Four-week-old male C57BL/6J mice (CRIFA, Spain) weighing 16–18g were housed under controlled light (12-h light/dark cycles from 8am to 8pm) and temperature (22–24°C) conditions with standard food and water ad libitum. After one week, animals with similar average body weight (BW) were divided into two groups and housed 8–10 per cage and assigned either to a low-fat or to a high-fat diet. Low-fat (10kcal% fat, 70kcal% carbohydrates and 20kcal% protein; 3.85kcal/g) or high-fat (45kcal% fat, 35kcal% carbohydrates, 20kcal% protein; 4.73kcal/g) diets were supplied by Research Diets (USA) and will be referred to as LF and HF, respectively. HF and their respective LF control mice had free access to food either during 8, 14 or 32 weeks. Diet composition (provided by the manufacturer) appears summarized in Table 1. BW and food intake were monitored twice per week. On the last day, mice were weighed and killed by decapitation between 8am and 9am. Blood was collected in chilled EDTA-coated polypropylene tubes and organs immediately dissected and weighed. Plasma samples were frozen for biochemical determinations. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH publication no. 85-23, revised 1996).

Recombinant murine leptin (1mg/kg) or saline was administered at 9am. After 90min, mice were killed by decapitation and liver, and lumbar and mesenteric adipose tissues were dissected and stored at −80°C until assay.

Plasma leptin concentration was analysed using a specific radioimmunoassay (RIA) kit for murine leptin (Linco Research, USA) (4.9% intra-assay variation, 3.3% inter-assay variation). Insulin was determined by means of a specific EIA kit for mouse insulin (Mercodia, Denmark) (2.2% intra-assay variation, 4.9% inter-assay variation). Glucose was measured by a spectrophotometric method (Glucose Trinder Method, Roche, Spain). Triglycerides and free fatty acids were determined using the GPO (Biolabo, France) and ACS-ACOD (Wako, Germany) methods, respectively.

Lipids were extracted from liver and mesenteric adipose tissue in chloroform/methanol (2/1) following the method of Folch7 with modifications.8

pSTAT3 was measured in liver and lumbar adipose tissue. Briefly, tissues were homogenized in ice-cold buffer containing 0.42M NaCl, 20mM Hepes (pH 7.9), 1mM Na4P2O7, 1mM EDTA, 1mM EGTA, 1mM dithiothreitol, 20% glycerol, 1μg/ml aprotinin, 1μg/ml leupeptin, 20mM sodium fluoride, 1mM trisodium orthovanadate and 2mM phenylmethylsulfonyl fluoride. Tubes containing homogenates were frozen at −80°C and thawed at 37°C three consecutive times, then centrifuged for 10min at 4°C. Equivalent amounts of proteins (50μg) present in the supernatant were loaded in Laemli buffer (50mM Tris pH=6.8, 10% SDS, 10% glycerol, 5% mercaptoethanol and 2mg/ml blue bromophenol) and size-separated in 15% SDS-polyacrylamide gel electrophoresis. Proteins were transferred to PVDF membranes (Amersham Pharmacia, Spain) using a transblot apparatus (Bio-Rad, Spain). For immunoblotting, membranes were blocked with 5% non-fat dried milk in Tween-PBS (TPBS) for 1h. A primary antibody against pSTAT3 (Tyr705) (Cell Signaling Technology, USA; 1/100 final dilution) or STAT3 (Santa Cruz Biotechnology, USA; 1/1000 final dilution) was applied at the convenient dilution overnight at 4°C. After washing, appropriate secondary antibodies (anti-rabbit IgG-peroxidase conjugated) were applied for 1h at a dilution of 1/5000. Blots were washed, incubated in commercial enhanced chemiluminescence reagents (ECL, Amersham Bioscence, UK) and exposed to autoradiographic film. Films were scanned using a GS-800 Calibrated Densitometer (Bio-Rad, Spain) and blots were quantified using Quantity One software (Bio-Rad, Spain). Values for pSTAT3 were normalized with STAT3.

All variables analysed in this study had a normal distribution. Variations were analysed by a one-way ANOVA, followed by Newman–Keuls’ post hoc test. Statistical significance was set at p<0.05.

Obesity was induced in C57BL/6J mice by exposure to a diet in which 45% cal derived from fat. A control group was fed with LF chow, which only yielded 10% cal from fat. As appears illustrated in Fig. 1, animals on HF did not exhibit hyperphagia throughout the treatment.

Figure 1.

Influence of HF on food intake. Data appear expressed in cal/day. (¿) High-fat diet; (¿) low-fat diet.

(0.07MB).

As summarized in Table 2, mice on HF diet exhibited a significant increase of BW after 8 (1-ANOVA F(1,39)=67,653; p<0.001), 14 (F(1,49)=372,907; p<0.001) and 32 weeks (F(1,32)=131,326; p<0.001) on dietary treatment. The HF diet induced a gain of white adipose tissue (WAT), illustrated by the significant increase of lumbar and mesenteric adipose pads. The size of lumbar WAT, compared to the size in the corresponding matched LF groups, increased approximately 24%, 42% and 60% after 8 (1-ANOVA F(1,17)=15,572; p<0.001), 14 (F(1,25)=94,322; p<0.001) and 32 weeks (F(1,32)=13,204; p<0.001) on treatment, respectively. In the case of mesenteric WAT, the increase was 41% (F(1,17)=36,261; p<0.01), 89% (F(1,19)=8731; p<0.01) and 51% (F(1,18)=50,696; p<0.01) at identical time points.

We have used the weight of a tissue divided by BW (g) as an index to estimate the influence of dietary treatment on weight accumulation in a particular organ/tissue. Thus, if the index remains unchanged during treatment, one can assume that organ/tissue size progresses in a similar extent than BW does. Our current data show that relative weights of lumbar and mesenteric WAT are increased in HF animals, independently of treatment duration (Table 2), indicating that these tissues gain proportionally more weight than overall BW. Comparison of relative weights between LF and HF groups revealed that the HF diet induced a 66% increase of the relative weight of lumbar WAT after 8 weeks of treatment. Variations determined 14 and 32 weeks after treatment were 85% and 27%, respectively. In the case of mesenteric WAT, variations were 91%, 53% and 19% after 8, 14 or 32 weeks on treatment (Fig. 2). All these data show that the partial contribution of lumbar and mesenteric WAT to BW decreases during the development of DIO, suggesting the accumulation of fat in non-adipose organs.

Figure 2.

(a) Percentage of variation of absolute weight of different organs/tissues induced by 8, 14 and 32 weeks HF diet (compared to low-fat matched controls). Percentages were calculated as [(tissue weight HF/tissue weight LF)−1]×100. (b) Percentage variation of relative weight of different organs/tissues induced by 8, 14 and 32 weeks HF diet. Percentages were calculated as [(relative tissue weight HF/relative tissue weight LF)−1]×100. (¿) Lumbar WAT; (▴) mesenteric WAT; (¿) liver.

(0.11MB).

In the liver, we detected an increase of weight after 14 (F(1,25)=41,597; p<0.001) and 32 weeks (F(1,32)=55,536; p<0.001) on dietary treatment. In this case relative weights exhibited a progressive 23% and 46% increase after 14 (F(1,24)=6287; p<0.05) and 32 weeks (F(1,32)=30,538; p<0.001), respectively. Fig. 2a (% variation of absolute tissue weight) and b (% variation of relative tissue weight) illustrate the magnitude of body mass flux from WAT deposits toward the liver.

As summarized in Table 3, plasma analysis revealed that HF induced significant hyperleptinemia regardless of the duration of dietary treatment. Maximal levels were detected after 14-week HF, and the concentration of the hormone declined from this time point. After 32 weeks of treatment, animals also exhibited hyperleptinemia, but leptin concentration was reduced almost 50%, compared to the maximal level after 14-week HF. Plasma insulin concentration remained unchanged after 8 weeks of treatment but marked hyperinsulinemia was observed both after 14- (p<0.001) and 32-week (p<0.001) HF. Plasma concentration of glucose and free fatty acids (FAs) was also analyzed and we detected an increase of glucose levels after 14-week HF. Free FAs appeared to be decreased both after 14 (p<0.001) and 32 weeks (p<0.05) of dietary treatment. Adiponectin level was not modified by HF until 32 weeks of treatment (p<0.001).

As appears summarized in Table 4, hepatic lipid remained unmodified after 8-week dietary treatment but one-way ANOVA revealed an effect of diet both after 14- (F(1,15)=7873; p<0.05) and 32-week HF (F(1,13)=13,871; p<0.01). In mesenteric adipose tissue total lipid content remained unmodified in lean mice and a significant increase of lipids was detected only after 8-week HF.

To assess the functionality of leptin receptors in different tissues at different time points of dietary treatment, we analyzed the responsiveness to acute leptin both in lumbar WAT and hepatic tissues. Recombinant murine leptin (1mg/kg) was administered i.p. to LF and HF-fed mice and phosphorylated STAT3 (pSTAT3) measured 90min after leptin administration by means of Western blot. Figs. 3 and 4 illustrate the effect of leptin on STAT3 phosphorylation on the above-mentioned tissues at different time points of HF treatment (8, 14 and 32 weeks).

Figure 3.

Effect of acute leptin (1mg/kg; i.p.) on STAT phosphorylation (pSTAT3) in liver from mice treated during 8 weeks (a), 14 weeks (b) or 32 weeks (c) with high-fat diet. Upper panels: Immunodetection of STAT3 and pSTAT3 in left ventricle of mice receiving either saline or 1mg/kg leptin. Lower panels: Effect of leptin on STAT3 phosphorylation as the mean±S.E.M. of the ratio pSTAT3/STAT3 (n=5–6 animals) (*p<0.05 compared to their respective saline groups. Newman–Keuls’ test).

(0.16MB).

Figure 4.

Effect of acute leptin (1mg/kg; i.p.) on STAT phosphorylation (pSTAT3) in lumbar adipose tissue from mice treated during 8 weeks (a), 14 weeks (b) or 32 weeks (c) with high-fat diet. Upper panels: Immunodetection of STAT3 and pSTAT3 in left ventricle of mice receiving either saline or 1mg/kg leptin. Lower panels: Effect of leptin on STAT3 phosphorylation as the mean±S.E.M. of the ratio pSTAT3/STAT3 (n=5–6 animals) (*p<0.05 compared to their respective saline groups. Newman–Keuls’ test).

(0.16MB).