic intracerebroventricular (icv) infusions of leptin
Rats were anesthetized with im ketamine/xylasine (45 and 9 mg/kg, respectively; Parke Davis and Bayer, Leverkusen, Germany) and were equipped with a cannula positioned in the right lateral ventricle. After 1 wk of recovery, osmotic minipumps (model 2001, Alzet, Cupertino, Canada) delivering 12 µg leptin/d for 7 d or its vehicle (isotonic saline) were connected to the icv infusion cannula via a polyethylene catheter under isofluorane (Abbott Laboratories, Chicago, IL) anesthesia. The leptin dose was chosen on the basis of previous studies in which we demonstrated that there was no leakage of the icv leptin infused into the general circulation (13). Three groups of rats were investigated: 1) rats infused with leptin eating ad libitum, 2) control rats infused with vehicle and allowed to eat ad libitum, and 3) control rats infused with the vehicle, but pair-fed to the amount of food consumed by the leptin-infused animals.
Total RNA isolation and Northern blot analysis
Total RNA was isolated by the method of Chomczynski and Sacchi (14). Fifteen micrograms of RNA were electrophoresed in a 1% agarose gel containing formaldehyde and transferred according to standard protocols. The probes used were a full-length mouse resistin cDNA (GenBank accession no. NM_022984). They were labeled by random priming with [
-32P]deoxy-CTP to a specific activity of approximately 1 x 109 dpm/µg DNA. Hybridizations were performed using QuikHyb solution, as previously described (15). Blots were exposed at –80 C to Hyperfilm enhanced chemiluminescence films (Amersham Biosciences UK Ltd., Buckinghamshire, UK). RNA levels were quantified by scanning photodensitometry of the autoradiograms using ImageQuant (version 3.3, Molecular Dynamics, Sunnyvale, CA). Membranes were subsequently reblotted with a probe against cyclophilin or ß-actin to ensure that equivalent amounts of RNA were loaded onto the gel. Representative blots are shown for each figure.
Quantitative PCR
cDNA templates for RT-PCR were obtained using 2.5 µg total RNA. The RT reaction was performed with random hexamers (Microsynth, Balgach, Switzerland); deoxy-NTPs; the ribonuclease inhibitor, RNasin (Catalys, Promega Corp., Madison, WI); and the Moloney murine leukemia virus reverse transcriptase enzyme kit (Life Technologies, Inc., Gaithersburg, MD).
The real-time PCR (LightCycler, Roche, Basel, Switzerland) reaction is an automated quantitative PCR obtained by the continuous monitoring of the fluorescence emitted on binding of the SYBR Green I dye to the double-stranded DNA after each amplification cycle. Amplification of cyclophilin A, fatty acid binding protein 4 (AP-2), resistin, and 11ß-hydroxysteroid dehydrogenase type I (11ßHSD1) was performed with the SYBR Green I DNA master kit (Roche, Mannheim, Germany), according to the light cycler standard protocol, using about 70 ng template cDNA. Primers for cyclophilin used at a final concentration of 0.5 µM were designed on-line with Primer 3 software (http://genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi) and synthesized by Microsynth. They were as follows: mouse cyclophilin A: sense primer, 5'-AGCACTGGGGAGAAAGGATT-3'; antisense primer, 5'-CATGCCTTCTTTCACCTTCC-3' (product size, 306); mouse AP-2: sense primer, 5'-GATGCCTTTGTGGGAACC-3'; antisense primer, 5'-CTCTTGTGGAAGTCACGC-3' (product size, 373); mouse 11ßHSD1: sense primer, 5'-GTCCCTGTTTGATGGCAGTT-3'; antisense primer, 5'-TTCCCTGGAGCATTTCTGGT-3' (product size, 114); rat resistin: sense primer, 5'-TGAAGCCATCAGCAAGAAGATC-3'; antisense primer, 5'-GACCAGCAATGTAGGACAGTGTTC-3' (product size, 84); and rat cyclophilin A: sense primer, 5'-CAAATGCTGGACCAAACACAA-3'; antisense primer, 5'-GCCATCCAGCCACTCAGTCT-3' (product size, 70).
After each run, a relative quantification of amplified PCR products in the different samples was performed. For this purpose, standard curves were constructed for the gene of interest as well as for cyclophilin, which was used as a housekeeping gene. The results are expressed as the ratio between the concentration of the target gene and that of cyclophilin.
Western blot
White adipose tissue was homogenized by potterization in PBS containing 1% sodium dodecyl sulfate and 0.06% protease inhibitor cocktail (Sigma-Aldrich Corp., St. Louis, MO; P8340). Fifteen micrograms of white adipose tissue homogenates were dried under vacuum and resuspended in 10 µl of a loading buffer containing 50% glycerol, 5% sodium dodecyl sulfate, 2.5% bromophenol blue, 0.5 M Tris-HCl (pH 6.8), and 20% ß-mercaptoethanol. The samples were electrophoresed on a 15% polyacrylamide, 0.1% sodium dodecyl sulfate gel, and transferred to a polyvinylidene difluoride membrane by electroblotting transfer with a buffer containing 10% methanol, 25 mM Tris-HCl (pH 6.8), and 190 mm glycine. The membrane was blocked with a PBS buffer containing 0.1% Tween and 5% nonfat dry milk. This same buffer was used for all antibody incubations. Resistin protein was detected using a rat monoclonal primary antibody raised against mouse resistin protein, which was provided by Dr. M. A. Lazar (University of Pennsylvania School of Medicine, Philadelphia, PA) and a 1:6000 diluted monoclonal antirat peroxidase-labeled secondary antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA; sc-2006). The ß-actin protein was detected as described above using a monoclonal antibody specific for mouse ß-actin (Sigma-Aldrich Corp.; A4700), and a 1:3000 diluted goat antimouse peroxidase-labeled secondary antibody on the same membrane to ensure that equivalent amounts of proteins were loaded onto the gel. The signals were detected by chemiluminescence using a standard enhanced chemiluminescence kit and developed on a Hyperfilm ECL. They were quantified by scanning photodensitometry. Representative blots are shown for each figure.
Plasma hormones and metabolites measurements
Plasma insulin concentrations were measured by RIA (16). Plasma glucose levels were determined using kits from Roche (Mannheim, Germany). Plasma leptin levels were measured by ELISA from Crystal Chemicals, Inc. (Chicago, IL).
Statistical significance
The results are given as the mean ± SEM. Statistical analysis was performed using one-way ANOVA, followed by the Tukey procedure, or using Kruskal-Wallis one-way ANOVA on ranks for multiple comparisons. Correlations were analyzed by forward stepwise regression and multiple linear regression. The calculations were performed using SigmaStat (SPSS, Inc., Chicago, IL). P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, resistin gene and protein expression were, respectively, 20- and 10-fold lower in obese compared with control animals. Although resistin was strongly decreased in adipose tissue of ob/ob mice, it was easily detected by Northern and Western blots, as shown by the signals of its mRNA and protein provided in the insets of Fig. 1.
|
). The decrease in food intake accounted for the effect on body weight, as suggested by a similar body weight loss in leptin-treated and pair-fed animals (P = 0.418, not significant; Table 1). In addition, leptin treatment of ob/ob mice was able to decrease both plasma glucose and insulin levels by 30% and 80%, respectively, whereas pair-feeding did not alter these parameters compared with those in ad libitum-fed control mice (Fig. 2, A and B). The calculated insulin/glucose index was strikingly reduced by 77% in the leptin-treated group, but it remained unchanged in the pair-fed group (Fig. 2C), suggesting that leptin administration was able to ameliorate the insulin sensitivity of ob/ob mice.
|
|
). This suggests that leptin is able to down-regulate resistin gene expression independently from its ability to decrease food intake.
Interestingly, we found a very strong correlation between resistin gene expression and body weight loss in the leptin-treated group (Fig. 3). This correlation was absent in the pair-fed group (data not shown). This further suggests that the inhibitory effect of leptin on resistin gene expression is not related to changes in food intake.
|
). We then looked at whether this correlation was also present in a second model of insulin resistance, that of the normal C57BL/6J mice fed a high-fat diet for 2, 4, or 6 d to study the early phase of development of the syndrome. As shown in Table 2, all of the mice exhibited increased body weight, caloric intake, and leptin levels at the various time points. Moreover, we observed that glycemia was only significantly increased on d 4 of the high- fat diet compared with that on d 0, although trends toward increases in glycemia were also observed on d 2 and 6 of high-fat feeding. When considering resistin gene expression in white adipose tissue, we observed that its expression was rapidly (d 2) increased and remained high in response to a high-fat diet, to further increase on d 6 (Fig. 5). We then looked for potential correlations between resistin gene expression and glycemia and found a significant positive correlation between these two parameters (Fig. 6). We did not find any correlation between resistin gene expression and insulinemia or change in body weight in high-fat-fed mice (data not shown).
|
|
|
|
). Of note is the finding of a strong positive correlation between AP-2 and resistin mRNA levels (r2 = 0.61; P < 0.001). Furthermore, it is well known that increased 11ßHSD1 activity and expression are closely associated with both increased adiposity and insulin resistance (18). Thus, we looked at the expression of 11ßHSD1 by semiquantitative PCR and observed that 11ßHSD was rapidly (d 2) increased (136%) and remained high in response to a high- fat diet (Table 2). In addition, we found a statistically significant positive correlation between 11ßHSD1 and resistin mRNA levels (r2 = 0.22; P = 0.04).
Leptin effect on resistin gene expression depends on the long leptin receptor isoform
To determine whether the leptin inhibitory effect on resistin gene expression is dependent on the long leptin receptor isoform, we treated db/db mice with leptin. As expected, leptin had no effect on food intake (5.2 ± 0.5 g/d for controls and 5.3 ± 0.3 g/d for leptin-treated animals), body weight changes (–0.3 ± 0.3 g for controls and –0.5 ± 0.3 g for leptin-treated animals), or glycemia (287 ± 16 mg/dl for controls and 260 ± 17 mg/dl for leptin-treated animals) in db/db mice. It also did not change resistin gene or protein expression in these mice, suggesting that the regulation of resistin by leptin requires the long leptin receptor isoform (data not shown).
Chronic icv leptin infusion down-regulates resistin gene expression in white adipose tissue of normal rats
To determine whether the leptin inhibitory effect on resistin gene expression is centrally mediated, we treated normal rats with icv leptin administered chronically over 7 d. As expected, leptin treatment reduced the food intake of normal rats and resulted in body weight loss in these animals (Table 3). The decrease in food intake accounted for the effect on body weight, as suggested by similar losses in body weight in leptin-treated and pair-fed animals.
|
). This further strengthens the observation mentioned above that leptin is able to down-regulate resistin gene expression independently from its effect to decrease food intake, and that this leptin effect is centrally mediated.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Leptin administration to ob/ob mice resulted in decreased food intake, body weight, and adipose resistin expression. Pair-fed ob/ob mice also displayed reduced body weight, but they had unchanged resistin expression. These observations suggest that leptin by itself may modulate resistin expression in ob/ob mice. Moreover, we observed a strong correlation between body weight loss and resistin mRNA levels in the leptin-treated mice, but not in the pair-fed group, again indicating that the decrease in resistin expression resulted from the presence of leptin itself, rather than being due to the leptin-induced decrease in food intake. However, the observation that several rodent models of obesity exhibiting a lack or an excess of leptin display a down-regulation of resistin imply that factors other than leptin may alter resistin expression in obese animals (9).
We found that leptin administration in db/db mice does not change body weight, food intake, or glucose metabolism, nor does it affect resistin expression. These results demonstrate that down-regulation of adipose resistin expression in leptin-treated mice requires the presence of the long leptin receptor isoform. Moreover, we observed that chronic icv leptin administration in rats decreases adipose resistin gene expression, suggesting that the effect of leptin on resistin results from a central effect of leptin. This observation is in keeping with a recent study of Yuzuriha et al. (19) in which the authors showed that icv administration of neuropeptide Y (NPY) in food-deprived mice led to increased resistin gene expression in white adipose tissue by 72% compared with artificial cerebrospinal fluid-treated controls. As leptin down-regulates NPY levels in the hypothalamus (12), one can hypothesize that its inhibitory effect on resistin expression is mediated by a decrease in hypothalamic NPY.
Interestingly, resistin gene expression was found to be correlated with plasma glucose levels in ob/ob mice. We also observed a correlation between these two parameters when C57BL/6J wild-type mice were placed on a high-fat diet, whereas there was no correlation with other metabolic parameters that vary in high-fat feeding, such as caloric intake, body weight, insulinemia, and resistin gene expression. It should be mentioned at that point that the correlations between resistin gene expression and plasma glucose levels were observed in two particular contexts of insulin sensitivity changes, i.e. the administration of leptin to ob/ob mice and the high fat feeding of C57BL/6J mice, and it is thus not possible to extend the validity of such correlations to the normal situation. This being said, our results obtained in ob/ob mice and high fat-fed obese mice are in keeping with in vivo studies by Rajala et al. (5) reporting that infusion of resistin in rats during euglycemic-hyperinsulinemic clamps markedly increased glucose production by the liver and thereby modulated hepatic insulin action. Moreover, the same group had previously demonstrated that acute hyperglycemia under normoinsulinemic conditions also led to up-regulation of resistin transcription in various adipose depots (4). Together these data suggest the existence of a cross-talk regulation between resistin and plasma glucose levels in models of insulin resistance. In this respect, it is interesting to note that an important set of genes involved in metabolic and energetic pathways are reportedly directly regulated by glucose (20). Smih et al. (21), for example, demonstrated that the proximal 137 bp of the hormone-sensitive lipase promoter contain a glucose-responsive region. As far as resistin is concerned, an up-regulation of its mRNA by high glucose concentrations was demonstrated in vitro (22). It would be of interest to determine whether a glucose-responsive element is present in the resistin promoter.
Regarding the model of high-fat feeding, resistin gene expression is rapidly up-regulated (d 2) in response to high- fat feeding. In addition, we observed a progressive increase in AP-2 gene expression, suggesting that high-fat feeding induced a progressive differentiation of adipose tissue. By analogy with the observation of Kim et al. (1) and Haugen et al. (17), who showed that resistin gene expression is induced during adipocyte differentiation, we noted a positive correlation between resistin mRNA and AP-2 mRNA in white adipose tissue of high fat-fed mice. Thus, it is conceivable that in our model, the increase in resistin gene expression is due to adipocyte differentiation. On the other hand, it has been shown that recombinant resistin decreased insulin-stimulated glucose uptake in 3T3-L1 adipocytes (3) and that conditioned medium from COS cells transfected with resistin inhibited the conversion of 3T3-L1 cells to adipocytes (1). Thus, we could speculate that the increase in adipocyte number leads to a rise in local resistin, which, in turn, would inhibit insulin action on glucose uptake in adipose tissue and would also prevent further adipocyte differentiation. Altogether, resistin might therefore be a feedback regulator of adipogenesis as was proposed by others (1, 6). Although this mechanism is purely speculative, it would be of interest to determine whether resistin might affect adipogenesis and, if so, whether it does so by interacting with the insulin signaling pathways.
In conclusion, resistin mRNA and protein expression are markedly reduced in white adipose tissue of ob/ob mice compared with those of wild-type controls. Leptin replacement by peripheral injection in ob/ob mice is able to ameliorate the insulin sensitivity of these mice, as indicated by the decrease in the insulin to glucose ratio. This amelioration state corresponds to a decrease in resistin gene expression, suggesting that resistin down-regulation might be implicated in the leptin effect on insulinogenic index. Leptin given icv is also able to promote white adipose tissue resistin down-regulation, showing that the leptin effect on resistin is centrally mediated. Finally, our findings that resistin gene expression and plasma glucose levels are correlated in leptin-treated ob/ob mice and high fat-fed mice suggest a direct or indirect cross-talk between glucose homeostasis and resistin.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: AP-2, Fatty acid binding protein 4; 11ßHSD1, 11ß-hydroxysteroid dehydrogenase type I; icv, intracerebroventricular; NPY, neuropeptide Y.
Received December 10, 2003.
Accepted for publication February 3, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in adipose tissue. Mol Endocrinol 16:1920–1930
agonists. J Biol Chem 276:25651–25653This article has been cited by other articles:
 |
|
 |
|
  S. Park, S. M. Hong, S. R. Sung, and H. K. Jung Long-Term Effects of Central Leptin and Resistin on Body Weight, Insulin Resistance, and {beta}-Cell Function and Mass by the Modulation of Hypothalamic Leptin and Insulin Signaling Endocrinology, February 1, 2008; 149(2): 445 - 454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Kusminski, N. F. da Silva, S. J. Creely, F. M. Fisher, A. L. Harte, A. R. Baker, S. Kumar, and P. G. McTernan The in Vitro Effects of Resistin on the Innate Immune Signaling Pathway in Isolated Human Subcutaneous Adipocytes J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 270 - 276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Qi, Z. Nie, Y.-S. Lee, N. S. Singhal, P. E. Scherer, M. A. Lazar, and R. S. Ahima Loss of Resistin Improves Glucose Homeostasis in Leptin Deficiency Diabetes, November 1, 2006; 55(11): 3083 - 3090. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Wang, K. A. Al-Regaiey, M. M. Masternak, and A. Bartke Adipocytokines and lipid levels in ames dwarf and calorie-restricted mice. J. Gerontol. A Biol. Sci. Med. Sci., April 1, 2006; 61(4): 323 - 331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. W. Rajala, Y. Qi, H. R. Patel, N. Takahashi, R. Banerjee, U. B. Pajvani, M. K. Sinha, R. L. Gingerich, P. E. Scherer, and R. S. Ahima Regulation of Resistin Expression and Circulating Levels in Obesity, Diabetes, and Fasting Diabetes, July 1, 2004; 53(7): 1671 - 1679. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
