Transgenic and Recombinant Resistin Impair Skeletal Muscle Glucose Metabolism in the Spontaneously Hypertensive Rat*

Increased serum levels of resistin, a molecule secreted by fat cells, have been proposed as a possible mecha-nistic link between obesity and insulin resistance. To further investigate the effects of resistin on glucose metabolism, we derived a novel transgenic strain of spontaneously hypertensive rats expressing the mouse resistin gene under the control of the fat-specific aP2 promoter and also performed in vitro studies of the effects of recombinant resistin on glucose metabolism in isolated skeletal muscle. Expression of the resistin transgene was detected by Northern blot analysis in adipose tissue and by real-time PCR in skeletal muscle and was associated with increased serum fatty acids and muscle triglycerides, impaired skeletal muscle glucose metabolism, and glucose intolerance in the absence of any changes in serum resistin concentrations. In skeletal muscle isolated from non-transgenic spontaneously hypertensive rats, in vitro incubation with recombinant resistin significantly inhibited insulin-stimulated glycogenesis and reduced glucose oxidation. These findings raise the possibility that autocrine effects of resistin in Biochemical Analyses— Blood glucose levels were measured by the glucose oxidase assay (Pliva-Lachema) using tail vein blood drawn into 5% trichloroacetic acid and promptly centrifuged. Serum non-esterified fatty acid (NEFA) levels were determined using an acyl-CoA oxidase- based colorimetric kit (Roche Diagnostics). Serum triglyceride concentrations were measured by standard enzymatic methods (Pliva-La- chema). Serum insulin concentrations were determined using a rat insulin radioimmunoassay kit (Amersham Biosciences). Serum levels of leptin were determined using a rat leptin radioimmunoassay kit from Linco Research (St. Charles, MO). Serum resistin concentrations were determined at Linco Research, Inc. with an immunoassay that cross-reacts with mouse and rat resistin (Linco Research, Inc.). Statistical Analysis— of the data are expressed as means (cid:5) S.E. Differences between control and experimental groups were evaluated by paired or non-paired Student’s t tests as appropriate. Statistical significance was defined as p (cid:6) 0.05.

Increased serum levels of resistin, a molecule secreted by fat cells, have been proposed as a possible mechanistic link between obesity and insulin resistance. To further investigate the effects of resistin on glucose metabolism, we derived a novel transgenic strain of spontaneously hypertensive rats expressing the mouse resistin gene under the control of the fat-specific aP2 promoter and also performed in vitro studies of the effects of recombinant resistin on glucose metabolism in isolated skeletal muscle. Expression of the resistin transgene was detected by Northern blot analysis in adipose tissue and by real-time PCR in skeletal muscle and was associated with increased serum fatty acids and muscle triglycerides, impaired skeletal muscle glucose metabolism, and glucose intolerance in the absence of any changes in serum resistin concentrations. In skeletal muscle isolated from non-transgenic spontaneously hypertensive rats, in vitro incubation with recombinant resistin significantly inhibited insulin-stimulated glycogenesis and reduced glucose oxidation. These findings raise the possibility that autocrine effects of resistin in adipocytes, leading to release of other prodiabetic effector molecules from fat and/or paracrine actions of resistin secreted by adipocytes embedded within skeletal muscle, may contribute to the pathogenesis of disordered skeletal muscle glucose metabolism and impaired glucose tolerance.
In industrialized societies, type 2 diabetes is a common cause of morbidity and mortality that is characterized by insulin resistance often in association with central obesity. However, the mechanisms that underlie the widely recognized relationship between obesity and insulin resistance remain to be defined. Although skeletal muscles are quantitatively the most important site of insulin-stimulated glucose disposal (1), adi-pose tissue clearly exerts a major influence on carbohydrate metabolism because changes in body fat mass can have substantial effects on insulin action and glucose tolerance. Moreover, recent studies demonstrating that adipose tissue can secrete a number of molecules that modulate carbohydrate and lipid metabolism strongly suggest that body fat is more than just a passive reservoir for fuel in the form of triglycerides.
Recently, a new hormone produced by fat cells and termed resistin was discovered that could represent an important link between obesity and insulin-resistant diabetes (2)(3)(4). Resistin is a cysteine-rich polypeptide expressed primarily in white adipose tissue that is induced during 3T3-L1 adipogenesis and may also serve as a feedback regulator to inhibit adipocyte generation (2)(3)(4). Message and protein levels of resistin are decreased by fasting and increased by refeeding, possibly in response to changes in insulin levels (2,4). Moreover, Steppan et al. (2) have found that treatment of mice with recombinant resistin can impair glucose tolerance and that administration of anti-resistin antibody improves blood glucose and insulin action in mice with diet-induced obesity (2). Incubation of 3T3-L1 adipocytes with recombinant resistin has also been reported to inhibit insulin-stimulated glucose uptake (2). In addition, resistin mRNA levels can be suppressed by exposure to either fatty acids or ligands for the peroxisome proliferatoractivated receptor ␥ (2). However, in various rodent models of obesity, conflicting results have been reported regarding the effects of systemically administered peroxisome proliferatoractivated receptor ␥ ligands on resistin expression, perhaps because of the fact that these ligands can influence a host of factors that may differentially regulate resistin (5)(6)(7)(8).
Based on measurements of resistin expression in fat tissue in humans and in animals with type 2 diabetes and or obesity, a number of investigators have recently raised questions regarding the potential relevance of resistin to the pathogenesis of insulin resistance. For example, whereas some investigators have found evidence of resistin expression in samples of human subcutaneous and abdominal fat (9, 10), others have found it difficult to detect mRNA for resistin in either adipocytes or subcutaneous adipose tissue isolated from insulin-resistant subjects (11). Expression of resistin in white adipose tissue has also been reported to be significantly decreased in several animal models of obesity-associated insulin resistance (6,12,13). However, the lack of correlation between resistin mRNA levels in isolated adipocytes and insulin resistance does not exclude the possibility that resistin may be contributing to the pathogenesis of disordered carbohydrate metabolism in either liver or skeletal muscle. In Sprague-Dawley rats, intra-arterial infusion of recombinant resistin over a period of 5 h has recently been reported to promote glucose intolerance by impairing insulin action on hepatic glucose metabolism (14). This observation raises the possibility that local secretion of resistin-like molecules into the portal venous circulation might play a role in the pathogenesis of type 2 diabetes. It is also conceivable that paracrine effects of resistin produced by adipocytes embedded deep within skeletal muscle might contribute to the pathogenesis of impaired glucose tolerance in the absence of changes in either circulating levels of resistin or resistin expression in visceral or superficial subcutaneous fat. However, few studies have been performed to directly investigate the effects of resistin on skeletal muscle glucose metabolism or the relationship between circulating levels of resistin and glucose tolerance.
In this study, we investigated the chronic effects of resistin on glucose metabolism in the spontaneously hypertensive rat (SHR), 1 a widely studied animal model of the hypertension metabolic syndrome that is predisposed to insulin resistance due at least in part to a genetic defect in the CD36 fatty acid transporter (15)(16)(17). We have found that transgenic expression of the mouse resistin gene under the control of the aP2 promoter in the SHR induces dyslipidema and increased muscle triglycerides, impairs oxidative and non-oxidative glucose disposal in skeletal muscle, and promotes glucose intolerance in the absence of detectable changes in circulating levels of resistin or insulin. In addition, we observed that recombinant resistin can inhibit glucose oxidation and insulin-stimulated glycogenesis in freshly isolated soleus muscle from non-transgenic SHR. These findings raise the possibility that paracrine actions of resistin secreted by adipocytes embedded within skeletal muscle or autocrine effects of resistin in adipocytes leading to the release of other prodiabetic effector molecules from fat or both may contribute to the pathogenesis of disordered skeletal muscle glucose metabolism and impaired glucose tolerance.

Generation of Transgenic Rats
Animals-The resistin transgene was expressed on the genetic background of the SHR/Ola strain (15)(16)(17). The rats were housed in an air-conditioned animal facility and allowed free access to food and water. Metabolic phenotypes were assessed in male control SHR (n ϭ 10) and male resistin-transgenic SHR (n ϭ 10) after the rats were fed a diet with 60% fructose (K4102.0 diet, Hope Farms, Woerden, The Netherlands) from the age of 8 weeks for 15 days. All of the experiments were performed in agreement with the Animal Protection Law of the Czech Republic (311/1997) and were approved by the Ethics Committee of the Institute of Physiology, Czech Academy of Sciences (Prague, Czech Republic).
Transgenic Strain Derivation-Transgenic SHR were derived by microinjection of zygotes with a mouse resistin cDNA construct that was prepared by reverse transcriptase PCR of RNA from fat tissue of a BALB/c mouse. Resistin primers were designed according to the published resistin sequence (2). The construct contained, in addition to cDNA of the mouse resistin gene, rabbit ␤-globin intron 2, a growth hormone poly(A) signal and the fat-specific aP2 promoter vector (kindly provided by Dr. Farid Chehab, University of California, San Francisco, CA). Microinjections of recombinant DNA into one or both pro-nuclei of fertilized ova were done according to Charreau et al. (18). Transgenic rats were detected by PCR using primers specific for the mouse resistin gene: upstream, 5Ј-tca aca aga agg agc tgt gg-3Ј, and downstream, 5Ј-cca gcc tgt ttt gtt tta ttt-Ј3. The official designation of the new resistintransgenic strain is SHR/Ola-TgN(Retn)201Ipcv (abbreviated herein as the SHR-TG strain).

Phenotypic Analysis
Gene Expression Analysis of Transgenic SHR Rats-Northern blot analysis was used to confirm expression of the mouse resistin transgene and endogenous rat resistin gene in adipose tissue. The probe for Northern analysis was prepared by random primer labeling of the first 580 bp of the mouse resistin gene cut and purified from the transgene construct. Real-time PCR analysis was used to test for possible expression of the mouse resistin transgene in skeletal muscle. The cyclophilin (peptidylprolyl isomerase A) gene was used as an internal control with expression of the mouse resistin transgene relative to cyclophilin being determined in triplicate using the preferred method of Muller et al. (19,20). The cDNA was prepared by reverse transcription of soleus muscle mRNA using random primers followed by real-time PCR amplification using QuantiTect SYBR Green reagents (Qiagen, Inc., Valencia, CA) on an Opticon continuous fluorescence detector (MJ Research, Waltham, MA). The upstream primers were 5Ј-caa atg ctg gac cca aca ca-3Ј (cyclophilin A) and 5Ј-aga agg cac agc agt ctt-3Ј (mouse resistin). The downstream primers were 5Ј-tgc cat cca acc act cag tc-3Ј (cyclophilin A) and 5Ј-tgt cca gtc tat cct tgc a-3Ј (mouse resistin).
Western Blot Analysis-Proteins were extracted from 2 g of epididymal fat. Fat tissue was homogenized with 4 ml of ice-cold extracting buffer (0.3 M Tris-Cl, 0.14 M NaCl, 0.03 M KCl, 1% (w/v) SDS, 1% (v/v) Tween 20, pH 7.4) and 80 l of protease inhibitors (protease inhibitor mixture for mammalian tissues, Sigma) for 1 min on ice. The samples then were agitated for 1 min by vortex and incubated for 10 min at 37°C. Afterward, the samples were centrifuged for 10 min, 5300 ϫ g, at 4°C. The water phase was used as a protein extract. For SDS electrophoresis, the protein extracts were agitated with 2 volumes of sample buffer containing ␤-mercaptoethanol (Laemmli buffer system according to Mini-PROTEAN 3 cell instruction manual from Bio-Rad) and heated to 95°C for 5 min. Samples were loaded on polyacrylamide gel (4% stacking gel and 13% resolving gel), and the electrophoresis was run at 70 V. The load of samples was 30 l. For Western blotting, a wet protein transfer was done using Mini Trans-Blot electrophoretic transfer cell (Bio-Rad) with transfer buffer (25 mM Tris, 192 mM glycine, 20% v/v methanol, 0.5% w/v SDS, pH 8.3) and nitrocellulose membrane (pore size 0.2 m, Bio-Rad). Conditions were 30 V for 16 h at 4°C. The membrane was blocked in 0.01 M phosphate-buffered saline, pH 7.4 (Sigma), containing 3% nonfat dry milk (Sigma) for 30 min at room temperature. Resistin protein was detected with a rabbit anti-mouse resistin antibody (Alpha Diagnostic International, Inc., San Antonio, TX). This antibody does not distinguish between mouse and rat resistin. Horseradish peroxidase-labeled donkey anti-rabbit IgG antibody was used as a secondary antibody. Primary and secondary antibodies were diluted in 0.01 M phosphate-buffered saline containing 1.5% nonfat dry milk. The incubation with the primary antibody was overnight at 4°C and with the conjugate for 1 h at room temperature. The membrane then was treated with the ECL (Amersham Biosciences), and the signal was detected using the Luminiscent Image analysis system (LAS-1000ϩ, Fuji) and quantified by AIDA image analyzer program (Raytest).
Oral Glucose Tolerance Testing-Oral glucose tolerance tests (OGTT) were performed using a glucose load of 300 mg/100 g body weight after 7 h of fasting. Blood was drawn from the tail without anesthesia before the glucose load (0-min time point) and at 30, 60, and 120 min thereafter.
Skeletal Muscle Glycogen Synthesis and Glucose Oxidation-Glycogen synthesis and glucose oxidation were determined in isolated soleus muscle by measuring the incorporation of [ 14 C-U]glucose into glycogen and CO 2 as described previously (21,22). The soleus muscles were attached to a stainless steel frame in situ at in vivo length by special clips and separated from other muscles and tendons and immediately incubated for 2 h in Krebs-Ringer bicarbonate buffer, pH 7.4, that contained 5.5 mM unlabeled glucose, 0.5 Ci/ml [ 14 C-U]glucose, and 3 mg/ml bovine serum albumin (Armour, Fraction V) with or without 250 microunits/ml insulin. After a 2-h incubation, 0.3 ml of 1 M hyamine hydroxide was injected into central compartment of the incubation vessel and 0.5 ml of 1 M H 2 SO 4 was injected into the main compartment to liberate CO 2 . The vessels were incubated for another 30 min, and the hyamine hydroxide was then quantitatively transferred into the scintillation vial containing 10 ml of toluene-based scintillation fluid for counting of radioactivity. For measurement of insulin-stimulated incorporation of glucose into glycogen, glycogen was extracted and glucose incorporation into glycogen was determined as described previously (21,22).
Tissue Triglyceride Measurements-For determination of triglycerides in liver and soleus muscle, tissues were powdered under liquid N 2 and extracted for 16 h in chloroform:methanol, after which 2% KH 2 PO 4 was added and the solution was centrifuged. The organic phase was removed and evaporated under N 2 . The resulting pellet was dissolved in isopropyl alcohol, and triglyceride content was determined by enzymatic assay (Pliva-Lachema, Brno, Czech Republic).

Effects of Recombinant Resistin on Insulin-stimulated Glucose Oxidation and Glycogen Synthesis in Skeletal Muscle
Isolated from the SHR-Glycogenesis and glucose oxidation were measured as described above in soleus muscles isolated from 7-week-old male SHR/Ola (n ϭ 5/each group) fed standard laboratory chow. Recombinant resistin (Alpha Diagnostic International, Inc.) was added to incubation media at a concentration of 600 ng/ml.
Biochemical Analyses-Blood glucose levels were measured by the glucose oxidase assay (Pliva-Lachema) using tail vein blood drawn into 5% trichloroacetic acid and promptly centrifuged. Serum non-esterified fatty acid (NEFA) levels were determined using an acyl-CoA oxidasebased colorimetric kit (Roche Diagnostics). Serum triglyceride concentrations were measured by standard enzymatic methods (Pliva-Lachema). Serum insulin concentrations were determined using a rat insulin radioimmunoassay kit (Amersham Biosciences). Serum levels of leptin were determined using a rat leptin radioimmunoassay kit from Linco Research (St. Charles, MO). Serum resistin concentrations were determined at Linco Research, Inc. with an immunoassay that crossreacts with mouse and rat resistin (Linco Research, Inc.).
Statistical Analysis-All of the data are expressed as means Ϯ S.E. Differences between control and experimental groups were evaluated by paired or non-paired Student's t tests as appropriate. Statistical significance was defined as p Ͻ 0.05.

RESULTS
Transgenic Expression of Resistin-Transgene-positive SHR and transgene-negative controls were identified by genotyping offspring derived from crosses of a founder male with transgene-negative SHR females. In the epididymal fat from the SHR transgene-positive line, Northern blot analysis confirmed the expression of both the mouse resistin transgene and the endogenous rat resistin gene (Fig. 1a). The transgene-negative controls showed expression of only the endogenous rat resistin gene. Fig. 1a shows the presence of the 0.8-and 1.4-kb pair transcripts that are characteristic of endogenous rat resistin in both the transgene-positive SHR and in a transgene-negative control. These two transcripts are similar in size to those previously reported for the rat, which is known to express two resistin transcripts, whereas the mouse expresses only one resistin transcript (4). In the SHR-transgenic line, a single transcript for the mouse resistin transgene could be detected in moderate amounts that is distinct in size from the endogenous rat resistin transcripts and that is slightly larger than the wild type mouse resistin transcript because of its longer 3Ј-untranslated tail. In addition to detecting mRNA for the mouse resistin transgene by Northern blot analysis in epididymal fat from the transgenic strain, we were able to detect low level skeletal muscle expression of the mouse resistin transgene relative to that of cyclophilin by real-time PCR (Fig. 1b). Transgene-negative controls showed no expression of the mouse resistin gene in fat or muscle when tested by either Northern blot analysis or real-time PCR. Western blot analysis of adipose tissue demonstrated greater expression of resistin protein in the transgenepositive rats than in transgene-negative controls (Fig. 1c).
Effects of Resistin Transgene on Body Weight, Serum Phenotypes, and Tissue Lipid Levels-At the time of sacrifice, there were no significant differences between transgene-positive rats and transgene-negative controls with respect to either body weight or epididymal fat weight (Table I). Serum resistin levels were similar between SHR transgene-positive rats and transgene-negative controls as measured by an immunoassay that cross-reacts with mouse and rat resistin (Table I). Serum concentrations of insulin and glucose in transgene-positive rats, 1.2 Ϯ 0.10 nmol/liter and 6.7 Ϯ 0.2 mmol/liter, were similar to those in transgene-negative controls, 1.2 Ϯ 0.13 nmol/liter and 6.8 Ϯ 0. 2 mmol/liter, respectively. In transgene-positive rats, serum levels of leptin, 3.8 Ϯ 0.8 ng/ml, were also not different compared with those in transgene negative controls, 3.9 Ϯ 0.7  ng/ml. In contrast, serum concentrations of NEFA were significantly increased in transgenic animals compared with controls ( Fig. 2a). Increased serum NEFA levels were also observed in the transgene-positive rats during the oral glucose tolerance test (Fig. 2a). In the transgene-positive rats, the increased serum NEFA levels were associated with increased muscle triglycerides (Fig. 2b), whereas no differences in hepatic triglycerides were observed between the transgene-positive rats (16.4 Ϯ 0.7 mol/g) and the transgene-negative controls (16.4 Ϯ 0.6 mol/g).

Effects of Resistin Transgene on Oral Glucose Tolerance-
The resistin-transgenic rats displayed impaired oral glucose tolerance (Fig. 3) with the area under the glucose tolerance curve being significantly greater in the transgene-positive line than in the transgene-negative controls (831 Ϯ 13 mmol/2 h and 760 Ϯ 12 mmol/2 h, respectively; p Ͻ 0.005) (Fig. 3). These results are consistent with the studies of Steppan et al. (2) in which systemic administration of recombinant resistin was found to impair oral glucose tolerance in C57BL/6J mice. However, in contrast to the studies of Steppan et al. (2) in which impaired oral glucose tolerance induced by injection of recombinant resistin was associated with increased serum levels of resistin, the current studies demonstrate that transgenic expression of mouse resistin on the SHR background can impair oral glucose tolerance in the absence of detectable changes in circulating resistin.
Effects of Resistin Transgene on Skeletal Muscle Glucose Metabolism-In soleus muscle isolated from transgenic SHR expressing the mouse resistin gene, glycogenesis and glucose oxidation were significantly reduced in both the presence and absence of insulin (Fig. 4, a and b). Thus, both non-oxidative and oxidative glucose metabolism were impaired in skeletal muscle of transgenic rats compared with controls. These findings are in accord with the recent studies of Moon et al. (23) in which recombinant resistin was found to inhibit glucose uptake in cultured L6 skeletal muscle cells in both the presence and absence of insulin. However, in the studies of Moon et al. (23), the effects of recombinant resistin on metabolic pathways of glucose disposal were not investigated and the extent to which the effects of recombinant resistin on glucose metabolism in L6 cells resemble those in skeletal muscle tissue is unknown. The current findings indicate that transgenic expression of resistin can impair both of the main pathways that mediate skeletal muscle glucose disposal even in isolated tissue that has not been previously exposed to increased circulating levels of resistin in vivo or to additional recombinant resistin in vitro.
Effects of Recombinant Resistin on Skeletal Muscle Glucose Metabolism-Incubation of isolated soleus muscle from nontransgenic SHR with recombinant resistin significantly im-paired insulin stimulated glycogenesis (Fig. 5a). Skeletal muscle glycogenesis was significantly lower in the presence of recombinant resistin plus insulin than in the presence of insulin alone. Although recombinant resistin also showed a tendency to reduce basal glycogenesis in the absence of insulin, the effect did not achieve statistical significance (Fig. 5a). Incubation of soleus muscle from non-transgenic SHR with recombinant resistin significantly reduced basal glucose oxidation (no insulin) and caused a borderline decrease (p ϭ 0.07) in glucose oxidation in the presence of insulin (Fig. 5b). These findings in SHR soleus muscle incubated with recombinant resistin show remarkable similarity to the results in soleus muscle isolated from transgenic SHR that express the mouse aP2 resistin transgene. DISCUSSION The discovery of resistin initially provoked great interest in the potential role of resistin-related molecules in the pathogenesis of insulin resistance and type 2 diabetes. More recently, a number of reports have been published that might seem to raise doubts regarding the possibility of an important relationship between resistin and various metabolic disturbances that are characteristic of obesity and type 2 diabetes. For example, some investigators have noted that in humans, little or no correlation exists between resistin mRNA levels in adipose tissue or adipocytes and diabetes, insulin resistance, or body mass index (10,11,24). In a variety of animal models of obesity that are associated with insulin resistance, expression of resistin in white adipose tissue has also been reported to be significantly decreased (6,12,13), prompting some investigators to conclude that "resistin seems to be at work to improve rather than cause insulin resistance in obesity" and even to suggest that "the use of the term resistin is not appropriate in this regard" (13). However, given the complex nature of multifactorial disorders similar to insulin resistance and type 2 diabetes, the often unpredictable relationship between mRNA levels and protein levels as well as the lack of knowledge regarding potential feedback relationships between insulin resistance and resistin expression, it is premature to discount a potential role for resistin in the pathogenesis of insulin resistance in either humans or in animals (25).
In the current studies, we have found that transgenic expression of mouse resistin under control of the aP2 promoter on the genetic background of the SHR induces increases in serum and skeletal muscle lipid levels, impairs oxidative and non-oxidative glucose disposal in skeletal muscle, and promotes glucose intolerance in the absence of detectable changes in circulating levels of resistin, insulin, or leptin. Similar results were obtained with transgenic expression of mouse resistin under the control of a universal elongation factor-1␣ promoter. 2 Thus, the current findings are probably related to the effects of transgenic expression of resistin rather than the effects of transgenic expression the aP2 promoter. In addition, we observed that recombinant resistin can attenuate glucose oxidation and greatly inhibit insulin-stimulated glycogenesis in freshly iso-lated soleus muscle from non-transgenic SHR. These findings are consistent with the studies of Steppan et al. (2) in which intraperitoneal injection of recombinant resistin was found to acutely impair glucose tolerance in C57BL/6J mice and suggest that, at least under the environmental and genetic circumstances of the current study, resistin may contribute to impaired glucose tolerance by way of effects on skeletal muscle glucose metabolism.
Although the current studies have focused on the effects of resistin on skeletal muscle glucose metabolism, they should not be taken to imply that the glucose intolerance observed in the SHR resistin-transgenic strain is only, or even largely, a consequence of skeletal muscle insulin resistance. In a comprehensive series of experiments using the pancreatic insulin clamp technique in conscious Sprague-Dawley rats, Rajala et al. (14) recently found that intra-arterial infusion of recombinant resistin over a period of 5 h can impair insulin action on hepatic glucose production. Other investigators have reported that recombinant resistin can inhibit insulin-stimulated glucose uptake in cultured 3T3-L1 adipocytes and in L6 skeletal muscle cells (2,23). These studies, together with the current findings in isolated skeletal muscle preparations from transgenic and non-transgenic rats, suggest that resistin may contribute to the pathogenesis of insulin resistance through effects on multiple target tissues. The relative importance of these effects of resistin in different target tissues might also depend on the environmental and genetic circumstances under which they are studied.
In the experiments of Rajala et al. (14) in Sprague-Dawley rats, the short term intra-arterial infusion of recombinant resistin acutely impaired hepatic gluconeogenesis without significantly affecting peripheral glucose uptake. However, short term infusions of recombinant resistin may not necessarily reflect the chronic metabolic effects of resistin in vivo and do not have the potential to reveal autocrine or paracrine meta-  open bars). b, glucose oxidation in soleus muscle isolated from the SHR-TG and SHR strains. *, p Ͻ 0.05; **, p Ͻ 0.01 versus identically treated soleus muscle from SHR transgene-negative controls. #, p Ͻ 0.05 compared with soleus muscle from the same strain without insulin incubation.
FIG. 5. Glucose metabolism in isolated soleus muscle incubated with recombinant resistin. a, glycogenesis in soleus muscle isolated from transgenenegative SHR and incubated in the presence and absence of recombinant resistin and insulin. b, glucose oxidation in soleus muscle isolated from transgene-negative SHR and incubated in the presence and absence of recombinant resistin and insulin. *, p Ͻ 0.05; **, p Ͻ 0.01 compared with soleus muscle not exposed to recombinant resistin but otherwise identically treated. #, p Ͻ 0.05; ##, p Ͻ 0.01 compared with soleus muscle not incubated with insulin but otherwise identically treated.
bolic effects of resistin that may occur independently of changes in circulating levels of resistin. Thus, notwithstanding the impressive studies of Rajala et al. (14), it is possible that chronic exposure of skeletal muscle to resistin from the circulation or to resistin that is locally secreted by adipocytes embedded within muscle tissue may impair skeletal muscle glucose disposal. In addition, although short term infusions of resistin failed to affect skeletal muscle glucose metabolism in Sprague-Dawley rats studied under the conditions of Rajala et al. (14), it is possible that acute or chronic exposure to resistin might influence skeletal muscle glucose disposal in Sprague-Dawley rats studied under other environmental or dietary conditions or in other strains that are genetically susceptible to insulin resistance. In the current studies, the results of two complementary approaches indicate that chronic exposure to resistin in vivo as well as acute exposure to resistin in vitro can impair skeletal muscle glucose metabolism in fructose-fed SHR, a well known animal model of the metabolic syndrome (15,26,27). This model resembles the metabolic syndrome in humans in that the pathogenesis of disordered glucose metabolism depends on the interaction of multiple environmental and genetic factors (15, 16, 26, 28 -30).
We originally designed the resistin-transgenic SHR with the goal of studying the metabolic effects of chronically increased plasma levels of resistin and were therefore surprised to observe alterations in glucose metabolism in the absence of detectable changes in circulating levels of resistin. Assuming that our single time point measurements of serum resistin accurately reflect circulating levels of resistin throughout the day and night, the current findings suggest that resistin may be modulating skeletal muscle glucose through paracrine or autocrine mechanisms or both. Moreover, even if serum resistin levels had been found to be increased in the transgenic strain, the observation of impaired glucose metabolism in isolated skeletal muscle studied in the absence of resistin in the media would still be intriguing. Such findings could imply the existence of some type of memory effect in which prior exposure to resistin in vivo is sufficient to impair skeletal muscle glucose metabolism studied in the absence of additional resistin in vitro.
Although we could not detect any differences in circulating levels of resistin, insulin, or leptin between resistin-transgenic SHR and non-transgenic SHR, we did observe significantly increased lipid levels in the serum and skeletal muscle of the transgenic strain. In light of the well known relationship between skeletal muscle insulin resistance and increased serum and skeletal muscle lipids (31,32), the current findings should motivate future studies into the effects of resistin on lipid metabolism as a potential link between adipocyte resistin levels and skeletal muscle glucose metabolism. Local action of resistin in fat tissue leading to release of other effector molecules from adipocytes that can modulate skeletal muscle glucose metabolism could constitute a type of autocrine mechanism that might promote impaired glucose tolerance in the absence of changes in serum resistin levels. In this regard, it will also be of interest to determine the extent to which the metabolic effects of resistin are being influenced by enhanced susceptibility to disordered lipid metabolism engendered by the genetically defective CD36 fatty acid transporter present in the SHR strain (15,16). The observation of impaired glucose metabolism in skeletal muscle isolated from transgenic SHR that exhibited low level expression of the mouse resistin transgene as well as in skeletal muscle from non-transgenic SHR that was exposed to recombinant resistin in vitro deserves further comment. These findings raise the possibility that paracrine effects of resistin released from adipocytes embedded in skeletal muscle might also be contributing to resistin-induced alterations in skeletal muscle glucose metabolism. This mechanism would be of particular interest given recent studies demonstrating a correlation between insulin resistance and the amount of adipose tissue beneath the fascia of skeletal muscle in the mid-thigh (33). In future studies, it may be of interest to investigate the metabolic effects of resistin transgenes that are expressed under the control of other restricted promoters such as the one for muscle creatine kinase.
In summary, in the SHR, we have found that transgenic expression of the mouse resistin gene under control of the aP2 promoter can impair glucose tolerance and inhibit oxidative and non-oxidative glucose metabolism in skeletal muscle in the absence of increased circulating levels of resistin. These findings raise the possibility that resistin may promote glucose intolerance through mechanisms that do not depend on changes in circulating levels of resistin and that could involve autocrine effects of resistin in fat, paracrine effects of resistin in skeletal muscle, or both (Fig. 6). Finally, regardless of the relative contributions of plasma or tissue levels of resistin or of resistin-induced effector molecules to insulin resistance, the current findings should motivate additional studies of the metabolic effects of resistin on lipid metabolism and skeletal muscle glucose metabolism under a variety of other environmental and genetic circumstances.