isocaloric intake of a high-fat diet promotes insulin resistance and inflammation in wistar rats

Isocaloric intake of a high-fat diet promotes insulin resistance andinflammation in Wistar rats

Patrícia Silva Jacob1, Tatiane Mieko de Meneses Fujii1, Monica Yamada1, Maria Carolina Borges1, LucasCarminatti Pantaleão2, Primavera Borelli3, Ricardo Fock3 and Marcelo Macedo Rogero1*1Department of Nutrition, School of Public Health, University of Sao Paulo, Sao Paulo, Brazil2Department of Human Physiology, Institute of Biomedical Sciences, University of Sao Paulo, Sao Paulo, Brazil3Department of Clinical and Toxicological Analyzes, Faculty of Pharmaceutical Sciences, University of Sao Paulo, Sao Paulo, Brazil

The aim of this study was to investigate the effect of isocaloric intake from a high-fat diet (HFD) on insulin resistance and inflammation in rats.MaleWistar rats were fed on an HFD (n=12) or control diet (n=12) for 12 weeks. Subsequently, all animals were euthanized, and blood glucose,insulin, free fatty acids, C-reactive protein, lipid profile, cytokines and hepatic-enzyme activity were determined. Carcass chemical compositionwas also analyzed. During the first and the twelfth weeks of the experimental protocol, the oral glucose tolerance test and insulin tolerance testwere performed and demonstrated insulin resistance (P< 0.05) in the HFD group. Although food intake (g) was lower (P< 0.05) in the HFDgroup compared with the control group, the concentration of total cholesterol, low-density lipoprotein, C-reactive protein and liver weight wereall significantly higher. The kinase inhibitor of kB, c-Jun N-terminal kinase and protein kinase B expressions were determined in the liver andskeletal muscle. After an insulin stimulus, the HFD group demonstrated decreased (P=0.05) hepatic protein kinase B expression, whereas thekinase inhibitor of kB phospho/total ratio was elevated in the HFD muscle (P=0.02). In conclusion, the isocaloric intake from the HFD inducedinsulin resistance, associated with impaired insulin signalling in the liver and an inflammatory response in the muscle. Copyright © 2012 JohnWiley & Sons, Ltd.

key words—obesity; inflammation; high-fat diet; insulin resistance; rats

INTRODUCTION

High consumption of diets rich in lipids, especially saturatedfats, is associated with metabolic disorders, such as insulinresistance and dyslipidemia, conditions that promote thedevelopment of obesity. The pathophysiology of chronicdiseases is associated with low-intensity chronic inflammation.1

In this state, there is a greater release of proinflammatorycytokines, chemokines and growth and angiogenic factors.2

Adipose tissue is the main organ responsible for the synthesisand release of biomarkers, such as tumour necrosis factor a(TNF-a) and interleukin 6 (IL-6), and it is considered to bethe origin of the systemic inflammatory response and insulinresistance.3 However, hypertrophy of this tissue leads to areduction in plasma adiponectin levels.4

There is some evidence to support the hypothesis that, inboth humans and animals, the intake of a high-fat diet(HFD) induces proinflammatory protein expression in theliver5 and skeletal muscle.6 The inflammatory responseinvolves the activation of the nuclear factor kappa B pathwayby activating protein kinases, such as inhibitor of kB kinase

(IKK-b) and c-Jun N-terminal kinase (JNK). In the liver andskeletal muscle, IKK-b and JNK together impair insulinsignalling by phosphorylation of insulin receptor substrate(IRS) proteins on serine residues, consequently reducing theactivation of protein kinase B (AKT) and decreasing glucoseuptake, thereby contributing to insulin resistance.7,8 However,is not clear whether this mechanism is associated with thehigher consumption of lipids and/or with the excess of energy.Thus, the aim of this study was to investigate the effect ofan isocaloric intake of an HFD on insulin resistance andinflammation in rats. For this purpose, the ingredients(except for starch and lard) of the diet were normalized,according to their energy densities.

MATERIALS AND METHODS

Animals and treatment

Male Wistar rats (initial weight, 298� 27 g), aged 2 monthsold, were obtained from the Animal Laboratory of theFaculty of Medicine at the University of Sao Paulo. Rats werehoused in plastic cages (2–3 rats per cage) in an atmosphere of55%� 10% relative humidity at 22 �C� 2 �C, with a 12-hlight/12-h dark cycle (lights on at 07:00 h). Rats were givenfree access to food and water. Body mass and diet intake

*Correspondence to: Marcelo Macedo Rogero, Department of Nutrition,School of Public Health, University of Sao Paulo, Avenida Doutor Arnaldo,715, Sao Paulo, SP 01246-904, Brazil.E-mail: [email protected]

Received 1 July 2012Revised 19 August 2012

Accepted 20 August 2012Copyright © 2012 John Wiley & Sons, Ltd.

cell biochemistry and functionCell Biochem Funct (2012)Published online in Wiley Online Library(wileyonlinelibrary.com) DOI: 10.1002/cbf.2894

were recorded three times a week. The study was approvedby the Ethics Committee on Animal Experimentation of theInstitute of Tropical Medicine, University of Sao Paulo,according to the guidelines of the Brazilian College onAnimal Experimentation (protocol number 048/2009).After acclimatization for 10 days on a semipurified

diet, based on the American Institute of Nutrition recommen-dations for adult rodents (AIN-93M),9 rats were randomlyassigned to two groups: the HFD-fed group and the control(CON) diet–fed group. For 12 weeks, the CON group(n = 12) received the AIN-93M diet (total energy: 75.8%carbohydrates, 9.3% fat and 14.9% protein), whereas theHFD group (n = 12) received the AIN-93M-based dietenriched with lard (total energy: 24.2% carbohydrates,60.9% fat and 14.9% protein). In a pilot study, rats fromthe HFD group consumed approximately 30% less dietthan the rats of the CON group. However, given the higherenergy density of the HFD (5.55 kcal/g), compared withCON diet (3.99 kcal/g), the daily energy intake did notdiffer between the groups. The decrease in the amount of dietconsumed by the HFD group would result in a decreasedintake of several micro- and macronutrients, which couldseriously affect the outcome of the study. Thus, to ensurecomparable intakes of such micro- and macronutrientsbetween the groups, all diet ingredients in the HFD (exceptfor starch and lard) were adjusted to be present at thesame amount as in the CON diet per kilocalorie of diet, asshown in Table 1. The adjustment of HFD ingredientsbased on their energy content, rather than on its weight, hasalready been performed by others10–13. Furthermore, becauseof the greater susceptibility of lard to oxidation, we addeda higher amount of tert-butylhydroquinone to the HFD14

(Table 1).

Obtaining biological material

Two experiments were conducted as follows: experiment1—collection of blood, adipose pads, heart, liver and subsequentevaluation of the chemical composition of the carcass; andexperiment 2—infusion of insulin via the portal vein withsubsequent collection of the liver andmuscle tissues for analysis

of protein expression by Western blotting (liver and soleusmuscle) and for histology (liver). All animals were anesthe-tized with xylazine (50mg/kg) and ketamine (100mg/kg)administered subcutaneously using an insulin syringe with a20-G needle.15

In the first experiment, after anesthetization, the animalswere killed by decapitation using an appropriate guillotine.This procedure was performed quickly to minimize anyform of suffering or distress of animals. Euthanasia wascarried out in the morning, between 8 and 12 h. In thesecond experiment, a sample of the soleus muscle of the rightpaw was withdrawn (50mg), the abdominal cavity wasopened, and then a sample of liver tissue (100mg) wastaken from the superior right lobe. To stimulate the insulin sig-nalling pathway, 10 U of regular human insulin (100 U/ml;NovolinRW; Novo Nordisk, Montes, MG, Brazil) wereinjected via the portal vein with an insulin syringe with aneedle length of 8mm and diameter of 0.3mm. After 30 s,another sample from the same lobe was extracted (100mg),and at 90 s after the injection of insulin, a sample of the soleusmuscle of the left paw of the animal was obtained (50mg).16

After the samples were taken from the soleus muscle and liver,both were immediately stored in liquid nitrogen. Subsequently,the extraction of total proteins in both tissues was carriedout and used for evaluating the expressions of JNK, IKK-band AKT, in its total and phosphorylated forms, by theWestern blotting.

Blood analyses

Fasting blood glucose from the tail vein was determined usingan Accu-Chek Advantage II meter (Roche Diagnostics, Basel,Switzerland). The serum total cholesterol, high-densitylipoprotein (HDL) cholesterol and triacylglycerol (TAG)levels were quantified using commercial kits for the automaticmultichannel analyzer (Architect C8000; Abbott Diagnostics,Abbott Park, IL, USA). Low-density lipoprotein (LDL) chol-esterol and very low density lipoprotein (VLDL) cholesterollevels were estimated using the following equations proposedby Friedewald et al.17 Serum insulin, adiponectin, C-reactiveprotein (CRP), TNF-a and IL-6 levels were quantified usingthe Lincoplex kit (Linco Research Inc., St Charles, MO,USA). Plasma total fatty acid concentration was determinedusing a commercial kit (Catalog #100-K612; Biovision,Milpitas, CA, USA). The serum activities of alanine amino-transferase (ALT) and aspartate aminotransferase (AST)were quantified using commercial kits for the automatic multi-channel analyzer (Architect C8000; Abbott Diagnostics).

Liver histology

Fragments of dissected liver were immediately fixed in 10%formalin. After a period of at least 48 h, representativefragments from each sample were subjected to paraffinembedding. All the cuts were stained with haematoxilynand eosin.18 Another sample of the liver was used toperform the Sudan Black method to evaluate lipid infiltrationin the hepatocytes.19

Table 1. Composition of experimental diets (g/1000 kcal; 4190 kJ)

Ingredients CON diet* HFD

Cornstarch 37.14 7.09Sucrose 5.98 5.98Casein 8.38 8.38Soybean oil 2.39 2.39Lard 0.00 13.25Cellulose 2.99 2.99Mineral mix 2.09 2.09Vitamin mix 0.60 0.60L-Cysteine 0.11 0.11Choline bitartrate 0.15 0.15Tert-butylhydroquinone 0.0005 0.0017

*According to AIN-93M.9

p. s. jacob ET AL.

Copyright © 2012 John Wiley & Sons, Ltd. Cell Biochem Funct (2012)

Glucose tolerance and insulin resistance

After 7–8 h of fasting, blood was collected from the tail veinto determine the glucose concentration using an Accu-ChekAdvantage II glucometer (Roche Diagnostics). For the oralglucose tolerance test (oGTT), blood samples were collectedat fasting (time 0) and 30, 60, 90 and 120min after theadministration of solution of 20% glucose (2 g/kg of bodyweight) by oral gavage.15

For the intraperitoneal insulin test (ipITT), blood sampleswas collected at fasting (time 0) and 5, 10, 15, 20, 25, 30and 40min after intraperitoneal insulin injection (0.75U/kgof body weight).

For both tests, the area under the curve (AUC) was calcu-lated. Homeostasis model assessment (HOMA) was alsoused as an indicator of peripheral insulin resistance, usingblood glucose (mmol/l) multiplied by fasting insulin (mIU/ml)as parameters, and divided by 22.5.20

Carcass chemical composition analyses

Water, lipid, protein and ash contents present in the carcassof rats were determined as described elsewhere.21

Fatty acid compositions of the diets and lard

Fatty acid compositions were determined by gas chromatog-raphy. Analyses of FAMEs were carried out using an Agilent7890A series gas chromatograph (Agilent Technologies,Santa Clara, CA, USA) equipped with a split injection port,flame ionization detector and fused silica capillary columnDB-23 (J&W 122–2361) of 60� 0.25mm id, with a filmthickness of 0.15mm (Agilent Technologies). The samples(1.0ml) were injected in the split mode (split ratio, 1:50).Helium was used as a carrier gas, and the fatty acids wereseparated using a 1 �C/min gradient from 140 �C to 225 �C.The injector temperature was set at 250 �C, and the detectortemperature was set at 260 �C. Peak identification wasperformed by comparing the relative retention times withthose of a commercial standard mixture of FAME Supelco37 Component mix; 47885-U and C4-C24 Even Carbon49453-U (Sigma Chemical, St Louis, MO, USA). The results

were expressed as percentage of the total fatty acids present,as shown in Table 2.

Analysis of protein phosphorylation and expression byWestern blotting

To obtain protein lysates, 20mg of liver and 25mg of musclewere homogenized with an extraction buffer (potassiumphosphate buffer, saccharose, dithiothreitol (DTT), ethylenedia-minetetraacetic acid, phenylmethylsulfonyl fluoride (PMSF),sodium fluoride (NaF), phosphatase inhibitor I, phosphataseinhibitor II, protease inhibitor and ultrapure water). Laemmlibuffer (1� concentrate, 240mM Tris–HCl, 200mMmercaptoethanol, 0.8% SDS, 40% glycerol, 0.02%bromophenol blue) was added to the lysates in a 1:1 ratio.The protein extracts were separated by sodium dodecylsulfate–polyacrylamide gel electrophoresis, transferred tonitrocellulose membranes and blotted with primary antibodies(Cell Signaling, Danvers,MA,USA): anti-IKK-b, anti-pIKK-b(Ser 180), anti-JNK, anti-pJNK (Tyr 185), anti-AKT andanti-pAKT (Ser 473). The proteins were then stained withantirabbit IgG (Cell Signaling) in conjunction with horseradishperoxidase diluted 1:2000 in PBST buffer (8% NaCl, 0.2%KCl, 0.2% KH2HPO4, 1.15% Na2HPO4 and 0.5% Tween)with 3% skimmed milk powder for 1 h with stirring. Resultswere analyzed using a digital image system (Image Quant ™

400 version 1.0.0; Amersham Biosciences, Pittsburgh,PA, USA) with ImageQuant™software Capture software.We used b-actin as a normalizer protein (Sigma-AldrichW

Co., St Louis, MO, USA).

Statistical analyses

All variables were tested for normality of distribution using theKolmogorov–Smirnov test and for homogeneity of variances.To compare results regarding the effect of isocaloric intake ofHFD, we used unpaired Student’s t-test or one-way ANOVA(Tukey’s post hoc). In cases where normal distribution orvariance homogeneity could not be observed, we usednonparametric tests: Mann-Whitney U test or Kruskal-Wallistest (Dunn’s post hoc). Statistical analyses were per-formed using the GraphPad PrismW software version 5.01

Table 2. Fatty acid compositions (%) of the experimental diets and lard

Fatty acids CON HFD Lard

C 14:0 Mirystic 0.81� 0.01 1.29� 0.03 1.49� 0.02C 16:0 Palmitic 14.06� 0.03 23.25� 0.13 25.66� 0.75C 18:0 Stearic 4.42� 0.01 12.01� 0.22 12.77� 0.82C 18:1 c Oleic 27.09� 0.03 40.07� 0.06 41.90� 1.86C 18:1t Elaidic ND ND NDC 18:2 c n6 Linoleic 48.37� 0.07 21.88� 0.22 17.32� 0.24C 18:3 n3 Linolenic 4.96� 0.01 1.48� 0.03 0.75� 0.03C 18:3 n6 g-Linolenic 0.23� 0.00 ND NDC 20:5 n3 Eicosapentaenoic 0.06� 0.00 ND NDTotal saturated 19.29� 0.03 35.56� 0.20 39.92� 1.60Total monounsaturated 27.09� 0.03 40.07� 0.06 41.90� 1.86Total polyunsaturated 53.62� 0.06 23.37� 0.25 18.07� 0.27

Results are expressed as mean�SD. ND, nonidentified fatty acids.

isocaloric intake of a high-fat diet promotes insulin resistance and inflammation


(GraphPadSoftware Inc., La Jolla, CA, USA), and the level ofsignificance adopted was 0.05.

RESULTS

Effect of isocaloric HFD intake on body weight and weighgain

After 12weeks of the experimental protocol, the HFD groupdemonstrated no statistical differences in body weight(Figure 1A) or in the percentage of weight gain (Figure 1B)in comparison with the CON group.

Effect of isocaloric HFD intake on food consumption

With regard to food intake, the HFD group demonstrated asignificant reduction (24%) for food consumed for 12weeksin comparison with the CON group (Figures 1C and 1D).

However, the average weekly energy intake did notdiffer between the groups, except in the twelfth week of theprotocol (Figure 1E). In addition, the total energy intakeduring the experimental protocol did not differ between thegroups (Figure 1 F).

Effect of isocaloric HFD intake on adipose pads and liverweight

The weight of the periepididymal adipose tissue as well asthe sum of the fatty pads (Table 3) did not differ betweengroups, although there was a trend towards an increase inthe HFD group (P= 0.09 and P= 0.06, respectively). Inaddition, the retroperitoneal adipose tissue weight was sig-nificantly higher in the HFD group in comparison withthe CON group. The HFD group presented a significantincrease in liver weight, in relation to the CON group(Table 3).

0 2 4 6 8 10 1210

15

20

25

30CONHFD

** * *

* * * * * * **

C

Weeks

Fo

od

inta

ke (

g/d

ay)

CONHFD

0

500

1000

1500

2000

2500

**

D

Groups

Tota

l fo

od

inta

ke (

g)

0 1 2 3 4 5 6 7 8 9 10 11 120

200

400

600CONHFD

A

Weeks

Wei

gh

t (g

)

CONHFD

0

10

20

30

40B

Groups

Wei

gh

t g

ain

(%

)

0 2 4 6 8 10 1240

60

80

100

120CONHFD

*

E

Weeks CONHFD

0

2000

4000

6000

8000

10000F

Groups

Tota

l en

erg

y in

take

(kc

al)

day

Figure 1. (A) Weight curve, (B) body weight gain, (C) average weekly consumption in grams, (D) total intake in grams, (E) average weekly consumption inenergy and (F) total energy intake of Wistar rats fed ad libitum with a CON diet or an HFD until the 12th week of the experimental protocol. Values areexpressed as mean�SD (n= 6/group). *P< 0.05 versus the CON group

p. s. jacob ET AL.


Effect of isocaloric HFD intake on carcass chemicalcomposition

The CON and the HFD groups did not differ in relation tothe carcass chemical composition (Table 3).

Effect of isocaloric HFD intake on glucose tolerance andinsulin resistance

The HFD group presented higher blood glucose levels at alltimes during the oGTT (Figure 2A) and a higher AUC incomparison to the CON group (Figure 2B). Similarly, the data

for the ipITT indicated resistance to the action of insulin in theHFD group because the glucose curve (Figure 2C) and theAUC (Figure 2D) of the HFD group were significantly higherthan those of the CON group. Confirming the resultsdescribed earlier, Table 4 shows that the animals in theHFD group had significantly higher fasting glucose levels

Table 3. Fat pads, tissue weight and body composition of rats fed on aCON diet or an HFD for 12weeks

CON HFD

Periepididymal pad (g) 14.20� 7.60 15.60� 3.56Retroperitoneal pad (g) 11.70� 5.90 14.80� 5.20*Sum of adipose pads (g) 25.80� 13.30 30.40� 8.50Liver (g) 16.80� 2.18 19.93� 0.38*Moisture (%) 51.12� 5.13 48.55� 4.61Lean mass (%) 81.40� 2.51 79.74� 3.73Lipid (%) 18.60� 2.51 20.26� 3.73Protein (%) 22.95� 4.15 18.93� 9.44Ash (%) 4.15� 2.13 3.62� 2.84

Results are expressed as mean�SD (n= 6 group).*P< 0.05 versus the CON group.

CONHFD

0

2000

4000

6000

8000

*

B

Groups

AU

C

0 50 100 15080

100

120

140

160

180CONHFD

*

* * **

A

Time (min)

Glu

cose

(m

g/d

L)

CONHFD

0

200

400

600

800

1000 *D

Groups

AU

C

0 5 10 15 20 25 30 35 4040

60

80

100

120

140

160

CON

HFD

**

**

C

Time (min)

Glu

cose

(m

g/d

L)

Figure 2. Glycemic curve in times 0, 30, 60, 90 and 120min in (A) oGTT, (B) AUC of oGTT, (C) glycemic curve in times 0, 5, 10, 15, 20, 25, 30 and 40minin ipITT and (D) AUC of ipITT of Wistar rats fed ad libitum with a CON diet or an HFD until the 12th week of the experimental protocol and submitted atoGTT. Values are expressed as mean�SD (n= 6/group). *P< 0.05 versus the CON group

Table 4. Blood biomarkers from rats fed on a CON diet or an HFD for12weeks

Parameters CON HFD

Glucose (mmol/l) 4.92� 0.43 5.52� 0.33*Insulin (mU/ml) 23.5� 16.0 22.4� 5.1HOMA-IR 5.27� 3.99 5.56� 1.55Total cholesterol (mg/dl) 68.0� 7.1 79.0� 6.6*HDL-c (mg/dl) 22.0� 2.6 20.0� 1.4LDL-c (mg/dl) 27.0� 7.5 38.0� 5.4*TAG (mg/dl) 96.0� 20.1 105.0� 20.4VLDL (mg/dl) 19.0� 4.0 21.0� 4.1Free fatty acids (nM/ml) 0.10� 0.12 0.10� 0.11ALT (U/l) 31.0� 7.3 35.0� 6.1AST (U/l) 106� 25 122� 17Leptin (pg/ml) 3,740� 2,036 3,389� 817IL-6 (pg/ml) 15.50� 4.51 21.68� 10.71TNF-a (pg/ml) 3.94� 0.82 4.13� 1.61Adiponectin (ng/ml) 13,514� 5,784 10,142� 3,373CRP (ng/ml) 611� 260 1,002� 168*

Results are expressed as mean�SD (n= 6/group).*P< 0.05 versus the CON group.



compared with animals of the CON group; however, therewas no difference in fasting serum insulin concentration andHOMA-IR.

Effect of isocaloric HFD intake on lipid profile and free fattyacids

HFD resulted in a statistically significant increase inserum total cholesterol and LDL cholesterol in relationto values for the CON diet group (Table 4). The groupsdid not differ significantly in relation to HDL-cholesterol,VLDL cholesterol and TAG. There was no statisticaldifference between the groups in relation to plasma totalfatty acids concentration.

Effect of isocaloric HFD intake on hepatic enzymes, CRPand cytokines

There were no effects of the HFD on serumALT, AST, leptin,IL-6, TNF-a and adiponectin levels; however, the serum CRPconcentration was higher (P=0.02) in the HFD group inrelation to the CON group (Table 4).

Effect of isocaloric HFD intake on hepatic lipid infiltration

Haematoxilyn and eosin stain demonstrated that one third ofthe HFD group showed discrete microgoticular degeneration,indicating lipid infiltration, as confirmed by the Sudan blackmethod (data not shown).

Effect of isocaloric HFD intake on the hepatic expression ofproteins

The CON and HFD groups did not present any significantdifferences with regard to the expression or phosphorylationof the inflammatory proteins, JNK and IKK-b, in the liver(Figures 3A–3D). However, there was a significant reductionin AKT phosphorylation in the liver of the HFD group afterinsulin stimulation when compared with the respective CONgroup (Figure 3H); these results corroborate the occurrenceof insulin resistance, as previously described.

Effect of isocaloric HFD intake on skeletal muscleexpression of proteins

IKK-b protein expression was determined in the muscletissue in its total (Figure 4A) and phosphorylated form(Figure 4B), and no statistical difference was observedbetween the groups. However, the ratio between the phos-phorylated and the total IKK-b (Figure 4C) was significantlyhigher (P= 0.02) in the HFD group in comparison with theCON group. We also observed that there was no significantdifference between the groups with regard to the expressionsof JNK and AKT.

DISCUSSION

Epidemiological studies suggest a positive associationbetween dietary fatty acid content with weight gain and

development of metabolic disorders, such as glucoseintolerance.22,23 Both the quantity and quality of fat intake,especially saturated fat, can influence insulin sensitivity, whichis associated with the aetiology of chronic morbidities such astype 2 diabetes mellitus and cardiovascular disease.24,25

Palmitic and stearic fatty acids are the common saturatedfatty acids present in the occidental diet,26,27 and these arepresent in high concentrations in lard, which was used in thepresent study to induce resistance to the action of insulin in rats.However, unlike observations from several studies using

this model,28,29 animals fed on an HFD showed no significantweight gain. First, it should be highlighted that the HFDingredients were adjusted for energy density to meet thenutrient recommendations of the experimental animals. Thisadjustment is extremely important to avoid a lower intake ofvitamins and minerals in the HFD group compared with theCON diet animals.30,31 Thus, the energy adjustment densityprovided an isocaloric intake between the groups, which, asexpected, did not cause any significant weight gain butresulted in significant metabolic changes.Safwat et al.13 used the same type of adjustment for the

preparation of the HFD (57% of calories from lipids), andafter 18weeks, they observed a significantly higher(~20%) energy intake in the HFD group. In the same study,TAG and LDL levels were also increased, as well as theaccumulation of fat in the hepatocytes, without any changein the liver weight. Also, the authors have not found alterationsin plasma glucose, adiponectin or TNF-a levels whencomparing the HFD group with the CON group.Although our study used the same type of adjustment for

the preparation of the HFD, we found important metabolicchanges. In this context, the HFD group showed higherfasting glucose, CRP, total cholesterol and LDL cholesterollevels as well as higher blood glucose levels in the oGTTand ipITT in comparison to the CON group. In addition,the HFD group showed discrete microgoticular degener-ation, indicating lipid infiltration, in comparison with theCON group. We hypothesized that the presence of dietarysaturated fat, per se, was the key factor in the inflammatoryresponse.With regard to the isocaloric intake between the groups, it

should be noted that HFD can induce anorexigenicmechanisms,which can increase the expression of anorexigenic peptidessuch as alpha-melanocyte stimulating hormone in the arcuatenucleus of the hypothalamus, which might explain theabsence of any hyperphagic behaviour in the HFD groupin the present study.32 Accordingly, De Meijer et al.33 testedthe hypothesis that several metabolic changes, includingincreased adipose pads, result from ingested lipid contentand not excess energy. For this purpose, the animals weregrouped in CON, HF and pair-fed HFD groups. The authorsobserved that, even after isoenergetic consumption, theanimals in this latter group presented significantly increasedweights for the inguinal, mesenteric, retroperitoneal andperiepididymal pads. In addition, Petro et al.34 designed astudy to separate the effects of fat from those of excess caloricconsumption in mice, and after 11weeks of the experimentalprotocol, the high-fat-restricted diet group demonstrated

p. s. jacob ET AL.


increased weight and glucose levels, but no alteration ininsulin, in comparison with the low-fat diet group.

The effect of the ingested lipid content on the metabolicchanges associated with HFD feeding may be related to thefact that overfeeding fat provokes a reduction in carbohydrateoxidation and no change in fat oxidation. In this context, theinability of the organism to adjust its metabolism to attenuatethe effect of changes in fat intake on fat balance provides ametabolic explanation for the epidemiologic finding thatobesity is enhanced by dietary fat intake.35 Furthermore, HFDcan result in a reduction in themetabolic rate, shifting the energyequation towards energy storage because rats fed on a 59% fatdiet demonstrated a significant reduction in energy expend-iture at 30 days in comparison with the CON group.36

In the present study, we did not observe significant changesin body weight, energy intake or body composition; however,fasting glucose was increased in the HFD group and this factmay lead to a resistance to insulin, changes in lipid profileand increased production and release of proinflammatorycytokines, also known as meta-inflammation.37 To evaluateinsulin resistance in the animals, oGTT and ipITT wereconducted. Glucose intolerance and insulin resistance werestrongly evident after 12weeks of the experimental protocolin the animals submitted to the HFD. For the oGTT, this resultis indicative of a decrease in the hepatic response of hepaticgluconeogenesis inhibition as well as a reduction in glucoseuptake by the muscles and the pancreatic secretion of insulin.Under normal conditions, b-pancreatic cells exposed to a high

CONHFD

0.0

0.5

1.0

1.5

2.0

D

Groups

Tota

l JN

K (

AU

)

Total JNK

actin

CONCON

HFDHFD

0.0

0.5

1.0

1.5

2.0

Insulin - + +-

I

Groups

Ph

osp

ho

AK

T/T

ota

l AK

T (

AU

)

CONCON

HFDHFD

0.0

0.5

1.0

1.5

2.0

Insulin - + +-

G

Groups

Tota

l AK

T (

AU

)

Total AKTactin

Phospho AKTactin

CONCON

HFDHFD

0.0

0.5

1.0

1.5

2.0

2.5

a,c

b

c

d

Insulin - + +-

H

Groups

Ph

osp

ho

AK

T (

AU

)

Total IKK

actin

CONHFD

0.0

0.5

1.0

1.5

2.0

A

Groups

Tota

l IK

K-b

eta

(AU

)

CONHFD

0.0

0.5

1.0

1.5

C

Groups

Ph

osp

ho

IKK

-bet

a/To

tal I

KK

-bet

a (A

U)Phospho IKK

actin

CONHFD

0.0

0.5

1.0

1.5

B

Groups

Ph

osp

ho

IKK

-bet

a (A

U)

CONHFD

0.0

0.5

1.0

1.5

E

Groups

Ph

osp

ho

JN

K(A

U)

Phospho JNK

actin

CONHFD

0.0

0.5

1.0

1.5

F

Groups

Ph

osp

ho

JN

K/T

ota

l JN

K (

AU

)

Figure 3. Expression of (A) total IKK-b, (B) phospho-IKK-b (Ser 180) and (C) ratio of phospho-IKK-b by total IKK-b, (D) total JNK, (E) phospho-JNK (Tyr185), (F) ratio of phospho-JNK by total JNK, (G) total AKT, (H) phospho-AKT (Ser 473) and (I) ratio of phospho-AKT by total AKT in hepatic tissue ofWistar rats fed ad libitum with a CON diet or an HFD until the 12th week of the experimental protocol. Values are expressed as mean�SD (n= 6/ group).Bars with different letters differ significantly (P< 0.05). AU, arbitrary units



concentration of glucose and free fatty acids demonstratean increased insulin release; however, the chronic exposureto free fatty acids may result in a low level inhibition of thesecretion of this hormone.38 Furthermore, the results forthe ipITT suggest the development of insulin resistance inthe muscles and in the adipose tissue.To understand the mechanisms related to the effect of HFD

on hepatic response, we have analyzed the phosphorylationand expression of the protein kinases IKK-b and JNK, whichhave been found to serve as critical molecular links betweenobesity, metabolic inflammation and disorders of glucosehomeostasis.39 We have also analyzed the phosphorylation

and expression of AKT, which is a relevant protein in theactivation of the insulin pathway.40 It should be noted thatthe suppression of the JNK pathway in the liver improvesinsulin resistance in the whole body and markedly amelioratesglucose intolerance in diabetic mice,41 whereas the IKK-boverexpression in hepatocytes causes local and systemicinduction of proinflammatory genes and systemic insulinresistance in the absence of obesity.42 A possible mechanismrelated to the insulin resistance induced by IKK-b and JNK isthe ability to directly induce serine phosphorylation of IRS1,decrease insulin-stimulated tyrosine phosphorylation of IRS1and inhibit insulin action.43 In our study, HFD did not increase

E

CONHFD

0.0

0.5

1.0

1.5

Groups

Ph

osp

ho

JN

K (

AU

)F

CONHFD

0.0

0.5

1.0

1.5

Groups

Ph

osp

ho

JN

K /

Tota

l JN

K

C

CONHFD

0

1

2

3

4

*

GroupsPh

osp

ho

IKK

bet

a/ T

ota

l IK

K b

eta

B

CONHFD

0

1

2

3

4

Groups

Ph

osp

ho

IKK

bet

a (A

U)

A

CONHFD

0.0

0.5

1.0

1.5

Groups

Tota

l IK

K b

eta

(AU

)

D

CONHFD

0.0

0.5

1.0

1.5

Groups

Tota

l JN

K (

AU

)

G

CONCON

HFDHFD

0.0

0.5

1.0

1.5

2.0

Insulin - + - +

Groups

To

tal A

KT

(A

U)

CONCON

HFDHFD

0

2

4

6

Insulin - + - +

Groups

Ph

osp

ho

AK

T (A

U)

H I

CONCON

HFDHFD

0.0

0.5

1.0

1.5

2.0

Insulin - + +-

Groups

Ph

osp

ho

AK

T/T

ota

l AK

T (

AU

)

Figure 4. Expression of (A) total IKK-b, (B) phospho-IKK- b (Ser 180), (C) ratio of phospho-IKK- b by total IKK-b, (D) total JNK, (E) phospho-JNK(Tyr 185), (F) ratio of phospho-JNK by total JNK, (G) total AKT, (H) phospho-AKT (Ser 473) and (I) ratio of phospho-AKT by total AKT in skeletal musculartissue of Wistar rats fed ad libitum with a CON diet or an HFD until the 12th week of the experimental protocol. Values are expressed as mean�SD(n = 4–6/ group). AU, arbitrary units. *P< 0.05 versus the CON group

p. s. jacob ET AL.


the expression or phosphorylation of the inflammatoryproteins, JNK and IKK-b, in the liver; however, AKT phos-phorylation in the liver was reduced in animals thatconsumed isocaloric HFD after 12 weeks, when stimulatedwith insulin and comparedwith the CONgroup. Indeed, the factthat there was no activation of the investigated inflammatoryproteins in the liver suggests that the reduction in AKTphosphorylation could be due to the activation of otherproteins such as protein kinase R, the ribosomal proteinS6K, protein kinase C, the extracellular signal-regulatedkinases and mammalian target of rapamycin, which also havethe capacity of inhibiting the phosphorylation of the IRS-1 intyrosine 1222.44

Forkhead box O (FOXO) transcription factors play animportant role in modulating metabolic functions. FOXO isregulated by several modifications, but one of the most criticalis phosphorylation and nuclear exclusion by AKT.45

As FOXO1 promotes the expression of gluconeogenicenzymes,46 the lower AKT phosphorylation in the HFD groupmay be associated with the decrease in hepatic gluconeogenesisinhibition and consequent increase in blood glucose levels inthis group. Furthermore, the chronic failure to CON thehepatic glucose production leads to hyperglycemia andcompensatory stimulation of insulin secretion by the pancreaticb-cells, which exacerbates peripheral insulin resistance inmuscle and adipose tissue.46,47 This fact may explain ourresults, as the HFD group presented higher blood glucoselevels at all times during the oGTT and higher fasting glucoselevels in comparison with the CON group; however, there wasno difference in the fasting serum insulin concentration. Theseresults suggest a compensatory stimulation of insulin secretionby the pancreatic b-cells, induced by hyperglycemia in theHFD group.

Another study showed the influence of inflammation on theestablishment of insulin resistance in rats submitted to HFDfor short- and long-term periods.48 Interestingly, after 3 daysof HFD consumption, animals showed impaired glucosetolerance, compared with CON animals. Similarly, this occur-rence was observed after 10weeks of HFD consumption.Using the euglycemic–hyperinsulinemic clamp, the authorsdetected that the fasting glucose and hepatic glucose produc-tion were increased after 3 days of the experimental protocolin HFD rats. All these results are consistent with the hypothesisthat hepatic glucose production is the determinant key factor inbasal hyperglycemia.

The evaluation of inflammatory protein phosphorylationand expression in the liver and skeletal muscles is extremelyimportant because these proteins are involved in the develop-ment of metabolic disorders such as insulin resistance.49

Although no significant alteration in JNK expression wasobserved in either tissues, the ratio between the phosphorylatedand the total IKK-b in the muscle was greater in the HFDgroup in comparison with the CON diet. This finding indicatesthat important inflammatory signalling pathways wereactivated by the HFD and may imply their involvement inthe hyperglycemia, dyslipidemia and increase in adiposityobserved. A study conducted by Yaspelkis et al.6 showed thatHFD administration to Sprague–Dawley male rats for

12weeks increased IKK-b phosphorylation, indicating theactivation of this protein in the HFD group. Bikman et al.15

also reported an increase of IKK-b phosphorylation in the skel-etal muscle of rats submitted to HFD.

In conclusion, the present study showed that the isocaloricintake of HFD increased body adiposity, resulting in ahyperglycemia, glucose intolerance and a dyslipidemic stateas well as an increase in the inflammatory response. Inaddition, HFD decreased the phosphorylation of AKT in theliver, promoting the development of insulin resistance, andincreased the muscular ratio of phosphorylated IKK-b-to-total IKK-b in response to the inflammatory process in thistissue. All these results suggest that the isocaloric intake ofHFD induced insulin resistance and glucose intolerance,which may be associated with an inflammatory response inthe muscle and impairments in hepatic insulin signalling.

CONFLICT OF INTEREST

The authors have declared that there is no conflict of interest.

ACKNOWLEDGEMENTS

This work was financially supported by the Fundação deAmparo à Pesquisa do Estado de São Paulo (FAPESP)(09/54395-0). The authors thank I.S.O. Pires and M.C.Ferreira for technical assistance.

REFERENCES

1. Ferguson LR, Philpott M. Nutrigenomics and chronic inflammation. InIn Personalized Nutrition: principles and applications, Bouwan L,Desiere F (eds.) eds. CRC Press: New York, 2007; 49–60.

2. Beutler B, Jiang Z, Georgel P, et al. Genetic analysis of host resistance:Toll-like receptor signaling and immunity at large. Annu Rev Immunol2006; 24: 353–89.

3. Steinberg GR. Inflammation in obesity is the common link betweendefects in fatty acid metabolism and insulin resistance. Cell Cycle2007; 6(8): 888–94.

4. Cancello R, Clément K. Is obesity an inflammatory illness? Role oflow-grade inflammation and macrophage infiltration in human whiteadipose tissue. BJOG 2006; 113(10): 1141–7.

5. Tobar N, Oliveira AG, Guadagnini D, et al. Diacerhein improvesglucose tolerance and insulin sensitivity in mice on a high-fat diet.Endocrinology 2011; 152(11): 4080–93.

6. Yaspelkis BB, 3rd, Kvasha IA, Figueroa TY. High-fat feedingincreases insulin receptor and IRS-1 coimmunoprecipitation withSOCS-3, IKKalpha/beta phosphorylation and decreases PI-3 kinaseactivity in muscle. Am J Physiol Regul Integr Comp Physiol 2009;296(6): R1709–15.

7. Gao Z, Hwang D, Bataille F, et al. Serine phosphorylation of insulinreceptor substrate 1 by inhibitor kappa B kinase complex. J Biol Chem2002; 277(50): 48115–21.

8. Zick Y. Ser/Thr phosphorylation of IRS proteins: a molecular basis forinsulin resistance. Sci STKE 2005; 2005(268): pe4.

9. Reeves PG, Nielsen FH, Fahey GC, Jr. AIN-93 purified diets forlaboratory rodents: final report of the American Institute of Nutritionad hoc writing committee on the reformulation of the AIN-76A rodentdiet. J Nutr 1993; 123(11): 1939–51.

10. Tallman DL, Taylor CG. Effects of dietary fat and zinc on adiposity,serum leptin and adipose fatty acid composition in C57BL/6J mice.J Nutr Biochem 2003; 14(1): 17–23.



11. Kishino E, Ito T, Fujita K, et al. A mixture of the Salacia reticulata(Kotala himbutu) aqueous extract and cyclodextrin reduces theaccumulation of visceral fat mass in mice and rats with high-fatdiet-induced obesity. J Nutr 2006; 136(2): 433–9.

12. Jobgen W, Meininger CJ, Jobgen SC, et al. Dietary L-arginine supple-mentation reduces white fat gain and enhances skeletal muscle and brownfat masses in diet-induced obese rats. J Nutr 2009; 139(2): 230–7.

13. Safwat GM, Pisanò S, D’Amore E, et al. Induction of non-alcoholicfatty liver disease and insulin resistance by feeding a high-fat diet inrats: does coenzyme Q monomethyl ether have a modulatory effect?Nutrition 2009; 25(11–12): 1157–68.

14. Pang J, Choi Y, Park T. Ilex paraguariensis extract ameliorates obesityinduced by high-fat diet: potential role of AMPK in the visceraladipose tissue. Arch Biochem Biophys 2008; 476(2): 178–85.

15. Bikman BT, Woodlief TL, Noland RC, et al. High-fat diet inducesIKKbeta and reduces insulin sensitivity in rats with low running capacity.Int J Sports Med 2009; 30(9): 631–5.

16. Kubota N, Kubota T, Itoh S, et al. Dynamic functional relay betweeninsulin receptor substrate 1 and 2 in hepatic insulin signaling duringfasting and feeding. Cell Metab 2008; 8(1): 49–64.

17. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concen-tration of low-density lipoprotein cholesterol in plasma, without useof the preparative ultracentrifuge. Clin Chem 1972; 18(6): 499–502.

18. Michalany J. The surgeon, histologic technics and surgical pathology.AMB Rev Assoc Med Bras 1979; 25(1): 44–6.

19. Pearse AGE. Histochemistry. Theoretical and applied: Little, Brown,Boston 1968.

20. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,Turner RC. Homeostasis model assessment: insulin resistance andbeta-cell function from fasting plasma glucose and insulin concentrationsin man. Diabetologia 1985; 28(7): 412–9.

21. Rogero MM, Tirapegui J, Vinolo MA, et al. Dietary glutamine supple-mentation increases the activity of peritoneal macrophages andhemopoiesis in early-weaned mice inoculated withMycobacterium bovisbacillus Calmette-Guérin. J Nutr 2008; 138(7): 1343–8.

22. Bray GA, Popkin BM. Dietary fat intake does affect obesity! Am J ClinNutr 1998; 68(6): 1157–73.

23. Maskarinec G, Takata Y, Pagano I, et al. Trends and dietarydeterminants of overweight and obesity in a multiethnic population.Obesity (Silver Spring) 2006; 14(4): 717–26.

24. Speakman J, Hambly C, Mitchell S, et al. Animal models of obesity.Obes Rev 2007; 8(Suppl 1): 55–61.

25. Winzell MS, Ahrén B. The high-fat diet-fed mouse: a model for studyingmechanisms and treatment of impaired glucose tolerance and type 2diabetes. Diabetes 2004; 53(Suppl 3): S215–9.

26. Okere IC, Chandler MP, McElfresh TA, et al. Differential effects ofsaturated and unsaturated fatty acid diets on cardiomyocyte apoptosis,adipose distribution, and serum leptin. Am J Physiol Heart CircPhysiol 2006; 291(1): H38–44.

27. MancoM, CalvaniM,MingroneG. Effects of dietary fatty acids on insulinsensitivity and secretion. Diabetes Obes Metab 2004; 6(6): 402–13.

28. Yang Y, Zhou L, Gu Y, et al. Dietary chickpeas reverse visceraladiposity, dyslipidaemia and insulin resistance in rats induced by achronic high-fat diet. Br J Nutr 2007; 98(4): 720–6.

29. Woods SC, Seeley RJ, Rushing PA, D’Alessio D, Tso P. A controlledhigh-fat diet induces an obese syndrome in rats. J Nutr 2003; 133(4):1081–7.

30. Cohen-Lahav M, Shany S, Tobvin D, et al. Vitamin D decreasesNFkappaB activity by increasing IkappaBalpha levels. NephrolDialTransplant 2006; 21(4): 889–97.

31. Wintergerst ES, Maggini S, Hornig DH. Contribution of selectedvitamins and trace elements to immune function. Ann Nutr Metab2007; 51(4): 301–23 Epub 2007 Aug 28.

32. Tian DR, Li XD, Shi YS, et al. Changes of hypothalamic alpha-MSHand CART peptide expression in diet-induced obese rats. Peptides2004; 25(12): 2147–2153.

33. De Meijer VE, Le HD, Meisel JA, et al. Dietary fat intake promotesthe development of hepatic steatosis independently from excesscaloric consumption in a murine model. Metabolism 2010; 59(8):1092–105.

34. Petro AE, Cotter J, Cooper DA, et al. Fat, carbohydrate, and calories inthe development of diabetes and obesity in the C57BL/6 J mouse.Metabolism 2004; 53(4): 454–457.

35. Schutz Y, Flatt JP, Jéquier E. Failure of dietary fat intake to promotefatoxidation: a factor favoring the development of obesity. Am J ClinNutr 1989; 50(2): 307–314.

36. Wilsey J, Zolotukhin S, Prima V, Scarpace PJ. Central leptin genetherapy fails to overcome leptin resistance associated with diet-induced obesity. Am J Physiol Regul Integr Comp Physiol 2003;285(5): R1011–1020.

37. Hotamisligil GS. Inflammation and metabolic disorders. Nature 2006;444(7121): 860–7.

38. Mingrone G. Dietary fatty acids and insulin secretion. Food Nut Res2006; 50: 79–84.

39. Solinas G, Karin M. JNK1 and IKKbeta: molecular links betweenobesity and metabolic dysfunction. FASEB J 2010; 24(8): 2596–611.

40. Arkan MC, Henever AL, Greten FR, et al. IKK-blinks inflammation toobesity induced insulin resistance. Nat Med 2005; 11(2): 191–198.

41. Nakatani Y, Kaneto H, Kawamori D, et al. Modulation of the JNKpathway in liver affects insulin resistance status. J Biol Chem 2004;279(44): 45803–9.

42. Cai D, Yuan M, Frantz DF, et al. Local and systemic insulin resistanceresulting from hepatic activation of IKK-beta and NF-kappaB. NatMed 2005; 11(2): 183–190.

43. Jiang S,Messina JL. Role of inhibitorykBkinase and c-JunNH2-terminalkinase in the development of hepatic insulin resistance in criticalillness diabetes. Am J Physiol Gastrointest Liver Physiol 2011;301(3): G454–63.

44. Boura-Halfon S, Zick Y. Phosphorylation of IRS proteins, insulinaction, and insulin resistance. Am J Physiol Endocrinol Metab 2009;296(4): E581–91.

45. Gross DN, Wan M, Birnbaum MJ. The role of FOXO in the regulationof metabolism. Curr Diab Rep 2009; 9(3): 208–14.

46. Cheng Z, White MF. Targeting Forkhead box O1 from the concept tometabolic diseases: lessons from mouse models. Antioxid Redox Signal2011; 14(4): 649–61.

47. White MF. Insulin signaling in health and disease. Science 2003; 302(5651): 1710–1711.

48. Lee YS, Li P, Huh JY, Hwang IJ, et al. Inflammation is necessary forlong-term but not short-term high-fat diet-induced insulin resistance.Diabetes 2011; 60(10): 2474–83.

49. Bulló M, Casas-Agustench P, Amigó-Correig P, et al. Inflammation,obesity and comorbidities: the role of diet. Public Health Nutr 2007;10(10A): 1164–72.

p. s. jacob ET AL.


isocaloric intake of a high-fat diet promotes insulin resistance and inflammation in wistar rats

Documents