OBJECTIVE—Since vascular dysfunction is a main trait of obese subjects, in the present study we evaluated the vascular impact of resistin, a recently discovered hormone markedly increased in obesity.
RESEARCH DESIGN AND METHODS—We performed our analysis on aortic and mesenteric segments from young and old C57BL/6 mice and on cultured endothelial cells. Resistin-induced vascular effect was evaluated in vitro and in vivo. Molecular analyses were performed by immunoprecipitation and Western blotting.
RESULTS—Recombinant murine resistin did not induce changes in either basal vascular tone or phenylephrine-induced vascular contraction. In contrast, both in vivo and in vitro administration of resistin significantly impaired dose-dependent insulin-evoked vasodilation by reducing endothelial nitric oxide synthase (eNOS) enzymatic activity. This effect of resistin was selective for insulin vascular action, since vasodilatation induced by increasing doses of acetylcholine or nitroglycerin was not influenced by the hormone. Molecular analysis of endothelial cells further detailed resistin-induced vascular resistance by showing impairment of insulin-evoked AKT and eNOS phosphorylations after exposure to resistin. Even this latter abnormality is selective of insulin signaling since AKT/eNOS phosphorylations are normally activated during acetylcholine stimulation. More important, the resistin-induced endothelial dysfunction depends on resistin's ability to alter insulin receptor substrate (IRS)-1 tyrosine/serine phosphorylation and its consequent interaction with phosphatidylinositol 3-kinase.
CONCLUSIONS—Our results demonstrate that resistin is able to induce a selective vascular insulin resistance-impairing endothelial IRS-1 signaling pathway that leads to eNOS activation and vasodilation.
The prevalence of obesity has risen dramatically in Western societies, and insulin resistance resulting from increased adipose tissue mass has been identified as a key factor for the increased cardiovascular risk present in obese subjects (1,2). Insulin resistance is defined as an impaired sensitivity of target tissues to the biological action of insulin (3,4). In particular, in obesity insulin resistance affects not only the classical target tissues of the hormone, such as skeletal muscle, but also the nonclassical actions of insulin, such as those on hemodynamics (5). In fact, obese subjects show a reduced insulin-stimulated skeletal muscle glucose uptake as well as an impaired insulin-evoked vasodilation (5). These observations have suggested that the pathophysiological mechanisms linking obesity to the development of cardiovascular diseases could go beyond the classical metabolic derangements. To unveil these other potential mechanisms leading to increased cardiovascular risk in obesity, in the last decade much effort has been made to understand the interaction between insulin resistance and vascular function, with particular emphasis on adipocyte-derived hormones and their effects on vascular homeostasis (6,7). Actually, it is well known that adipose tissue regulates skeletal muscle insulin sensitivity via a number of circulating adipocyte-derived hormones, such as leptin, adiponectin, and tumor necrosis factor-α (8). It has become clear that the majority of these adipocyte-derived hormones also have important vascular effects (7).
During a search for adipocyte targets of thiazolidinediones, a class of insulin sensitizers (9,10), Steppan et al. (11) discovered a novel adipocyte-derived hormone called resistin, which was suppressed by thiazolidinedione treatment. Resistin, expressed exclusively in white adipose tissue, is a member of the newly discovered family of cysteine-rich proteins called resistin-like molecules (12). Resistin is secreted to the medium by cultured adipocytes and circulates in plasma, indicating that it is a secretory product of adipose tissue. Initial studies have demonstrated that obesity induced by a high-fat diet, or by mutation of the leptin gene (ob/ob mice) or of the leptin receptor gene (db/db mice), is associated with elevated circulating resistin plasma levels (11). Intraperitoneally administered resistin augments blood glucose and plasma insulin levels and limits the hypoglycemic response to insulin infusion (11). Furthermore, resistin suppresses insulin-stimulated glucose uptake in cultured adipocytes, and this effect is prevented by exposure to anti-resistin antibodies. Finally, treatment with these antibodies decreases blood glucose and improves insulin sensitivity in obese mice (11). All these data suggest that resistin could contribute to the insulin resistance observed in obese subjects by decreasing insulin sensitivity at least in skeletal muscle tissue (12). Nevertheless, it is still unexplored whether resistin could affect nonclassical actions of insulin, such as those on vascular function.
The aim of the present study was to explore whether resistin, used at the same dose previously described to induce insulin resistance at the skeletal muscle level, is also able to interfere with vascular function and, more strictly, with insulin-evoked vasorelaxant effects. Moreover, in addition to functional studies, a more detailed molecular analysis of the interplay between resistin and insulin at vascular level has been assessed.
RESEARCH DESIGN AND METHODS
Animal studies.
The studies were performed on male C57BL/6 mice aged 8–12 weeks (n = 36) and aged 6 months (n = 22). All experimental procedures were in accordance with the guidelines of our institution regarding research in animals.
On the day of the experiments, mice were weighed, sedated with CO2, and then decapitated. Thoracic aorta and mesenteric arteries were dissected from each mouse, cleaned of adhering perivascular tissue, and placed in cold Krebs-Henseleit bicarbonate buffer solution with the following composition (mmol/l): NaCl 118.3, KCl 4.7, CaCl2 2.5, MgSO4 · 7H2O 1.2, KH2PO4 1.2, NaHCO3 25, and glucose 5.6.
Chronic resistin treatment.
In some mice, resistin (1 μg/day) or vehicle was administered chronically for 28 days by osmotic minipumps implanted subcutaneously (Alzet). This dose of resistin allows mice to reach resistin plasma levels comparable with those observed in hyperglycemic patients (13).
Blood pressure and heart rate were evaluated by the tail-cuff technique at 7, 14, 21, and 28 days (Visitech Systems) after a period of acclimation. At the end of treatment, blood pressure was measured in the right carotid artery via cannulated micromanometers. Plasma levels of resistin were measured using an enzyme-linked immunosorbent assay kit for mouse (Linco Research).
Vascular reactivity studies.
Vascular reactivity studies have been performed on aortas, as previously described (14). Mesenteric arteries were mounted on a micromyograph, as previously described (15). On aortic and mesenteric rings preconstricted with phenylephrine (10−6 mol/l; Sigma), we evaluated the vascular effect evoked by resistin (5 nmol/l). The dose used in these experiments and in those in vascular cells is in the same range as the peak plasma level reached in in vivo studies (16). Moreover, contraction to phenylephrine (10−9 − 10−6 mol/l) and subsequent vasodilation to insulin (10−9 − 10−7 mol/l), acetylcholine (10−9 − 10−5 mol/l), and nitroglycerin (10−9 − 10−6 mol/l) were tested in vessels before and after 30 min of incubation with 5 nmol/l resistin or 300 μmol/l l-NG-nitro-l-arginine methyl ester (l-NAME). Subsequently, to further clarify the role of resistin on vascular insulin resistance, we extended our studies on mice with chronic resistin administration. Therefore, at the end of chronic infusion, we evaluated, on aortic and mesenteric rings preconstricted with phenylephrine, the vasorelaxation responses evoked by increasing doses of insulin (10−9 − 10−7 mol/l), acetylcholine (10−9 − 10−5 mol/l), and nitroglycerin (10−9 − 10−6 mol/l). The maximal contraction evoked by phenylephrine was considered baseline when subsequent vasorelaxations were evoked. Vasorelaxation was expressed as percent reduction in contraction (the maximal vasorelaxation attained with papaverine being 100% vasorelaxation).
Evaluation of endothelial nitric oxide synthase enzymatic activity.
Endothelial nitric oxide synthase (eNOS) activity by Ca2+-independent stimuli was evaluated measuring the conversion of 14C-arginine to 14C-citrulline in vitro in a calcium-free buffer; meanwhile, calcium was present in the buffer when Ca2+-dependent agonists were challenged. Aortas were treated with insulin (50 nmol/l for 10 min), acetylcholine (1 μmol/l for 10 min), resistin (5 nmol/l for 40 min), and l-NAME (300 μmol/l for 40 min) alone or in combination and then rapidly frozen in liquid nitrogen and homogenized in lysis buffer (in mmol/l: 25 Tris-HCl, pH 7.4; 1 EDTA; 1 EGTA; 0.1 NaF; and 0.1 Na3VO4). A total of 50 μg of protein extract were incubated at room temperature for 30 min with reaction buffer (50 mmol/l Tris-HCl, pH 7.4; 0.3 μmol/l BH4; 1 μmol/l FAD; 1 μmol/l FMN; 1 mmol/l NADPH; and 0.5 μCi 14C-arginine, with or without 1 mmol/l CaCl2). The reaction was blocked on ice for 10 min with stop buffer (20 mmol/l HEPES and 5 mmol/l EDTA). The reaction mix was applied to a 1-ml column containing Dowex AG 50 × 8 resin that had been equilibrated with stop buffer. 14C-citrulline was eluted twice with 2 ml of water, and radioactivity was determined by scintillation counting.
Cell cultures.
Since the vascular effect of resistin was mainly targeted to the endothelial function, we focused our molecular analysis on endothelial cells. Actually, vascular tissues are composed mostly of smooth muscle cells, thus diluting endothelial signaling in the background noise of smooth muscle. Then, we choose to use human endothelial cells (HAEC and HUVEC; Biowhittaker, Verviers, Belgium) placed in endothelial cell growth medium containing 10% fetal bovine serum on 60-mm plates and studied them before cell confluence at passages four to six. Plates were washed with PBS and starved for 24 h in endothelial cell basal medium containing 1% fetal bovine serum.
Immunoblot and immunoprecipitation.
Endothelial cells were incubated with insulin (50 nmol/l for 10 min) and/or resistin (5 nmol/l for 40 min). Proteins were extracted as previously described. Four milligrams of protein extract were subjected to immunoprecipitation with anti–insulin receptor substrate (IRS)-1 antibodies (8 μg/ml; UpState) and separated on 6% SDS-PAGE. Gels were electroblotted on nitrocellulose membrane (Bio-Rad), filters were blocked 1 h at room temperature in Tris-buffered saline with Tween 20 buffer (20 mmol/l Tris-HCl, 150 mmol/l NaCl, and 0.1% Tween 20) containing 3% nonfat dry milk. Blots were then incubated overnight at 4°C with anti–IRS-1 (1.5 μg/ml; UpState), anti-p85 (1.5 μg/ml; UpState), anti–phospho-IRS (pY 632 or pS307, 1:200; Santa Cruz Biotechnologies), or anti-ubiquitin (1:100; Sigma).
A total of 35 μg of proteins in cell lysates were separated by 8% SDS PAGE (Bio-Rad). Filters were blocked 1 h at room temperature in Tris-buffered saline with Tween 20 buffer containing 3% nonfat dry milk then with anti–phospho-(Ser473)Akt (1:500; Cell Signaling) or phospho-(Ser1,177) eNOS (1:500; Cell Signaling). Protein detection was performed with an enhanced chemiluminescence kit (Amersham). To normalize for protein quantity, membranes were stripped and incubated with anti-Akt (1:1,000; Cell Signaling) and anti-eNOS (1 μg/ml; Transduction Laboratories), respectively. Proteins were revealed as above. The intensity of the bands was quantified by scanning densitometry using NIH Image 1.61 software.
Statistical analysis.
All data shown are presented as means ± SE. Statistical analysis was performed by two-way ANOVA followed by a Bonferroni post hoc test. A P value <0.05 was assigned statistical significance.
RESULTS
Resistin selectively blunted insulin-induced vasorelaxation in both young and old mice.
In 8- to 12-week-old mice, resistin did not alter aortic and mesenteric passive wall tensions and phenylephrine-induced contractions (aortic max contraction: 1,080 ± 30 mg vs. 1,100 ± 20 mg, P = NS; mesenteric max contraction: 700 ± 20 mg vs. 685 ± 30 mg, P = NS). Interestingly, in both aortic (Fig. 1B) and mesenteric arteries (data not shown) resistin blunted insulin-evoked dose-dependent vasorelaxation, thus suggesting that the impairment of insulin induced-vasodilation was due to a selective interaction between resistin and insulin. Furthermore, resistin did not modify the vasorelaxation responses evoked by increasing doses of acetylcholine (Fig. 1C) or nitroglycerine (data not shown.). The administration of l-NAME impaired insulin vasodilation in both aorta (Fig. 1B) and mesenteric vessels (data not shown), and in this condition resistin did not exert any effect. To evaluate the influence of age on resistin vascular effects, we studied 6-month-old mice. Even in these mice, resistin impaired insulin vasorelaxation (Fig. 1D) without influencing other vascular responses such as that evoked by acetylcholine (Fig. 1E).
Chronic resistin administration impaired insulin-induced vasorelaxation.
To test the effect of resistin in vivo, we infused the hormone for 4 weeks. As shown in Table 1, chronic resistin administration increased plasma levels of the hormone and did not significantly influence blood pressure levels and heart rate compared with vehicle-treated mice. At the end of treatment, the evaluation of vascular reactivity showed that resistin did not modify phenylephrine-induced contraction (Table 1). However, even in vivo resistin administration impaired the vasorelaxant response evoked by increasing doses of insulin without influencing acetylcholine (Fig. 2) or nitroglycerine vascular responses (data not shown).
Resistin selectively impaired insulin-induced eNOS enzymatic activity.
Since insulin-evoked vasodilation was dependent on endothelial NO production, we explored whether resistin was able to affect eNOS enzymatic activity in aortic rings. Our results demonstrated that preincubation with acetylcholine or insulin markedly stimulated enzymatic activity, and this latter effect was impaired by l-NAME exposure. More interestingly, pretreatment with resistin, while not affecting basal eNOS enzymatic activity, significantly blunted only insulin's effect on eNOS activity (Fig. 3). This evidence suggests that resistin can interfere selectively with the insulin signaling that leads to activation of eNOS.
Resistin selectively altered insulin-induced endothelial Akt/eNOS phosphorylations.
Since the vascular effect of resistin was mainly targeted to the endothelial function, we focused our molecular analysis on endothelial cells. Thus, we analyzed the phosphorylation of eNOS and of its activator Akt. As shown in Fig. 4, both insulin and acetylcholine increased eNOS phosphorylation on Ser1,177 and Akt phosphorylation on Ser473, though to a different extent. In contrast, resistin, per se, did not influence eNOS and Akt phosphorylations compared with control conditions. More important, pretreatment with resistin was able to markedly impair insulin-evoked eNOS-Ser1,177 and Akt-Ser473 phosphorylations, although it did not modify the protein phosphorylation effects of acetylcholine.
Resistin impaired endothelial insulin–induced IRS-1/PI3K interaction.
It is well known that insulin can induce Akt phosphorylation through the activation of the IRS-1/phosphatidylinositol 3-kinase (PI3K) cascade that occurs after the physical interaction between IRS-1 and the PI3K subunit p85. As expected, the immunoprecipitation of IRS-1 allowed us to observe on the same blot the concomitant presence of p85, indicating that insulin is able to stimulate the interaction between IRS-1 and p85 (Fig. 5A). More interestingly, endothelial exposure to resistin significantly decreased IRS-1 intracellular levels and markedly impaired the ability of insulin to induce IRS-1/PI3K interaction. This decreased molecular interaction during resistin plus insulin stimulation was not simply ascribable to the reduced intracellular levels of IRS-1 induced by resistin, but it resulted also by an impaired ability of IRS-1 to recruit p85 as evidenced by the analysis of the p85–to–IRS-1 ratio that was significantly decreased compared with that observed in cells stimulated with insulin alone (Fig. 5B). These data suggest that resistin is able to blunt the insulin signaling pathway acting on both IRS-1 levels and its ability to activate PI3K.
Resistin increased endothelial IRS-1 ubiquitination.
The decrease in IRS-1 protein amount could result from either reduction of its expression or its increased degradation. In our experimental settings, exposure to resistin lasts just 50 min, so that an effect of resistin on genetic expression of IRS-1 is less plausible. Thus, we focused our attention on the molecular mechanisms involved in protein catabolism, in particular on ubiquitination, which is the first step of degradation process. As shown in Fig. 5A, immunoprecipitated IRS-1 was also recognized by anti-ubiquitin antibodies. Interestingly, insulin exposure is able to slightly increase the positiveness to ubiquitin antibodies, but, more interestingly, treatment with resistin dramatically increases this phenomenon.
Resistin altered insulin-evoked endothelial Tyr632/Ser307–IRS-1 phosphorylations.
IRS-1 contains several tyrosine and serine/threonine phosphorylation sites that have opposite effects on IRS-1 ability to transduce intracellular signaling pathway. In fact, IRS-1 is activated when phosphorylated in tyrosine and is inhibited when phosphorylated in serine. As shown in Fig. 5A, in our cellular system, the analysis of immunoprecipitated IRS-1 demonstrated that, as expected, insulin induced IRS-1 Tyr632 phosphorylation, while resistin alone was not able to influence it. Interestingly, in presence of resistin, insulin-induced IRS-1 Tyr632 phosphorylation was markedly blunted. This latter result cannot be ascribed only to the reduced intracellular IRS-1 levels, since the pY632-IRS-1–to–IRS-1 ratio, which adjusts for protein amount, was significantly reduced (Fig. 5C). Moreover, the analysis of IRS-1 Ser307 phosphorylation demonstrated that neither insulin nor resistin alone were able to influence it. Nevertheless, in presence of resistin IRS-1 Ser307 phosphorylation was significantly increased by insulin.
DISCUSSION
Our data demonstrate that the resistin also has effects on vascular tissues, selectively impairing insulin-evoked vasorelaxation. This vascular action of resistin is mainly realized in endothelium, by altering the insulin-induced IRS1-signaling leading to Akt/eNOS phosphorylations and, consequently, blunting NO-dependent vasorelaxation (15,17).
It is becoming clear that increased cardiovascular risk associated with obesity could be not only related to metabolic derangements, the typical trait of the disease, but also be a consequence of the vascular dysfunction observed in overweight subjects (18). Indeed, several studies have been recently described in which obesity is able to induce vascular abnormalities independently of diastolic blood pressure, insulin, and lipid abnormalities (13,20). An important feature of obese subjects is adipose mass increase, mainly of the white visceral adipose tissue. The latest discovered hormone produced in this adipose tissue is the peptide resistin, so named for its ability to provoke resistance to insulin-induced skeletal muscle glucose uptake (11).
The plasma concentration of resistin in patients with insulin resistance remains to be carefully defined, even if preliminary reports indicate that mean circulating resistin levels in obese subjects are increased about fourfold compared with lean subjects (21,22). The present study reports that levels of resistin comparable with those observed in obese subjects, besides their deleterious impact on metabolic targets, also exert a direct negative influence on vascular tissue. In particular, our results demonstrate that resistin selectively blunts the vasodilation induced by insulin on both conductance and resistance vessels. The selectivity of resistin with regard to insulin vascular action finds a strong support in those studies that have demonstrated that resistin selectively antagonizes insulin biologic action also at the metabolic level (11,23–25). Indeed, some studies (16,26) have been unable to reveal resistin impairment of insulin effects. With regard to this controversy, our results confirm, and extend at vascular level, that resistin is an adipocyte hormone that selectively antagonizes insulin action. This vascular effect of resistin is observed at either young or old age, thus suggesting that an increase in this adipocyte hormone is deleterious in the whole lifespan. More important, our data obtained with chronic infusion of resistin in vivo demonstrate that hyperresistinemia is able to induce a status of vascular insulin resistance. This biological action supports the idea of resistin as a cardiovascular risk factor. On this note, epidemiological studies so far have been inconclusive. A recent study (27) did not find an association between resistin levels and cardiovascular events in Korean diabetic patients, and Pilz et al. (28) related resistin levels to inflammatory processes and renal function but not to the incidence of some cardiovascular diseases. In contrast, several other studies (29,30) found that resistin levels are higher in different cardiovascular conditions. More important, the same Pilz et al. (28) found an increase in mortality in hyperresistinemic patients, and Lubos et al. (31) demonstrated that resistin levels predict cardiovascular mortality in patients with coronary artery disease. Our study found a mechanism through which resistin can affect cardiovascular physiology and contribute to human disease, thus extending recent observations (32) showing that vascular cells are a target of resistin also independently of metabolic effects.
It has been clearly demonstrated that the vasodilation evoked by insulin is mediated by NO released from the endothelium by eNOS (33,34), and we observed that resistin decreased eNOS enzymatic activity. Thus, we have extended our analysis to the molecular mechanisms involved in eNOS activation. eNOS activity can depend either on Ca2+ mobilization (35) or eNOS phosphorylation at different sites, among which Ser1,177 has been the most extensively characterized (15,16,36). On this issue, insulin has become a prototype of endothelial agonist, which activates eNOS Ser1,177 phosphorylation by an increased phosphorylation of Akt/protein kinase B, a downstream target of the intracellular insulin–evoked signaling pathway (16). In agreement with the reduction of insulin vasodilatation by resistin, the hormone also was able to blunt insulin-evoked Akt and eNOS phosphorylations, while it was unable to influence eNOS/Akt phosphorylations evoked by acetylcholine. This evidence demonstrates, at the molecular level, that resistin selectively impairs the effect of insulin on eNOS enzymatic activity and indicates a mechanism through which resistin can reduce insulin-evoked vasorelaxation. The fact that resistin selectively impairs insulin-evoked but not acetylcholine-evoked Akt/eNOS phosphorylations, as well as vasorelaxation, suggests that these effects are dependent on an interference exerted upstream of Akt in a more specific insulin postreceptor signaling event. Our results are in agreement with a recent study (37) reporting that the insulin resistance induced by resistin on rat skeletal muscle cells involves interference on Akt phosphorylation.
It is well known that insulin can induce Akt activation through the IRS/PI3K cascade. In response to insulin, IRS-1 recruits a number of SH2-containing signal transducers including PI3K, which in turn, stimulates its main downstream effectors such as Akt (38) and the atypical protein kinase C, driving the insulin action on GLUT4 translocation and glucose transport (39). Here we show that, in endothelial cells, resistin reduces the ability of IRS-1 to recruit the PI3K subunit p85, clearly indicating that resistin induces Akt-dependent endothelial NO dysfunction through the inhibition of the IRS-1 signaling pathway. Moreover, we observed that IRS-1 itself is present in a lower amount in cells challenged with insulin and pretreated with resistin. This finding suggests that resistin interferes with the insulin-stimulated IRS-1–dependent signaling pathway, acting both on the IRS-1 protein and on its ability to activate PI3K. The decrease in IRS-1 cellular content could result from its degradation or from reduction in its expression (38). However, our data are obtained after acute resistin stimulation. Since the activation of the expression machinery needs a longer time, our data suggest that the effect of resistin on IRS-1 expression could be ruled out. In our experimental system, we show that immunoprecipitated IRS-1 was also recognized by anti-ubiquitin antibodies in cells stimulated with insulin and pretreated with resistin, indicating that resistin induces IRS-1 ubiquitination, the first step of the degradation process. In fact, it has been reported that degradation of IRS-1 through the ubiquitin-proteasome pathway is a physiological event that is a part of the negative-feedback control mechanism induced by insulin that terminates its action (40) and that occurs, in vitro, after at least 2 h of insulin stimulation (38). Accelerated IRS-1 degradation through the ubiquitin proteasome pathway has been ascribed among the several mechanisms observed in animal models of insulin resistance (41). Thus, our findings clearly suggest that the negative vascular effect induced by resistin is mediated by the ability of this latter to interfere with the proper IRS-1–mediated signaling pathway, accelerating the physiological mechanism through which insulin itself abolishes its effect.
In our system, we show that resistin is able to impair insulin-induced IRS-1 Tyr632 phosphorylation. The role of IRS-1 Tyr phosphorylation has been largely elucidated (42). In particular, it has been described that the phosphorylation of Tyr632 generates the major docking sites for PI3K (43). Thus, our finding suggests that this resistin-mediated phosphorylation impairment is responsible for the lack of interaction between IRS-1 and PI3K. More interestingly, here we show that in presence of insulin, resistin is able to induce IRS-1 Ser307 phosphorylation. An increased serine phosphorylation of IRS-1 may make this form of IRS a “poor” insulin receptor substrate (44) or result in its dissociation from the insulin receptor and trigger proteasome-dependent degradation (45). Thus, our observation on IRS-1 Ser307 phosphorylation and ubiquitination can be correlated in light of this phenomenon. Our results support the theory according to which a delicate balance between “positive” IRS-1 tyrosine phosphorylation versus “negative” IRS-1 serine phosphorylation could regulate IRS-1 functions, and modification of this equilibrium could lead to pathological situations.
Thus, in this study we describe that in endothelial cells resistin, altering the balance between the two phosphorylated forms of IRS-1, is able to impair the IRS-1–dependent activation of the PI3K/Akt pathway with consequent inhibition of eNOS activity and control of the vascular tone. Our results could lie beneath the finding that thiazolidinediones, drugs that inhibit the adipocyte generation of resistin, improve not only insulin-induced skeletal muscle glucose metabolism but also insulin resistance-induced endothelial dysfunction (9,10). Indeed, it has been recently demonstrated that rosiglitazone treatment improves downstream insulin-stimulated IRS-1 tyrosine phosphorylation and increases the association of the p85 regulatory subunit of PI3K with IRS-1 as well as PI3K activity (46). Since endothelial NO has a crucial role not only in modulating vascular tone but also in antiatherogenic protection (47) by inhibiting inflammation, oxidation, vascular smooth muscle cell proliferation, and migration, we can speculate that endothelial NO dysfunction induced by resistin could also participate to the enhanced atherosclerotic process that occurs in obese subjects.
A: Representative chart obtained during vascular reactivity studies. Arrows indicate the time of addition of phenylephrine (Phe) and of increasing doses of insulin (Ins) (one square = 100 mg). Vascular response of phenylephrine-precontracted aortic rings from young (8–12 weeks, n = 15) (B and C) and old (6 months, n = 10) (D and E) C57BL/6 mice to increasing doses of insulin (B and D) and acetylcholine (C and E) alone (▪, Ctrl) and in presence of resistin (▴, +Res) or l-NAME (•, +L-N). *P < 0.01 vs. insulin alone.
A: Representative chart obtained during vascular reactivity studies. Arrows indicate the time of addition of phenylephrine (Phe) and of increasing doses of insulin (Ins) (one square = 100 mg). Vascular response of phenylephrine-precontracted aortic rings from young (8–12 weeks, n = 15) (B and C) and old (6 months, n = 10) (D and E) C57BL/6 mice to increasing doses of insulin (B and D) and acetylcholine (C and E) alone (▪, Ctrl) and in presence of resistin (▴, +Res) or l-NAME (•, +L-N). *P < 0.01 vs. insulin alone.
Vascular response of phenylephrine-precontracted aortic rings from C57BL/6 mice chronically treated for 4 weeks with resistin (▴, +Res) or with vehicle (▪, Ctrl) to increasing doses of insulin (A) and acetylcholine (B). *P < 0.01 vs. vehicle.
Vascular response of phenylephrine-precontracted aortic rings from C57BL/6 mice chronically treated for 4 weeks with resistin (▴, +Res) or with vehicle (▪, Ctrl) to increasing doses of insulin (A) and acetylcholine (B). *P < 0.01 vs. vehicle.
eNOS enzymatic activity in aortic rings in control conditions (Ctrl) and in the presence of insulin (Ins) (50 nmol/l), acetylcholine (Ach), resistin (Res) (5 nmol/l), and l-NAME, alone and in combination (n = 10). *P < 0.01 vs. control; #P < 0.01 vs. insulin alone.
eNOS enzymatic activity in aortic rings in control conditions (Ctrl) and in the presence of insulin (Ins) (50 nmol/l), acetylcholine (Ach), resistin (Res) (5 nmol/l), and l-NAME, alone and in combination (n = 10). *P < 0.01 vs. control; #P < 0.01 vs. insulin alone.
A: Representative Western blotting showing eNOS and Akt phosphorylations. Endothelial cells were treated with insulin (50 nmol/l) or acetylcholine in presence or in absence of resistin (5 nmol/l). B: Quantification of eNOS Ser1,177 (□) and Akt Ser473 (▪) phosphorylations normalized for eNOS and Akt protein content, respectively (n = 5). *P < 0.01 vs. control; #P < 0.01 vs. insulin alone.
A: Representative Western blotting showing eNOS and Akt phosphorylations. Endothelial cells were treated with insulin (50 nmol/l) or acetylcholine in presence or in absence of resistin (5 nmol/l). B: Quantification of eNOS Ser1,177 (□) and Akt Ser473 (▪) phosphorylations normalized for eNOS and Akt protein content, respectively (n = 5). *P < 0.01 vs. control; #P < 0.01 vs. insulin alone.
A: Representative Western blotting of IRS-1 immunoprecipitated from endothelial cells in control conditions (Ctrl); treated with insulin (Ins) (50 nmol/l), resistin (Res) (5 nmol/l), or in combination (Ins+Res); or incubated with antibodies against IRS-1, against p85, against ubiquitin (n = 5), against phospho-Tyr608–IRS-1 (pY), and against anti–phospho-Ser307–IRS-1 (pS). B: Quantification of the p85–to–IRS-1 ratio (n = 5). *P < 0.01 vs. control. C: Quantification of phospho-Tyr608–IRS-1 (□) and anti–phospho-Ser307–IRS-1 (▪) normalized for protein content (n = 5). *P < 0.01 vs. control.
A: Representative Western blotting of IRS-1 immunoprecipitated from endothelial cells in control conditions (Ctrl); treated with insulin (Ins) (50 nmol/l), resistin (Res) (5 nmol/l), or in combination (Ins+Res); or incubated with antibodies against IRS-1, against p85, against ubiquitin (n = 5), against phospho-Tyr608–IRS-1 (pY), and against anti–phospho-Ser307–IRS-1 (pS). B: Quantification of the p85–to–IRS-1 ratio (n = 5). *P < 0.01 vs. control. C: Quantification of phospho-Tyr608–IRS-1 (□) and anti–phospho-Ser307–IRS-1 (▪) normalized for protein content (n = 5). *P < 0.01 vs. control.
Effects of chronic resistin administration
| Control | Chronic resistin | P | |
|---|---|---|---|
| Plasma resistin (ng/ml) | 7.9 ± 3.5 | 45.1 ± 6.3 | <0.001 |
| Plasma glucose level (mmol/l) | 7.24 ± 0.35 | 7.32 ± 0.61 | NS |
| Mean blood pressure (mmHg) | |||
| 1st week (tail-cuff) | 98 ± 3 | 94 ± 4 | NS |
| 2nd week (tail-cuff) | 96 ± 4 | 99 ± 3 | NS |
| 3rd week (tail-cuff) | 101 ± 2 | 97 ± 5 | NS |
| Final (intra-arterial method) | 96 ± 3 | 100 ± 2 | NS |
| Heart rate (beats/min) | 560 ± 40 | 580 ± 50 | NS |
| Aortic maximal constriction to phenylephrine (mg) | 1,060 ± 20 | 1,020 ± 30 | NS |
| Mesenteric maximal constriction to phenylephrine (mg) | 660 ± 30 | 680 ± 20 | NS |
| Control | Chronic resistin | P | |
|---|---|---|---|
| Plasma resistin (ng/ml) | 7.9 ± 3.5 | 45.1 ± 6.3 | <0.001 |
| Plasma glucose level (mmol/l) | 7.24 ± 0.35 | 7.32 ± 0.61 | NS |
| Mean blood pressure (mmHg) | |||
| 1st week (tail-cuff) | 98 ± 3 | 94 ± 4 | NS |
| 2nd week (tail-cuff) | 96 ± 4 | 99 ± 3 | NS |
| 3rd week (tail-cuff) | 101 ± 2 | 97 ± 5 | NS |
| Final (intra-arterial method) | 96 ± 3 | 100 ± 2 | NS |
| Heart rate (beats/min) | 560 ± 40 | 580 ± 50 | NS |
| Aortic maximal constriction to phenylephrine (mg) | 1,060 ± 20 | 1,020 ± 30 | NS |
| Mesenteric maximal constriction to phenylephrine (mg) | 660 ± 30 | 680 ± 20 | NS |
Data are means ± SE. NS, not significant.
Published ahead of print at http://diabetes.diabetesjournals.org on 7 December 2007. DOI: 10.2337/db07-0557.
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.
