Consumption of a Fat-Rich Diet Activates a Proinflammatory Response and Induces Insulin Resistance in the Hypothalamus
http://www.100md.com
《内分泌学杂志》
Departments of Internal Medicine (C.T.D.S., E.P.A., R.A., R.L.Z., M.J.A.S., L.A.V.) and Physiology and Biophysics (A.C.B.), State University of Campinas, 13084-970 Campinas-SP
Department of Physiology and Biophysics (S.B.), University of So Paulo, 05508-900 So Paulo-SP, Brazil
Abstract
Obesity has reached epidemic proportions in several regions of the world. General changes in lifestyle, including consumption of fat-rich food, are among the most important factors leading to an unprecedented increase in the prevalence of this disease. Weight gain results from an imbalance between caloric intake and energy expenditure. Both of these parameters are under the tight control of specialized neurons of the hypothalamus that respond to peripheral anorexigenic and adipostatic signals carried by leptin and insulin. Here we show, by macroarray analysis, that high-fat feeding [hyperlipidic diet (HL)] induces the expression of several proinflammatory cytokines and inflammatory responsive proteins in hypothalamus. This phenomenon is accompanied by increased activation of c-Jun N-terminal kinase and nuclear factor-B. In addition, HL feeding leads to impaired functional and molecular activation of the insulin-signaling pathway, which is paralleled by increased serine phosphorylation of the insulin receptor and insulin receptor substrate-2. Intracerebroventricular treatment of HL rats with a specific inhibitor of c-Jun N-terminal kinase (SP600125) restores insulin signaling and leads to a reduced caloric intake and weight loss. We conclude that HL feeding induces a local proinflammatory status in the hypothalamus, which results in impaired anorexigenic insulin signaling.
Introduction
DURING THE LAST decade, great advances were made in the characterization of the role played by the hypothalamus in the coordination of food intake and energy expenditure (1, 2). Leptin and insulin were shown to provide the most robust signals to specialized neurons of the hypothalamus to inform about energy stocks in the periphery and trigger physiologic responses aimed at controlling hunger and thermogenesis (1, 3, 4). Both hormones act through receptors located, mostly, in at least two subpopulations of neurons in the arcuate nucleus of the hypothalamus. Low levels of insulin or leptin, as observed during fasting or when total body fat mass is depleted, lead to the activation of neuropeptide Y/agouti gene-related peptidergic neurons, providing stimulatory signals to melanin-concentrating hormone- and orexin-producing neurons of the lateral hypothalamus. The consequence is increased hunger and decreased energy expenditure. In contrast, after a meal or when body fat stores are replete, the levels of insulin and leptin rise, leading to inhibition of neuropeptide Y/agouti gene-related peptidergic neurons and stimulating proopiomelanocortin/cocaine- and amphetamine-induced transcriptergic neurons; in turn, a reduction of orexin and melanin-concentrating hormone occurs in the lateral hypothalamus, accompanied by increased production of CRH and TRH in the periventricular nucleus, which promotes satiety and increases thermogenesis (4). Establishing a parallel between the development of type 2 diabetes mellitus and obesity, it has been proposed that hypothalamic resistance to insulin and leptin action plays a central role in the loss of coordinated control of food intake and energy expenditure (4, 5). In some animal models, this phenomenon has been documented (6, 7), but little progress has been achieved in the characterization of mechanisms that trigger and mediate insulin and leptin resistance in the hypothalamus.
In peripheral tissues, insulin resistance develops as a consequence of several variable factors, beginning with an appropriate genetic background which must, to a certain degree, be associated with environmental variables such as ageing, feeding patterns, sedentary lifestyle, stress, and infections among others. An environmental factor strongly linked with the development of peripheral insulin resistance, type 2 diabetes, and obesity is the consumption of fat-rich diets (8).
The primary objective of the present study was to evaluate the effect of long-term consumption of hyperlipidic (HL) diet on the expression of several mRNA specificities in hypothalamus of rats. Because HL diet promoted the regulation of expression of a number of mRNAs encoding for proteins that participate in the inflammatory response, we next evaluated the role of this proinflammatory phenomenon in insulin signal transduction and feeding behavior of rats.
Materials and Methods
Experimental animals and HL diet
Male 4-wk-old Wistar rats from the University of Campinas Breeding Center were randomly divided into two groups, control (Ctr), fed standard rodent chow ad libitum (protein, 20 kcal%; carbohydrate, 70 kcal%; lipid, 10 kcal%) and HL, fed a fat-rich chow ad libitum (protein, 20 kcal%; carbohydrate, 35 kcal%; lipid, 45 kcal%). This diet composition has been previously used (9). For macroarray analysis, hypothalami were obtained from rats of each group after 16 wk of dieting. For evaluation of cytokine expression by ELISA and Western blot and evaluation of c-Jun N-terminal kinase (JNK) and nuclear factor-B (NFB) activation, hypothalami were obtained from control rats at 4 wk of age (Ctr0) or control or HL rats at 13 (Ctr13, HL13) or 16 wk (Ctr16, HL16) after beginning a specific diet. For RT-PCR evaluation of cytokine expression, immunohistochemistry, and signal transduction studies, hypothalami were obtained from control and HL rats at 16 wk (Ctr16, HL16) after beginning a specific diet. As a rule, the hypothalamic fragments used for mRNA and protein extraction were 4.0 mm3 and had the optic chiasm as the rostral limit (bregma –0.25 mm), the infundibular stem as the caudal limit (bregma –4.20 mm) and were 4.0 mm wide and 3.0 mm deep. Fragments for immunohistochemical studies were larger (8.0–9.0 mm3), and the correct topographical localization of the sections was obtained elsewhere (10). In some rats the JNK-specific inhibitor SP600125 (Tocris, Ellisville, MO) was administered intracerebroventricularly (icv), twice a day, for 1 wk (from wk 15 to 16 after beginning a specific diet) (Ctr16SP and HL16SP). The doses of 12, 23, and 46 μmol/dose were tested in dose-response experiments and, because similar effects to inhibit [Thr183] phosphorylation of JNK were obtained with 23 and 46 μmol/dose, the dose of 23 μmol/dose was used for signal transduction and functional experiments. Some rats of the HL group were submitted to 1 wk pair feeding (HL16PF) with rats treated with SP600125 (HL16SP).
Intracerebroventricular cannulation
For SP600125 treatment, insulin-induced inhibition of food intake and insulin or leptin signal transduction experiments rats were stereotaxically instrumented using a Stoelning stereotaxic apparatus, according to a method previously described (6). Cannula patency was tested 1 wk after cannulation by the evaluation of the drinking response elicited by icv angiotensin II (11).
Body and epididymal mass, icv insulin-induced inhibition of 12-h food intake, and daily caloric intake
All these parameters were evaluated following protocols previously described (6, 7).
RNA preparation for macroarray and RT-PCR
Total RNA was extracted using Trizol reagent (Life Technologies, Gaithersburg, MD), according to the manufacturer’s recommendations. Total RNA was rendered genomic DNA free by digestion with Rnase-free Dnase (RQ1; Promega, Madison, WI).
Macroarray analysis
Rat 1.2 Atlas Array was from Clontech (Palo Alto, CA). One microgram of poly(A)+ RNA was converted into 33P-labeled first strand cDNA by Moloney murine leukemia virus reverse transcriptase. Unincorporated 33P-labeled nucleotides were removed by chromatography using a NucleoSpin extraction spin column (Clontech). Purified cDNA probes were hybridized to the Atlas membranes. Hybridization occurred overnight at 68 C. After washing, membranes were sealed in hybridization bags and exposed to imaging plates for 1 d. After exposure, the imaging plates were scanned using a BAS-1500 (Fujifilm, Fuji, Japan), and hybridization signals were counted. Hybridization signals of each gene were normalized by a positive control (signals of the housekeeping gene), and gene expression was compared between the preconditioning and control groups.
Semiquantitative RT-PCR
Seven micrograms of total RNA were reverse transcribed with SuperScript reverse transcriptase (200 U/μl) using oligo (dT) (50 mM) in a 30 μl reaction volume (5x reverse transcription buffer, 10 mM deoxynucleotide triphosphate, and 40 U/μl Rnase-free inhibitor). The reverse transcriptions involved a 50-min incubation at 42 C and a 15-min incubation at 70 C. The PCR products were submitted to 1.5% agarose gel electrophoresis containing ethidium bromide and visualized by excitation under UV light. Photodocumentation was performed using the Nucleovision system (NucleoTech, San Mateo, CA), and band quantification was performed using the Gel Expert software (NucleoTech). In all samples the amplification of cyclophiline was performed and used as an internal control for quantity and quality. The semiquantitative expression of cytokines was calculated using the formula: semiquantitative expression = pixel area of product/pixel area of cyclophilin x 100. The primers used were: TNF sense, 5'-GTG CCT CAG CCT CTT CTC ATT CC-3', antisense, 5'-GCT CCT CCG CCT GGT GGT TT-3' (product 218 bp); IL-1 sense, 5'-GGA TGA TGA CGA CCT GC-3', antisense, 5'-TCC CGA CCC TTG CTG TT-3' (product 450 bp); IL-6 sense, 5'-CCT TCT TGG GAC TGA TGT-3', antisense, 5'-CTC TGG CTT TGT CTT TCT-3' (product 384 bp); and cyclophilin sense, 5'-GAC AGC AGA AAA CTT TCG TGC-3', antisense, 5'-GGT TCT GAC TCA CCG ACC T-3' (product 276 bp).
ELISA for cytokine determination
Tissue cytokine levels were determined in samples of hypothalamic protein extracts (2.0 mg/ml) by ELISA (Pierce Biotechnology, Rockford, IL), following the recommendations of the manufacturer.
Determination of NFB activation
Basal hypothalamic NFB activation was determined in nuclear extracts by ELISA (Pierce Biotechnology) according to the recommendations of the manufacturer and using positive and negative controls supplied with the kit.
Tissue extraction, immunoblotting and immunoprecipitation
The icv cannulated rats were anesthetized and acutely treated with saline (2.0 μl), insulin (10–6 M, 2.0 μl), or leptin (10–6 M, 2.0 μl). After 2 min [for insulin receptor (IR)], 5 min [for insulin receptor substrate (IRS)-2, phosphatidylinositol 3-kinase (PI3-kinase), and Akt], or 10 min [for signal transducer and activator of transcription (STAT)-3], the hypothalami were obtained and immediately homogenized in solubilization buffer at 4 C [1% Triton X-100, 100 mM Tris-HCl (pH 7.4), 100 mM sodium pyrophosphate, 100 mM sodium fluoride, 10 mM EDTA, 10 mM sodium orthovanadate, 2.0 mM phenylmethylsulfonyl fluoride, and 0.1 mg aprotinin/ml] with a Polytron PTA 20S generator (model PT 10/35; Brinkmann Instruments, Westbury, NY). Insoluble material was removed by centrifugation for 20 min at 9000 x g in a 70. Ti rotor (Beckman, Fullerton, CA) at 4 C. The protein concentration of the supernatants was determined by the Bradford dye binding method. Aliquots of the resulting supernatants containing 2.0 mg of total protein were used for immunoprecipitation with antibodies against IR, IRS-2, and STAT-3 at 4 C overnight, followed by SDS-PAGE, transfer to nitrocellulose membranes, and blotting with antiphosphotyrosine, antiphosphoserine, anti-IR, anti-IRS-2, anti-STAT-3, or anti-p85/PI3 kinase. In direct immunoblot experiments, 0.2 mg of protein extracts were separated by SDS-PAGE, transferred to nitrocellulose membranes, and blotted with anti-TNF, anti-IL-1, anti-IL-6, anti-IR, anti-IRS-2, anti-STAT-3, anti-suppressor of cytokine signaling (SOCS)-3, anti-phospho-[Ser473]Akt, and anti-phospho-[Thr183]-JNK antibodies, as described (12). The protein loading of immunoblots was always evaluated by two methods: before blotting, the nitrocellulose membranes were stained with Coomassie Blue, and after blotting the membranes were reprobed with an antivimentin antibody.
Optic and confocal microscopy
Paraformaldehyde-fixed hypothalami were sectioned (5 μm) and used in regular single- or double-immunofluorescence staining using TNF, NeuN, or phospho-[Thr183]-JNK antibodies according to a previously described protocol (6, 13). Analysis and documentation of results were performed using an LSM 510 laser confocal microscope (Zeiss, Gttingen, Germany) and a BX60 microscope (Olympus, Tokyo, Japan). Semiquantitative assessment of NeuN, TNF and phospho-[Thr183]-JNK was achieved by counting the number of positively stained cells. Every second of all consecutive sections were counted in arcuate nucleus and lateral hypothalamus, sites of predominant staining for both TNF and phospho-[Thr183]-JNK. The anatomical correlations were made according to the landmarks given in a stereotaxic atlas (10). The topographical views of the regions to be studied were obtained by hematoxylin-eosin staining of consecutive sections.
Statistical analysis
The results of blots are presented as direct comparisons of bands in autoradiographs and quantified by densitometry using the Scion Image software (Scion Corp., Frederick, MD). All data, except those from immunohistochemistry and macroarray, were analyzed by the two-tailed unpaired Student’s t test or by repeat-measures ANOVA (one-way or two-way ANOVA) followed by post hoc analysis of significance (Bonferroni test) when appropriate, comparing experimental and control groups. In immunohistochemistry the average number of positive cells per section was obtained for each animal. Differences between groups were evaluated by the nonparametric Kruskal-Wallis and Mann-Whitney U test. The data are expressed as percent variation from respective control. In macroarray analysis, genes exhibiting differential expression in the preconditioning group were selected only if the hybridization signals were either increased or decreased by at least 2-fold, compared with those of the control group, and according to the statistical analysis performed by ANOVA. The functional clustering of genes was performed by combining the information obtained from public databases such as the National Institutes of Health (www.ncbi.nlm.nih.gov) and the Rat Genome Database (www.rgd.mcw.edu). In all experiments, the level of significance was set at P < 0.05.
Results and Discussion
To evaluate the role of HL feeding on hypothalamic protein expression, Wistar rats treated during 16 wk with a diet containing 45 kcal% fat were used for macroarray analysis of 1176 mRNAs in hypothalamus in comparison with rats treated with a regular rodent chow, containing 10 kcal% fat. HL feeding modulated the expression of 170 mRNAs specificities (92 up-regulated and 78 down-regulated; complete data presented in supplemental Table 1, published on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org), including hormones, growth factors, transcription factors, enzymes of lipid metabolism, and membrane and nuclear receptors among others. HL feeding modulated the expression of mRNAs encoding proteins known to participate in the control of feeding, thermogenesis, and other important metabolic functions. These mRNAs included the IR (5.42 ± 0.03-fold increase), leptin receptor ObR (8.89 ± 0.07-fold increase), CRH binding protein (3.52 ± 0.03-fold increase), and N-methyl-D-aspartate receptor 2B (3.86 ± 0.04-fold increase). All these mRNAs were functionally clustered as metabolic-related sequences, and the data are presented in supplemental Table 2. However, after complete functional clustering, we observed that immune-related proteins’ mRNAs represented the largest group modulated by HL feeding. Table 1 presents all mRNA specificities that encode immune-related proteins that were modulated by HL feeding.
In recent years, a bulk of evidence has been obtained to show that certain proinflammatory cytokines, produced by the adipose tissue, participate in the induction of insulin resistance in tissues that act as traditional targets for insulin action (14, 15). This phenomenon is mediated by the activation of intracellular serine kinases such as JNK and IB kinase, which act as intermediaries for proinflammatory signaling, but are also able to induce serine phosphorylation of key elements of the insulin signal transduction pathway and by this mechanism blunt insulin action (14, 15, 16). TNF, IL-1, and IL-6 have been described as inducers of insulin resistance by the aforementioned mechanism (14, 15, 16). In the macroarray analysis, the mRNAs of all these three cytokines were increased by the consumption of the HL diet. Therefore, we used distinct methods to evaluate the expression/concentration of these cytokines in the hypothalamus of rats. By ELISA (Fig. 1A), it was shown that the hypothalamic levels of TNF and IL-1 were significantly increased at 16 wk, compared with respective controls (Ctr16), whereas IL-6 was increased at 16 wk, compared with the basal control (Ctr0). By immunoblot (Fig. 1B), all three cytokines were shown to be significantly increased in the hypothalamus of rats fed on the HL diet for 16 wk, compared with respective controls (Ctr16) and basal controls (Ctr0). Moreover, by the same method, it was shown that the consumption of HL diet for 13 wk was also sufficient to promote a significant increase on TNF and IL-6 expression (Fig. 1B). In addition, by RT-PCR (Fig. 1C), the levels of mRNAs encoding for TNF and IL-1 were found to be significantly increased.
By confocal microscopy, we observed that low levels of TNF could be detected in the arcuate nucleus and lateral hypothalamus of rats fed on a regular diet (Fig. 1D). Feeding the rats with the HL diet for 16 wk promoted an increase in TNF in both hypothalamic nuclei. In the arcuate nucleus and lateral hypothalamus TNF-specific staining was found both coinciding and not coinciding with neuron bodies. Specifically in the lateral hypothalamus, the HL diet produced a 44 ± 12% increase (P < 0.05) in the number of TNF-stained neuron bodies. In other regions of the hypothalamus and the remainder of the brain, TNF staining was minimal. There were no significant differences in the number of NeuN-positive cells in lateral hypothalamus and arcuate nucleus between the groups.
To evaluate whether the higher concentration of proinflammatory cytokines in the hypothalamus may lead to the activation of proinflammatory intracellular signaling, we performed immunoblots to determine phospho-[Thr183]-JNK and ELISA to determine NFB activation. As depicted in Fig. 1E, the consumption of the HL diet for 13 or 16 wk significantly increased the levels of phospho-[Thr183]-JNK, whereas the consumption of HL diet for 16 wk activated NFB. No phospho-[Thr183]-JNK signal could be detected by confocal microscopy in rats fed on the control diet. However, specific staining was detected in several neuron bodies of the arcuate nucleus and lateral hypothalamus in HL rats (Fig. 1F).
The icv injection of a single dose of insulin inhibits 12 h spontaneous food intake. Most of this effect is dependent on insulin binding and activation of the IR present in neurons of the arcuate nucleus and activation of an intracellular cascade that depends on the engagement of IRS-2, PI3-kinase, and Akt (6, 7, 17). The treatment of rats with a HL diet during 16 wk did not modulate the protein expression of IR (Fig. 2A) and IRS-2 (Fig. 2B) but significantly reduced insulin-induced tyrosine phosphorylation of IR (Fig. 2A) and IRS-2 (Fig. 2B). This was accompanied by reduced insulin-induced association of IRS-2 with PI3-kinase (Fig. 2C) and reduced insulin-induced [Ser473]-phosphorylation of Akt (Fig. 2D). Because proinflammatory signaling is known to induce molecular resistance to insulin by promoting serine phosphorylation of key elements of the insulin signaling pathway (14, 15), we performed immunoblot analysis of IR and IRS-2 immunoprecipitates obtained from hypothalamic protein extracts of rats fed on control and HL diets. As depicted in Fig. 2E, both IR and IRS-2 presented higher levels of serine phosphorylation in samples from rats fed on HL diet, compared with the control rats. Finally, insulin-induced inhibition of spontaneous food intake was severely affected by the consumption of the HL diet (Fig. 2F), demonstrating that, in parallel to the molecular resistance to insulin, high-fat feeding negatively modulates a functional phenomenon controlled by insulin in the hypothalamus.
The reason the IR protein level was not modulated by HL feeding whereas IR mRNA was significantly increased is not known. We suspect that, as in other circumstances, proinflammatory events may induce higher protein turnover and differential regulation of transcription and translation (18). In this context, a recent study has shown that the cytokine signaling suppressors, SOCS-1 and -3, direct elements of the insulin signaling pathway to proteosomic degradation, and therefore, accelerate the rate of protein turnover (19). The fact that the IR protein amount was not effectively increased by HL feeding may have implications not only in insulin signal transduction in hypothalamic neurons but also in the rate of insulin transport across the blood-brain barrier. We have not measured this transport capacity, but previous studies have shown that, in animal models of obesity and insulin resistance, the binding of insulin to brain capillaries is impaired, suggesting the participation of this mechanism in the inappropriate action of insulin in the central nervous system (20, 21).
To test the hypothesis that proinflammatory signaling stimulated by HL feeding may be involved in the inhibition of insulin signaling in hypothalamus, we treated control and HL rats with a specific inhibitor of JNK, SP600125. The compound was icv injected and after dose optimization based on previous reports (22), we observed that an almost complete inhibition of [Thr183] phosphorylation of JNK was obtained in control and HL rats (Fig. 3A). The treatment with the compound was also able to significantly reduce daily caloric intake (Fig. 3B) and promote a loss of body mass (Fig. 3C). Mean body mass at the beginning of the period of treatment was 338 ± 24, 342 ± 26, 403 ± 20, and 405 ± 22 g for control, control treated with SP600125, HL, and HL treated with SP600125, respectively. At the end of the treatment period (8 d), the body mass was 350 ± 23, 345 ± 22, 420 ± 24, and 397 ± 19 g for control, control treated with SP600125, HL, and HL treated with SP600125, respectively. In control rats, the compound also exerted some effect promoting a significant reduction in daily caloric intake (Fig. 3B). However, this was of a significantly (P < 0.05) lower magnitude than in HL rats and led only to a reduction of the rate of body mass gain (Fig. 3C). Interestingly, the loss of body mass promoted by the treatment with SP600125 was, in part, due to a significant fall in intraabdominal fat, as determined by the measurement of epididymal fat mass (Fig. 3D).
Next, we evaluated the effect of icv injection of SP600125 on insulin signaling and action in the hypothalamus of rats fed on the HL diet. As depicted in Fig. 4, A and B, the compound exerted no effect on IR and IRS-2 protein expression. It did, however, revert the effect of HL feeding on insulin-induced tyrosine phosphorylation of IR and IRS-2 in such a way that no difference was found between control and rats fed on the HL diet and treated with the compound. Restoration of insulin signaling through PI3-kinase and Akt was also obtained by the icv injection of SP600125 in HL rats (Fig. 4, C and D). All these phenomena were accompanied by a reduction in Ser phosphorylation of IR and IRS-2, which returned to levels similar to those of the control (Fig. 4E). Moreover, the treatment of HL rats with SP600125 significantly increased insulin-induced inhibition of 12 h spontaneous food intake, whereas no effect of the compound was detected in control rats (Fig. 4F). To determine whether the effect of SP600125 on hypothalamic insulin signal transduction was dependent on the reduction in food intake and loss of body mass, we evaluated insulin-induced tyrosine phosphorylation of the IR and IRS-2 and [Ser 473]-phosphorylation of Akt in pair-fed, HL diet-treated rats. Pair feeding promoted a similar body mass loss to that produced by SP600125 treatment. However, as depicted in Fig. 4G, pair feeding was not enough to restore baseline insulin-induced molecular activation of these proteins. Therefore, we suspect that most of the effect of SP600125 is mediated by its specific action on JNK and not through indirect actions on food intake and body mass.
Hypercaloric and particularly hyperlipidic feeding is implicated as one of the most important environmental factors leading to obesity. Recent studies have shown that ingestion of a fat-rich diet leads to hypothalamic resistance to leptin (9, 23). This phenomenon is partially dependent on the immune-related signaling suppressor SOCS-3, which interacts with and inhibits elements of the ObR signaling pathway (24, 25). To evaluate whether the proinflammatory phenomenon, herein described, may somehow interact and modulate canonical leptin signaling, we determined SOCS-3 expression and leptin-induced activation of STAT-3 in hypothalamus. As shown in Fig. 4 h, the consumption of the HL diet for 16 wk promoted a significant increase in the hypothalamic expression of SOCS-3, which was not reverted by a 1-wk treatment with SP600125. In addition, HL diet consumption led to the inhibition of leptin-induced tyrosine phosphorylation of STAT-3. Once more, inhibition of JNK with SP600125 did not revert the inhibitory effect of HL feeding on the ability of leptin to induce STAT-3 molecular activation.
In the present study, we demonstrate that the consumption of a fat-rich diet modulates the expression of several hypothalamic mRNAs encoding proteins with immune-related activities. The ability of certain types of fatty acids to modulate the production of cytokines has been known for quite some time and both pro- and antiinflammatory outcomes have been described, depending on the type of fatty acid studied (26, 27). Distinct mechanisms seem to participate in fatty acid-dependent regulation of the immune response. These mechanisms range from the control of prostaglandin E2 and leukotriene B4 production, which may affect signal transduction through IB kinase (28, 29), to direct action on nuclear receptors controlling the production of immune-related factors (30). Although we cannot be sure whether the proinflammatory phenomenon, herein described, is a direct or indirect consequence of the consumption of fat-rich food, we believe our data have established an important link between a common environmental event, i.e. consumption of fat-rich food, and the generation of a microenvironmental phenomenon of inflammation at an anatomical site involved in the central control of metabolic functions. Moreover, this study shows that inhibition of a protein that plays a central role in one of the most important proinflammatory intracellular signaling pathways, i.e. JNK, improves metabolic parameters and restores hypothalamic insulin signaling in rats fed a HL diet. These data place local inflammation into a pivotal position in the pathophysiological mechanisms involved in diet-induced obesity and opens novel therapeutic perspectives for this medical condition.
Acknowledgments
We are grateful to Dr. Nicola Conran for English grammar review.
Footnotes
This work was supported by grants from Fundao de Amparo à Pesquisa do Estado de So Paulo and Conselho Nacional de Pesquisa.
1 C.T.D.S. and E.P.A. contributed equally to these studies.
Abbreviations: Ctr, Control; HL, hyperlipidic; icv, intracerebroventricular; IR, insulin receptor; IRS, insulin receptor substrate; JNK, c-Jun N-terminal kinase; NFB, nuclear factor-B; PI3-kinase, phosphatidylinositol 3-kinase; SOCS, suppressor of cytokine signaling; STAT, signal transducer and activator of transcription.
References
Flier JS 2004 Obesity wars: molecular progress confronts an expanding epidemic. Cell 116:337–350
Friedman JM 2000 Obesity in the new millennium. Nature 404:632–634
Friedman JM, Halaas JL 1998 Leptin and the regulation of body weight in mammals. Nature 395:763–770
Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671
Schwartz MW, Kahn SE 1999 Insulin resistance and obesity. Nature 402:860–861
Torsoni MA, Carvalheira JB, Pereira-Da-Silva M, de Carvalho-Filho MA, Saad MJ, Velloso LA 2003 Molecular and functional resistance to insulin in hypothalamus of rats exposed to cold. Am J Physiol Endocrinol Metab. 285:E216–E223
Carvalheira JB, Ribeiro EB, Araujo EP, Guimaraes RB, Telles MM, Torsoni M, Gontijo JA, Velloso LA, Saad MJ 2003 Selective impairment of insulin signalling in the hypothalamus of obese Zucker rats. Diabetologia. 46:1629–1640
Stein CJ, Colditz GA 2004 The epidemic of obesity. J Clin Endocrinol Metab. 89:2522–2525
El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, Flier JS 2000 Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obesity. J Clin Invest. 105:1827–1832
Paxinos G, Watson C 1997 The rat brain in stereotaxic coordinates. 4th ed. San Diego: Academic Press
Johnson AK, Epstein AN 1975 The cerebral ventricles as the avenue for the dipsogenic action of intracranial angiotensin. Brain Res. 86:399–418
Velloso LA, Kampe O, Hallberg A, Christmanson L, Betsholtz C, Karlsson FA 1993 Demonstration of GAD-65 as the main immunogenic isoform of glutamate decarboxylase in type 1 diabetes and determination of autoantibodies using a radioligand produced by eukaryotic expression. J Clin Invest. 91:2084–2090
Araujo EP, Amaral ME, Filiputti E, De Souza CT, Laurito TL, Augusto VD, Saad MJ, Boschero AC, Velloso LA, Carneiro EM 2004 Restoration of insulin secretion in pancreatic islets of protein-deficient rats by reduced expression of insulin receptor substrate (IRS)-1 and IRS-2. J Endocrinol. 181:25–38
Hotamisligil GS 2003 Inflammatory pathways and insulin action. Int J Obes Relat Metab Disord. 27(Suppl 3):S53–S55
Shoelson SE, Lee J, Yuan M 2003 Inflammation and the IKK/IB/NF-B axis in obesity- and diet-induced insulin resistance. Int J Obes Relat Metab Disord. 27(Suppl 3):S49–S52
Hotamisligil GS 2000 Molecular mechanisms of insulin resistance and the role of the adipocyte. Int J Obes Relat Metab Disord. 24(Suppl 4):S23–S27
Niswender KD, Morrison CD, Clegg DJ, Olson R, Baskin DG, Myers Jr MG, Seeley RJ, Schwartz MW 2003 Insulin activation of phosphatidylinositol 3-kinase in the hypothalamic arcuate nucleus: a key mediator of insulin-induced anorexia. Diabetes 52:227–231
Colangelo V, Schurr J, Ball MJ, Pelaez RP, Bazan NG, Lukiw WJ 2002 Gene expression profiling of 12633 genes in Alzheimer hippocampal CA1: transcription and neurotrophic factor down-regulation and up-regulation of apoptotic and pro-inflammatory signaling. J Neurosci Res. 70:462–473
Rui L, Yuan M, Frantz D, Shoelson S, White MF 2002 SOCS-1 and SOCS-3 block insulin signaling by ubiquitin-mediated degradation of IRS1 and IRS2. J Biol Chem. 277:42394–42398
Schwartz MW, Figlewicz DF, Kahn SE, Baskin DG, Greenwood MR, Porte Jr D1990 Insulin binding to brain capillaries is reduced in genetically obese, hyperinsulinemic Zucker rats. Peptides 11:467–472
Pardridge WM 1993 Transport of insulin-related peptides and glucose across the blood-brain barrier. Ann NY Acad Sci. 692:126–137
Nadruz Jr W, Kobarg CB, Kobarg J, Franchini KG 2004 c-Jun is regulated by combination of enhanced expression and phosphorylation in acute-overloaded rat heart. Am J Physiol Heart Circ Physiol. 286:H760–H777
Munzberg H, Flier JS, Bjorbaek C 2004 Region-specific leptin resistance within the hypothalamus of diet-induced-obese mice. Endocrinology 145:4880–4889
Howard JK, Cave BJ, Oksanen LJ, Tzameli I, Bjorbaek C, Flier JS 2004 Enhanced leptin sensitivity and attenuation of diet-induced obesity in mice with haploinsufficiency of Socs3. Nat Med. 10:734–738
Mori H, Hanada R, Hanada T, Aki D, Mashima R, Nishinakamura H, Torisu T, Chien KR, Yasukawa H, Yoshimura A 2004 Socs3 deficiency in the brain elevates leptin sensitivity and confers resistance to diet-induced obesity. Nat Med. 10:739–743
Wallace FA, Miles EA, Calder PC 2000 Activation state alters the effect of dietary fatty acids on pro-inflammatory mediator production by murine macrophages. Cytokine 12:1374–1379
Blok WL, Katan MB, van der Meer JW 1996 Modulation of inflammation and cytokine production by dietary (n-3) fatty acids. J Nutr. 126:1515–1533
Brach MA, de Vos S, Arnold C, Gruss HJ, Mertelsmann R, Herrmann F 1992 Leukotriene B4 transcriptionally activates interleukin-6 expression involving NK-B and NF-IL6. Eur J Immunol. 22:2705–2711
Rossi A, Kapahi P, Natoli G, Takahashi T, Chen Y, Karin M, Santoro MG 2000 Anti-inflammatory cyclopentenone prostaglandins are direct inhibitors of IB kinase. Nature 403:103–108
Chinetti G, Fruchart JC, Staels B 2000 Peroxisome proliferator-activated receptors (PPARs): nuclear receptors at the crossroads between lipid metabolism and inflammation. Inflamm Res. 49:497–505(Cláudio T. De Souza1, Eli)
Department of Physiology and Biophysics (S.B.), University of So Paulo, 05508-900 So Paulo-SP, Brazil
Abstract
Obesity has reached epidemic proportions in several regions of the world. General changes in lifestyle, including consumption of fat-rich food, are among the most important factors leading to an unprecedented increase in the prevalence of this disease. Weight gain results from an imbalance between caloric intake and energy expenditure. Both of these parameters are under the tight control of specialized neurons of the hypothalamus that respond to peripheral anorexigenic and adipostatic signals carried by leptin and insulin. Here we show, by macroarray analysis, that high-fat feeding [hyperlipidic diet (HL)] induces the expression of several proinflammatory cytokines and inflammatory responsive proteins in hypothalamus. This phenomenon is accompanied by increased activation of c-Jun N-terminal kinase and nuclear factor-B. In addition, HL feeding leads to impaired functional and molecular activation of the insulin-signaling pathway, which is paralleled by increased serine phosphorylation of the insulin receptor and insulin receptor substrate-2. Intracerebroventricular treatment of HL rats with a specific inhibitor of c-Jun N-terminal kinase (SP600125) restores insulin signaling and leads to a reduced caloric intake and weight loss. We conclude that HL feeding induces a local proinflammatory status in the hypothalamus, which results in impaired anorexigenic insulin signaling.
Introduction
DURING THE LAST decade, great advances were made in the characterization of the role played by the hypothalamus in the coordination of food intake and energy expenditure (1, 2). Leptin and insulin were shown to provide the most robust signals to specialized neurons of the hypothalamus to inform about energy stocks in the periphery and trigger physiologic responses aimed at controlling hunger and thermogenesis (1, 3, 4). Both hormones act through receptors located, mostly, in at least two subpopulations of neurons in the arcuate nucleus of the hypothalamus. Low levels of insulin or leptin, as observed during fasting or when total body fat mass is depleted, lead to the activation of neuropeptide Y/agouti gene-related peptidergic neurons, providing stimulatory signals to melanin-concentrating hormone- and orexin-producing neurons of the lateral hypothalamus. The consequence is increased hunger and decreased energy expenditure. In contrast, after a meal or when body fat stores are replete, the levels of insulin and leptin rise, leading to inhibition of neuropeptide Y/agouti gene-related peptidergic neurons and stimulating proopiomelanocortin/cocaine- and amphetamine-induced transcriptergic neurons; in turn, a reduction of orexin and melanin-concentrating hormone occurs in the lateral hypothalamus, accompanied by increased production of CRH and TRH in the periventricular nucleus, which promotes satiety and increases thermogenesis (4). Establishing a parallel between the development of type 2 diabetes mellitus and obesity, it has been proposed that hypothalamic resistance to insulin and leptin action plays a central role in the loss of coordinated control of food intake and energy expenditure (4, 5). In some animal models, this phenomenon has been documented (6, 7), but little progress has been achieved in the characterization of mechanisms that trigger and mediate insulin and leptin resistance in the hypothalamus.
In peripheral tissues, insulin resistance develops as a consequence of several variable factors, beginning with an appropriate genetic background which must, to a certain degree, be associated with environmental variables such as ageing, feeding patterns, sedentary lifestyle, stress, and infections among others. An environmental factor strongly linked with the development of peripheral insulin resistance, type 2 diabetes, and obesity is the consumption of fat-rich diets (8).
The primary objective of the present study was to evaluate the effect of long-term consumption of hyperlipidic (HL) diet on the expression of several mRNA specificities in hypothalamus of rats. Because HL diet promoted the regulation of expression of a number of mRNAs encoding for proteins that participate in the inflammatory response, we next evaluated the role of this proinflammatory phenomenon in insulin signal transduction and feeding behavior of rats.
Materials and Methods
Experimental animals and HL diet
Male 4-wk-old Wistar rats from the University of Campinas Breeding Center were randomly divided into two groups, control (Ctr), fed standard rodent chow ad libitum (protein, 20 kcal%; carbohydrate, 70 kcal%; lipid, 10 kcal%) and HL, fed a fat-rich chow ad libitum (protein, 20 kcal%; carbohydrate, 35 kcal%; lipid, 45 kcal%). This diet composition has been previously used (9). For macroarray analysis, hypothalami were obtained from rats of each group after 16 wk of dieting. For evaluation of cytokine expression by ELISA and Western blot and evaluation of c-Jun N-terminal kinase (JNK) and nuclear factor-B (NFB) activation, hypothalami were obtained from control rats at 4 wk of age (Ctr0) or control or HL rats at 13 (Ctr13, HL13) or 16 wk (Ctr16, HL16) after beginning a specific diet. For RT-PCR evaluation of cytokine expression, immunohistochemistry, and signal transduction studies, hypothalami were obtained from control and HL rats at 16 wk (Ctr16, HL16) after beginning a specific diet. As a rule, the hypothalamic fragments used for mRNA and protein extraction were 4.0 mm3 and had the optic chiasm as the rostral limit (bregma –0.25 mm), the infundibular stem as the caudal limit (bregma –4.20 mm) and were 4.0 mm wide and 3.0 mm deep. Fragments for immunohistochemical studies were larger (8.0–9.0 mm3), and the correct topographical localization of the sections was obtained elsewhere (10). In some rats the JNK-specific inhibitor SP600125 (Tocris, Ellisville, MO) was administered intracerebroventricularly (icv), twice a day, for 1 wk (from wk 15 to 16 after beginning a specific diet) (Ctr16SP and HL16SP). The doses of 12, 23, and 46 μmol/dose were tested in dose-response experiments and, because similar effects to inhibit [Thr183] phosphorylation of JNK were obtained with 23 and 46 μmol/dose, the dose of 23 μmol/dose was used for signal transduction and functional experiments. Some rats of the HL group were submitted to 1 wk pair feeding (HL16PF) with rats treated with SP600125 (HL16SP).
Intracerebroventricular cannulation
For SP600125 treatment, insulin-induced inhibition of food intake and insulin or leptin signal transduction experiments rats were stereotaxically instrumented using a Stoelning stereotaxic apparatus, according to a method previously described (6). Cannula patency was tested 1 wk after cannulation by the evaluation of the drinking response elicited by icv angiotensin II (11).
Body and epididymal mass, icv insulin-induced inhibition of 12-h food intake, and daily caloric intake
All these parameters were evaluated following protocols previously described (6, 7).
RNA preparation for macroarray and RT-PCR
Total RNA was extracted using Trizol reagent (Life Technologies, Gaithersburg, MD), according to the manufacturer’s recommendations. Total RNA was rendered genomic DNA free by digestion with Rnase-free Dnase (RQ1; Promega, Madison, WI).
Macroarray analysis
Rat 1.2 Atlas Array was from Clontech (Palo Alto, CA). One microgram of poly(A)+ RNA was converted into 33P-labeled first strand cDNA by Moloney murine leukemia virus reverse transcriptase. Unincorporated 33P-labeled nucleotides were removed by chromatography using a NucleoSpin extraction spin column (Clontech). Purified cDNA probes were hybridized to the Atlas membranes. Hybridization occurred overnight at 68 C. After washing, membranes were sealed in hybridization bags and exposed to imaging plates for 1 d. After exposure, the imaging plates were scanned using a BAS-1500 (Fujifilm, Fuji, Japan), and hybridization signals were counted. Hybridization signals of each gene were normalized by a positive control (signals of the housekeeping gene), and gene expression was compared between the preconditioning and control groups.
Semiquantitative RT-PCR
Seven micrograms of total RNA were reverse transcribed with SuperScript reverse transcriptase (200 U/μl) using oligo (dT) (50 mM) in a 30 μl reaction volume (5x reverse transcription buffer, 10 mM deoxynucleotide triphosphate, and 40 U/μl Rnase-free inhibitor). The reverse transcriptions involved a 50-min incubation at 42 C and a 15-min incubation at 70 C. The PCR products were submitted to 1.5% agarose gel electrophoresis containing ethidium bromide and visualized by excitation under UV light. Photodocumentation was performed using the Nucleovision system (NucleoTech, San Mateo, CA), and band quantification was performed using the Gel Expert software (NucleoTech). In all samples the amplification of cyclophiline was performed and used as an internal control for quantity and quality. The semiquantitative expression of cytokines was calculated using the formula: semiquantitative expression = pixel area of product/pixel area of cyclophilin x 100. The primers used were: TNF sense, 5'-GTG CCT CAG CCT CTT CTC ATT CC-3', antisense, 5'-GCT CCT CCG CCT GGT GGT TT-3' (product 218 bp); IL-1 sense, 5'-GGA TGA TGA CGA CCT GC-3', antisense, 5'-TCC CGA CCC TTG CTG TT-3' (product 450 bp); IL-6 sense, 5'-CCT TCT TGG GAC TGA TGT-3', antisense, 5'-CTC TGG CTT TGT CTT TCT-3' (product 384 bp); and cyclophilin sense, 5'-GAC AGC AGA AAA CTT TCG TGC-3', antisense, 5'-GGT TCT GAC TCA CCG ACC T-3' (product 276 bp).
ELISA for cytokine determination
Tissue cytokine levels were determined in samples of hypothalamic protein extracts (2.0 mg/ml) by ELISA (Pierce Biotechnology, Rockford, IL), following the recommendations of the manufacturer.
Determination of NFB activation
Basal hypothalamic NFB activation was determined in nuclear extracts by ELISA (Pierce Biotechnology) according to the recommendations of the manufacturer and using positive and negative controls supplied with the kit.
Tissue extraction, immunoblotting and immunoprecipitation
The icv cannulated rats were anesthetized and acutely treated with saline (2.0 μl), insulin (10–6 M, 2.0 μl), or leptin (10–6 M, 2.0 μl). After 2 min [for insulin receptor (IR)], 5 min [for insulin receptor substrate (IRS)-2, phosphatidylinositol 3-kinase (PI3-kinase), and Akt], or 10 min [for signal transducer and activator of transcription (STAT)-3], the hypothalami were obtained and immediately homogenized in solubilization buffer at 4 C [1% Triton X-100, 100 mM Tris-HCl (pH 7.4), 100 mM sodium pyrophosphate, 100 mM sodium fluoride, 10 mM EDTA, 10 mM sodium orthovanadate, 2.0 mM phenylmethylsulfonyl fluoride, and 0.1 mg aprotinin/ml] with a Polytron PTA 20S generator (model PT 10/35; Brinkmann Instruments, Westbury, NY). Insoluble material was removed by centrifugation for 20 min at 9000 x g in a 70. Ti rotor (Beckman, Fullerton, CA) at 4 C. The protein concentration of the supernatants was determined by the Bradford dye binding method. Aliquots of the resulting supernatants containing 2.0 mg of total protein were used for immunoprecipitation with antibodies against IR, IRS-2, and STAT-3 at 4 C overnight, followed by SDS-PAGE, transfer to nitrocellulose membranes, and blotting with antiphosphotyrosine, antiphosphoserine, anti-IR, anti-IRS-2, anti-STAT-3, or anti-p85/PI3 kinase. In direct immunoblot experiments, 0.2 mg of protein extracts were separated by SDS-PAGE, transferred to nitrocellulose membranes, and blotted with anti-TNF, anti-IL-1, anti-IL-6, anti-IR, anti-IRS-2, anti-STAT-3, anti-suppressor of cytokine signaling (SOCS)-3, anti-phospho-[Ser473]Akt, and anti-phospho-[Thr183]-JNK antibodies, as described (12). The protein loading of immunoblots was always evaluated by two methods: before blotting, the nitrocellulose membranes were stained with Coomassie Blue, and after blotting the membranes were reprobed with an antivimentin antibody.
Optic and confocal microscopy
Paraformaldehyde-fixed hypothalami were sectioned (5 μm) and used in regular single- or double-immunofluorescence staining using TNF, NeuN, or phospho-[Thr183]-JNK antibodies according to a previously described protocol (6, 13). Analysis and documentation of results were performed using an LSM 510 laser confocal microscope (Zeiss, Gttingen, Germany) and a BX60 microscope (Olympus, Tokyo, Japan). Semiquantitative assessment of NeuN, TNF and phospho-[Thr183]-JNK was achieved by counting the number of positively stained cells. Every second of all consecutive sections were counted in arcuate nucleus and lateral hypothalamus, sites of predominant staining for both TNF and phospho-[Thr183]-JNK. The anatomical correlations were made according to the landmarks given in a stereotaxic atlas (10). The topographical views of the regions to be studied were obtained by hematoxylin-eosin staining of consecutive sections.
Statistical analysis
The results of blots are presented as direct comparisons of bands in autoradiographs and quantified by densitometry using the Scion Image software (Scion Corp., Frederick, MD). All data, except those from immunohistochemistry and macroarray, were analyzed by the two-tailed unpaired Student’s t test or by repeat-measures ANOVA (one-way or two-way ANOVA) followed by post hoc analysis of significance (Bonferroni test) when appropriate, comparing experimental and control groups. In immunohistochemistry the average number of positive cells per section was obtained for each animal. Differences between groups were evaluated by the nonparametric Kruskal-Wallis and Mann-Whitney U test. The data are expressed as percent variation from respective control. In macroarray analysis, genes exhibiting differential expression in the preconditioning group were selected only if the hybridization signals were either increased or decreased by at least 2-fold, compared with those of the control group, and according to the statistical analysis performed by ANOVA. The functional clustering of genes was performed by combining the information obtained from public databases such as the National Institutes of Health (www.ncbi.nlm.nih.gov) and the Rat Genome Database (www.rgd.mcw.edu). In all experiments, the level of significance was set at P < 0.05.
Results and Discussion
To evaluate the role of HL feeding on hypothalamic protein expression, Wistar rats treated during 16 wk with a diet containing 45 kcal% fat were used for macroarray analysis of 1176 mRNAs in hypothalamus in comparison with rats treated with a regular rodent chow, containing 10 kcal% fat. HL feeding modulated the expression of 170 mRNAs specificities (92 up-regulated and 78 down-regulated; complete data presented in supplemental Table 1, published on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org), including hormones, growth factors, transcription factors, enzymes of lipid metabolism, and membrane and nuclear receptors among others. HL feeding modulated the expression of mRNAs encoding proteins known to participate in the control of feeding, thermogenesis, and other important metabolic functions. These mRNAs included the IR (5.42 ± 0.03-fold increase), leptin receptor ObR (8.89 ± 0.07-fold increase), CRH binding protein (3.52 ± 0.03-fold increase), and N-methyl-D-aspartate receptor 2B (3.86 ± 0.04-fold increase). All these mRNAs were functionally clustered as metabolic-related sequences, and the data are presented in supplemental Table 2. However, after complete functional clustering, we observed that immune-related proteins’ mRNAs represented the largest group modulated by HL feeding. Table 1 presents all mRNA specificities that encode immune-related proteins that were modulated by HL feeding.
In recent years, a bulk of evidence has been obtained to show that certain proinflammatory cytokines, produced by the adipose tissue, participate in the induction of insulin resistance in tissues that act as traditional targets for insulin action (14, 15). This phenomenon is mediated by the activation of intracellular serine kinases such as JNK and IB kinase, which act as intermediaries for proinflammatory signaling, but are also able to induce serine phosphorylation of key elements of the insulin signal transduction pathway and by this mechanism blunt insulin action (14, 15, 16). TNF, IL-1, and IL-6 have been described as inducers of insulin resistance by the aforementioned mechanism (14, 15, 16). In the macroarray analysis, the mRNAs of all these three cytokines were increased by the consumption of the HL diet. Therefore, we used distinct methods to evaluate the expression/concentration of these cytokines in the hypothalamus of rats. By ELISA (Fig. 1A), it was shown that the hypothalamic levels of TNF and IL-1 were significantly increased at 16 wk, compared with respective controls (Ctr16), whereas IL-6 was increased at 16 wk, compared with the basal control (Ctr0). By immunoblot (Fig. 1B), all three cytokines were shown to be significantly increased in the hypothalamus of rats fed on the HL diet for 16 wk, compared with respective controls (Ctr16) and basal controls (Ctr0). Moreover, by the same method, it was shown that the consumption of HL diet for 13 wk was also sufficient to promote a significant increase on TNF and IL-6 expression (Fig. 1B). In addition, by RT-PCR (Fig. 1C), the levels of mRNAs encoding for TNF and IL-1 were found to be significantly increased.
By confocal microscopy, we observed that low levels of TNF could be detected in the arcuate nucleus and lateral hypothalamus of rats fed on a regular diet (Fig. 1D). Feeding the rats with the HL diet for 16 wk promoted an increase in TNF in both hypothalamic nuclei. In the arcuate nucleus and lateral hypothalamus TNF-specific staining was found both coinciding and not coinciding with neuron bodies. Specifically in the lateral hypothalamus, the HL diet produced a 44 ± 12% increase (P < 0.05) in the number of TNF-stained neuron bodies. In other regions of the hypothalamus and the remainder of the brain, TNF staining was minimal. There were no significant differences in the number of NeuN-positive cells in lateral hypothalamus and arcuate nucleus between the groups.
To evaluate whether the higher concentration of proinflammatory cytokines in the hypothalamus may lead to the activation of proinflammatory intracellular signaling, we performed immunoblots to determine phospho-[Thr183]-JNK and ELISA to determine NFB activation. As depicted in Fig. 1E, the consumption of the HL diet for 13 or 16 wk significantly increased the levels of phospho-[Thr183]-JNK, whereas the consumption of HL diet for 16 wk activated NFB. No phospho-[Thr183]-JNK signal could be detected by confocal microscopy in rats fed on the control diet. However, specific staining was detected in several neuron bodies of the arcuate nucleus and lateral hypothalamus in HL rats (Fig. 1F).
The icv injection of a single dose of insulin inhibits 12 h spontaneous food intake. Most of this effect is dependent on insulin binding and activation of the IR present in neurons of the arcuate nucleus and activation of an intracellular cascade that depends on the engagement of IRS-2, PI3-kinase, and Akt (6, 7, 17). The treatment of rats with a HL diet during 16 wk did not modulate the protein expression of IR (Fig. 2A) and IRS-2 (Fig. 2B) but significantly reduced insulin-induced tyrosine phosphorylation of IR (Fig. 2A) and IRS-2 (Fig. 2B). This was accompanied by reduced insulin-induced association of IRS-2 with PI3-kinase (Fig. 2C) and reduced insulin-induced [Ser473]-phosphorylation of Akt (Fig. 2D). Because proinflammatory signaling is known to induce molecular resistance to insulin by promoting serine phosphorylation of key elements of the insulin signaling pathway (14, 15), we performed immunoblot analysis of IR and IRS-2 immunoprecipitates obtained from hypothalamic protein extracts of rats fed on control and HL diets. As depicted in Fig. 2E, both IR and IRS-2 presented higher levels of serine phosphorylation in samples from rats fed on HL diet, compared with the control rats. Finally, insulin-induced inhibition of spontaneous food intake was severely affected by the consumption of the HL diet (Fig. 2F), demonstrating that, in parallel to the molecular resistance to insulin, high-fat feeding negatively modulates a functional phenomenon controlled by insulin in the hypothalamus.
The reason the IR protein level was not modulated by HL feeding whereas IR mRNA was significantly increased is not known. We suspect that, as in other circumstances, proinflammatory events may induce higher protein turnover and differential regulation of transcription and translation (18). In this context, a recent study has shown that the cytokine signaling suppressors, SOCS-1 and -3, direct elements of the insulin signaling pathway to proteosomic degradation, and therefore, accelerate the rate of protein turnover (19). The fact that the IR protein amount was not effectively increased by HL feeding may have implications not only in insulin signal transduction in hypothalamic neurons but also in the rate of insulin transport across the blood-brain barrier. We have not measured this transport capacity, but previous studies have shown that, in animal models of obesity and insulin resistance, the binding of insulin to brain capillaries is impaired, suggesting the participation of this mechanism in the inappropriate action of insulin in the central nervous system (20, 21).
To test the hypothesis that proinflammatory signaling stimulated by HL feeding may be involved in the inhibition of insulin signaling in hypothalamus, we treated control and HL rats with a specific inhibitor of JNK, SP600125. The compound was icv injected and after dose optimization based on previous reports (22), we observed that an almost complete inhibition of [Thr183] phosphorylation of JNK was obtained in control and HL rats (Fig. 3A). The treatment with the compound was also able to significantly reduce daily caloric intake (Fig. 3B) and promote a loss of body mass (Fig. 3C). Mean body mass at the beginning of the period of treatment was 338 ± 24, 342 ± 26, 403 ± 20, and 405 ± 22 g for control, control treated with SP600125, HL, and HL treated with SP600125, respectively. At the end of the treatment period (8 d), the body mass was 350 ± 23, 345 ± 22, 420 ± 24, and 397 ± 19 g for control, control treated with SP600125, HL, and HL treated with SP600125, respectively. In control rats, the compound also exerted some effect promoting a significant reduction in daily caloric intake (Fig. 3B). However, this was of a significantly (P < 0.05) lower magnitude than in HL rats and led only to a reduction of the rate of body mass gain (Fig. 3C). Interestingly, the loss of body mass promoted by the treatment with SP600125 was, in part, due to a significant fall in intraabdominal fat, as determined by the measurement of epididymal fat mass (Fig. 3D).
Next, we evaluated the effect of icv injection of SP600125 on insulin signaling and action in the hypothalamus of rats fed on the HL diet. As depicted in Fig. 4, A and B, the compound exerted no effect on IR and IRS-2 protein expression. It did, however, revert the effect of HL feeding on insulin-induced tyrosine phosphorylation of IR and IRS-2 in such a way that no difference was found between control and rats fed on the HL diet and treated with the compound. Restoration of insulin signaling through PI3-kinase and Akt was also obtained by the icv injection of SP600125 in HL rats (Fig. 4, C and D). All these phenomena were accompanied by a reduction in Ser phosphorylation of IR and IRS-2, which returned to levels similar to those of the control (Fig. 4E). Moreover, the treatment of HL rats with SP600125 significantly increased insulin-induced inhibition of 12 h spontaneous food intake, whereas no effect of the compound was detected in control rats (Fig. 4F). To determine whether the effect of SP600125 on hypothalamic insulin signal transduction was dependent on the reduction in food intake and loss of body mass, we evaluated insulin-induced tyrosine phosphorylation of the IR and IRS-2 and [Ser 473]-phosphorylation of Akt in pair-fed, HL diet-treated rats. Pair feeding promoted a similar body mass loss to that produced by SP600125 treatment. However, as depicted in Fig. 4G, pair feeding was not enough to restore baseline insulin-induced molecular activation of these proteins. Therefore, we suspect that most of the effect of SP600125 is mediated by its specific action on JNK and not through indirect actions on food intake and body mass.
Hypercaloric and particularly hyperlipidic feeding is implicated as one of the most important environmental factors leading to obesity. Recent studies have shown that ingestion of a fat-rich diet leads to hypothalamic resistance to leptin (9, 23). This phenomenon is partially dependent on the immune-related signaling suppressor SOCS-3, which interacts with and inhibits elements of the ObR signaling pathway (24, 25). To evaluate whether the proinflammatory phenomenon, herein described, may somehow interact and modulate canonical leptin signaling, we determined SOCS-3 expression and leptin-induced activation of STAT-3 in hypothalamus. As shown in Fig. 4 h, the consumption of the HL diet for 16 wk promoted a significant increase in the hypothalamic expression of SOCS-3, which was not reverted by a 1-wk treatment with SP600125. In addition, HL diet consumption led to the inhibition of leptin-induced tyrosine phosphorylation of STAT-3. Once more, inhibition of JNK with SP600125 did not revert the inhibitory effect of HL feeding on the ability of leptin to induce STAT-3 molecular activation.
In the present study, we demonstrate that the consumption of a fat-rich diet modulates the expression of several hypothalamic mRNAs encoding proteins with immune-related activities. The ability of certain types of fatty acids to modulate the production of cytokines has been known for quite some time and both pro- and antiinflammatory outcomes have been described, depending on the type of fatty acid studied (26, 27). Distinct mechanisms seem to participate in fatty acid-dependent regulation of the immune response. These mechanisms range from the control of prostaglandin E2 and leukotriene B4 production, which may affect signal transduction through IB kinase (28, 29), to direct action on nuclear receptors controlling the production of immune-related factors (30). Although we cannot be sure whether the proinflammatory phenomenon, herein described, is a direct or indirect consequence of the consumption of fat-rich food, we believe our data have established an important link between a common environmental event, i.e. consumption of fat-rich food, and the generation of a microenvironmental phenomenon of inflammation at an anatomical site involved in the central control of metabolic functions. Moreover, this study shows that inhibition of a protein that plays a central role in one of the most important proinflammatory intracellular signaling pathways, i.e. JNK, improves metabolic parameters and restores hypothalamic insulin signaling in rats fed a HL diet. These data place local inflammation into a pivotal position in the pathophysiological mechanisms involved in diet-induced obesity and opens novel therapeutic perspectives for this medical condition.
Acknowledgments
We are grateful to Dr. Nicola Conran for English grammar review.
Footnotes
This work was supported by grants from Fundao de Amparo à Pesquisa do Estado de So Paulo and Conselho Nacional de Pesquisa.
1 C.T.D.S. and E.P.A. contributed equally to these studies.
Abbreviations: Ctr, Control; HL, hyperlipidic; icv, intracerebroventricular; IR, insulin receptor; IRS, insulin receptor substrate; JNK, c-Jun N-terminal kinase; NFB, nuclear factor-B; PI3-kinase, phosphatidylinositol 3-kinase; SOCS, suppressor of cytokine signaling; STAT, signal transducer and activator of transcription.
References
Flier JS 2004 Obesity wars: molecular progress confronts an expanding epidemic. Cell 116:337–350
Friedman JM 2000 Obesity in the new millennium. Nature 404:632–634
Friedman JM, Halaas JL 1998 Leptin and the regulation of body weight in mammals. Nature 395:763–770
Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671
Schwartz MW, Kahn SE 1999 Insulin resistance and obesity. Nature 402:860–861
Torsoni MA, Carvalheira JB, Pereira-Da-Silva M, de Carvalho-Filho MA, Saad MJ, Velloso LA 2003 Molecular and functional resistance to insulin in hypothalamus of rats exposed to cold. Am J Physiol Endocrinol Metab. 285:E216–E223
Carvalheira JB, Ribeiro EB, Araujo EP, Guimaraes RB, Telles MM, Torsoni M, Gontijo JA, Velloso LA, Saad MJ 2003 Selective impairment of insulin signalling in the hypothalamus of obese Zucker rats. Diabetologia. 46:1629–1640
Stein CJ, Colditz GA 2004 The epidemic of obesity. J Clin Endocrinol Metab. 89:2522–2525
El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, Flier JS 2000 Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obesity. J Clin Invest. 105:1827–1832
Paxinos G, Watson C 1997 The rat brain in stereotaxic coordinates. 4th ed. San Diego: Academic Press
Johnson AK, Epstein AN 1975 The cerebral ventricles as the avenue for the dipsogenic action of intracranial angiotensin. Brain Res. 86:399–418
Velloso LA, Kampe O, Hallberg A, Christmanson L, Betsholtz C, Karlsson FA 1993 Demonstration of GAD-65 as the main immunogenic isoform of glutamate decarboxylase in type 1 diabetes and determination of autoantibodies using a radioligand produced by eukaryotic expression. J Clin Invest. 91:2084–2090
Araujo EP, Amaral ME, Filiputti E, De Souza CT, Laurito TL, Augusto VD, Saad MJ, Boschero AC, Velloso LA, Carneiro EM 2004 Restoration of insulin secretion in pancreatic islets of protein-deficient rats by reduced expression of insulin receptor substrate (IRS)-1 and IRS-2. J Endocrinol. 181:25–38
Hotamisligil GS 2003 Inflammatory pathways and insulin action. Int J Obes Relat Metab Disord. 27(Suppl 3):S53–S55
Shoelson SE, Lee J, Yuan M 2003 Inflammation and the IKK/IB/NF-B axis in obesity- and diet-induced insulin resistance. Int J Obes Relat Metab Disord. 27(Suppl 3):S49–S52
Hotamisligil GS 2000 Molecular mechanisms of insulin resistance and the role of the adipocyte. Int J Obes Relat Metab Disord. 24(Suppl 4):S23–S27
Niswender KD, Morrison CD, Clegg DJ, Olson R, Baskin DG, Myers Jr MG, Seeley RJ, Schwartz MW 2003 Insulin activation of phosphatidylinositol 3-kinase in the hypothalamic arcuate nucleus: a key mediator of insulin-induced anorexia. Diabetes 52:227–231
Colangelo V, Schurr J, Ball MJ, Pelaez RP, Bazan NG, Lukiw WJ 2002 Gene expression profiling of 12633 genes in Alzheimer hippocampal CA1: transcription and neurotrophic factor down-regulation and up-regulation of apoptotic and pro-inflammatory signaling. J Neurosci Res. 70:462–473
Rui L, Yuan M, Frantz D, Shoelson S, White MF 2002 SOCS-1 and SOCS-3 block insulin signaling by ubiquitin-mediated degradation of IRS1 and IRS2. J Biol Chem. 277:42394–42398
Schwartz MW, Figlewicz DF, Kahn SE, Baskin DG, Greenwood MR, Porte Jr D1990 Insulin binding to brain capillaries is reduced in genetically obese, hyperinsulinemic Zucker rats. Peptides 11:467–472
Pardridge WM 1993 Transport of insulin-related peptides and glucose across the blood-brain barrier. Ann NY Acad Sci. 692:126–137
Nadruz Jr W, Kobarg CB, Kobarg J, Franchini KG 2004 c-Jun is regulated by combination of enhanced expression and phosphorylation in acute-overloaded rat heart. Am J Physiol Heart Circ Physiol. 286:H760–H777
Munzberg H, Flier JS, Bjorbaek C 2004 Region-specific leptin resistance within the hypothalamus of diet-induced-obese mice. Endocrinology 145:4880–4889
Howard JK, Cave BJ, Oksanen LJ, Tzameli I, Bjorbaek C, Flier JS 2004 Enhanced leptin sensitivity and attenuation of diet-induced obesity in mice with haploinsufficiency of Socs3. Nat Med. 10:734–738
Mori H, Hanada R, Hanada T, Aki D, Mashima R, Nishinakamura H, Torisu T, Chien KR, Yasukawa H, Yoshimura A 2004 Socs3 deficiency in the brain elevates leptin sensitivity and confers resistance to diet-induced obesity. Nat Med. 10:739–743
Wallace FA, Miles EA, Calder PC 2000 Activation state alters the effect of dietary fatty acids on pro-inflammatory mediator production by murine macrophages. Cytokine 12:1374–1379
Blok WL, Katan MB, van der Meer JW 1996 Modulation of inflammation and cytokine production by dietary (n-3) fatty acids. J Nutr. 126:1515–1533
Brach MA, de Vos S, Arnold C, Gruss HJ, Mertelsmann R, Herrmann F 1992 Leukotriene B4 transcriptionally activates interleukin-6 expression involving NK-B and NF-IL6. Eur J Immunol. 22:2705–2711
Rossi A, Kapahi P, Natoli G, Takahashi T, Chen Y, Karin M, Santoro MG 2000 Anti-inflammatory cyclopentenone prostaglandins are direct inhibitors of IB kinase. Nature 403:103–108
Chinetti G, Fruchart JC, Staels B 2000 Peroxisome proliferator-activated receptors (PPARs): nuclear receptors at the crossroads between lipid metabolism and inflammation. Inflamm Res. 49:497–505(Cláudio T. De Souza1, Eli)