Materials

Thiobarbituric acid, CoA, N-ethylmaleimide, NAD+, NADH(H), NADPH(H), standards for oxidized and reduced glutathione, serotonin (5-HT), organic and inorganic components for buffer preparations, teraetoxypropane and butylated hydroxytoluene were purchased from Sigma Chemical Co (St. Louis, Mo, USA). Sodium nitrite (99.8% pure) was from Mallinckrodt (Griesheim, Germany). All solvents were HPLC-grade and provided by Carlo Erba (Milano, Italy). Lipid standards for c-gas liquid chromatography (c-GLC) identification were from Serdary Research Laboratories (London, Ontario, Canada). All standards were stored in appropriate solvents under nitrogen atmosphere at -20°C. Concentrations and purities were routinely checked by HPLC and/or c-GLC. Other chemicals used were reagent grade and obtained from local commercial sources.

Animals, plasma determinations and characteristics of the FRD-induced insulin resistance model

Normal male Wistar rats weighing 180-200 g were maintained in a temperature-controlled environment (23°C) with a fixed 12-hour light/dark cycle and fed a standard laboratory rat diet (Ganave, Argentina) for one week (stabilization period); the compositional analysis of this diet has been described elsewhere [33]. Rats were fed according to the American Institute of Nutrition [34]. After that period, rats were divided into two groups and fed the above mentioned diet and tap water (control diet, CD) during 3 weeks, or the same diet with the addition of 10% fructose into the drinking water (FRD). The fructose solution was freshly prepared and changed daily. In both groups, the relative FA content in the diet was 1.5% 16:0, 2.1% 18:0, 26.6% 18:1, 57.3% 18:2 n-6, 1.4% 18:3 n-3, 0.2% 20:3 n-6, 0.2% 20:4 n-9, 0.2% 20:1, and other FAs in trace amounts (< 0.2%) sufficient to cover the animals' nutritional requirements. Before sacrifice, blood samples were obtained from the retroorbital plexus under isoflurane (Florane, Abbot, Argentine) anesthesia; the samples were placed into chilled tubes with heparin, rapidly centrifuged at 4°C at 300 × g, and immediately analyzed or stored at -20°C. Plasma triglycerides, glucose and fructosamine were measured using commercial kits (Bio Systems Diagnostics, Argentina) implemented in an automated clinical analyzer. Insulin levels were determined by radioimmunoassay using rat insulin standard. Insulin resistance was assessed with the HOMA score, calculated with the formula: fasting plasma insulin (μU/ml) × fasting plasma glucose (mmol/l)/22.5, as described by Matthews et al. [35].

All the experimental procedures followed during this investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication no. 85-23, revised 1996) and to the guidelines established by the Animal Welfare Committee of La Plata School of Medicine.

PVAT measurements

After sacrifice, thoracic aortas were quickly removed and placed in Petri dishes with chilled PBS. The adhered PVAT was dissected under a stereoscopic zoom microscope (Model SMZ645/660, Nikon Instruments Inc., Melvilla, NY, USA). Aliquots of fat tissue weighing 60 to 80 mg were homogenized in a buffer solution containing 50 mM Tris-HCl (pH 7.4), 1.15% KCl, 1 mM N-ethylmaleimide and 0.1% BHT (v/v) with a glass-teflon microhomogeneizer. The homogenates were centrifuged at 4°C (600 × g) for 20 min and the resulting supernatants were assayed immediately or stored at -70°C under argon atmosphere. The extent of lipid peroxidation was determined by measuring thiobarbituric acid reactive species (TBARS) by the method of Yagi [36] and expressed as nmol TBARS/mg wet tissue. Teraetoxypropane (Sigma) was used as standard for calibration.

Glutathione and enzymatic antioxidant activity were also determined in PVAT. Total glutathione (GSSG + GSH) was measured following the recycling enzymatic assay described by Brigelius et al. [37]. Superoxide dismutase activities (SOD) were assayed as described by Flohé and Ötting [38]. Glutathione peroxidase (GSHPx) and glutathione reductase (GSHRd) were determined according to Habig et al. [39] and Carlberg and Mannervik [40], respectively.

Nitric oxide (NO) content in the homogenates was estimated as the sum of nitrates plus nitrites (NOx), which results as the spontaneous and stable end-point degradation of NO. NOx level was measured with the spectrophotometric method described by Miranda et al. [41] using sodium nitrite as standard.

Lipid composition analysis

Appropriate aliquots of tissue homogenates were processed by c-GLC fatty acyl composition analysis. Briefly, total lipids were extracted by the method of Folch et al. [42]. c-GLC of fatty acid methyl esters (FAME) was performed as previously reported [43], except that in this case we used a capillary column (Supelco, Avondale, PA, USA) mounted on a Hewlett Packard HP 6890 Series GC System Plus (Avondale, PA) equipped with a terminal computer integrator system. FAME were identified by comparison of their relative retention times with authentic standards, and mass distribution was calculated electronically by quantification of the peak areas.

Vascular reactivity studies

The aortas were cut into 3-4 mm wide rings and gently suspended between two stainless steel wires in a water-jacketed organ bath containing Krebs-Ringer bicarbonate (KRB) solution with the following composition (in mM): 130 NaCl, 4.7 KCl, 1.17 Na₂HPO₄, 1.16 MgSO₄, 24.0 NaHCO₃, 2.5 CaCl₂ and 6 glucose. The solution was continuously bubbled with a mixture of 5% CO₂ and 95% O₂, pH 7.4, and kept at 37°C. The lower wire was fixed to a vertical plastic rod immersed in the organ bath, while the upper one was rigidly attached to a force transducer (Letica TRI-201). Initially, arterial rings were stretched until attaining a force of ≈ 2 g; thereafter, the rings were allowed to stabilize for 60 min and washed with fresh KRB every 20 min. The signals from the force transducers were amplified and driven into an analog-digital board (DT16EZ, Data Translation, Inc., Marlboro, MA, USA) mounted in a desktop computer. On-line recordings and files for later processing were obtained with the appropriate software (Labtech Notebook Pro, Laboratory Technology Corp., Wilmington, MA, USA).

In some experiments, a high-K⁺ KRB solution with the following composition was used (in mM): 54.7 NaCl, 80 KCl, 1.17 Na₂HPO₄, 1.16 MgSO₄, 24.0 NaCO3 H, 2.5 CaCl₂ and 6.0 glucose.

To evaluate the potential modulatory role of adipose tissue, whereas some rings were dissected to remove the adherent PVAT, in others PVAT was left intact. In rings with an intact PVAT, a special protocol was developed to study the possible FRD-induced modifications on the relaxant effects of this tissue. For this purpose, a contraction was induced by 5 μM 5-HT, which acts through the activation of membrane receptors of vascular smooth muscle. Thereafter, the rings were carefully dismounted from the chambers and PVAT was carefully removed with micro scissors under a stereoscopic zoom microscope. Afterwards, the rings were remounted in the chambers as described before (including the 1-h stabilization), and the 5-HT challenge protocol was repeated. In order to minimize differences among vascular rings, the results of this second exposure to 5-HT were presented as paired to the first one, i.e., the change in the maximal force of the second contraction was expressed as a percentage of the first one. To test the effect of a different vessel contraction stimulatory pathway, we used a similar protocol (challenge of the intact ring followed by the same stimulus applied to the same ring after PVAT removal), but muscle contraction was triggered by high K⁺-induced depolarization.

In order to study whether vascular smooth muscle was affected by the FRD, appropriate controls were performed removing PVAT from CD and FRD rings after isolation of the aorta; the denuded rings were then subjected to the same two sequential steps of the protocol described above, including the demounting and remounting step between the two contractions. Although initially studies were performed using either intact or fat-deprived aorta rings, we later adopted the above mentioned procedure to obtain most consistent and least dispersed data.

Statistical analyses

Results are expressed as means ± SEM. One-way ANOVA followed by Bonferroni's test and Student's t-test were used for establishing statistically significant differences between groups. In all instances of statistical analysis, P < 0.05 was considered significant.

Food intake, body weight and blood parameters

A comparable caloric intake was recorded in both experimental groups. CD and FRD rats consumed an average of 19.2 and 13.3 g/animal/day; FRD animals also consumed 5.26 g fructose (calculated from the amount of fructose solution drinked daily). Accordingly, the food composition for CD vs. FRD rats was 45% carbohydrates, 43% proteins and 12% lipids vs. 60% carbohydrates, 31% proteins and 9% lipids. Consequently, there was a small but not significant difference in caloric intake between CD and FRD rats (55.65 ± 4.54 vs. 50.79 ± 3.43 kcal/day, respectively), No differences were detected in initial and final body weight between the experimental groups.

There were no significant differences in fasting plasma glucose between groups (FRD vs. CD, 6.28 ± 0.15 vs. 6.05 ± 0.15 mmol/L). FRD rats, however, had significantly higher plasma triglyceride, fructosamine and insulin levels (1.61 ± 0.14 vs. 1.13 ± 0.08 mmol/L, P < 0.02; 162.6 ± 3.3 vs. 146.2 ± 5.6 μmol/L, P < 0.03; 1.12 ± 0.04 vs. 0.86 ± 0.10, P < 0.05) as well as HOMA insulin resistance index (7.66 ± 0.31 vs. 5.16 ± 0.21, P < 0.001, respectively) than CD rats. These results validate the presence of an early fructose-induced metabolic disturbance and glucoxidative state in FRD rats, as described previously [12, 13, 15].

PVAT: FA composition, antioxidant defense system, TBARS and NOx

Vascular reactivity studies

Since it is known that PVAT attenuates the smooth muscle force development induced by 5-HT [24, 25], we tested its effect (5 μM concentration) upon two paired contractions (intact and PVAT-deprived rings) of aorta rings obtained from CD and FRD rats (see above). Removal of PVAT increased the developed force; as shown in Figure 1A, the second contraction (ring without PVAT) was bigger than the first one (intact ring). In the case of aortic rings from FRD animals, however, the increased force recorded after PVAT removal was significantly smaller than the one observed in rings obtained from CD animals (Figure 1A). This effect was no longer observed when the two contractions were performed in PVAT-deprived rings (Figure 1B). Additionally, the second contractions in these rings showed only a small and comparable increase in either CD or FRD rats (data not shown).

A similar experimental protocol was performed using high extracellular K⁺, that triggers smooth muscle cell contraction by depolarization and subsequent opening of voltage-operated Ca²⁺ channels. Results showed that PVAT does not affect the force development induced by depolarization; in fact, the percentage of increase of the second contraction (without PVAT) was small and of the same magnitude as that recorded in PVAT-deprived rings (Figure 1A). The lack of effect of PVAT on smooth muscle contractions induced by high extracellular K⁺ was also observed in aortic rings obtained from FRD animals (Figure 1A). Furthermore, no differences were seen between CD and FRD PVAT-deprived rings in the maximal force produced by depolarization with high K⁺ (Figure 1B).

Blood parameters measured in the FRD- rats model

As already reported, we have currently recorded a significant increase in plasma fructosamine level in rats fed a FRD for 3 weeks, indicating an increment in systemic protein glycosylation [15, 19]. Since no differences were recorded in fasting glucose levels but FRD rats had IGT, we must assume that these rats present abnormally high postprandial glucose levels. FRD rats also had higher circulating insulin levels and HOMA score values, thus supporting the presence of an IR state. Hypertriglyceridemia - another IR-inducer [3, 12, 15, 22] - was also present in these rats due to a combined effect of excess FA provision to the liver, decreased esterification in adipose tissue, and attenuation of the lowering effect of insulin on hepatic VLDL secretion. Such changes can be ascribed to a) the increased hepatic de novo lipogenesis observed during high carbohydrate feeding [44], b) the inhibition of lipoproteinlipase (LPL) activity reported in FRD rats [45], and c) the consequent decrease in VLDL and chylomicron hydrolysis.

PVAT measurements

Final remarks and conclusions

The current results evidence that an increased intake of fructose induces multiple metabolic and endocrine systemic alterations which are concomitant with significant changes in PVAT fatty acyl composition and oxidative stress biomarkers. These changes were similar to those measured in abdominal adipose tissue in this rat model. While at this stage (3 weeks of treatment) the vascular smooth muscle response was still unaffected, alterations in the vascular relaxant properties dependent on the PVAT were already evident. These results should contribute to elucidate, at least partly, the strong association between adipose tissue and vascular dysfunction. They also suggest that at early stages of this rat model that resemble the human MS, alterations in the paracrine function of PVAT precede those of vascular smooth muscle. Finally, the FRD rat model represents a valuable experimental tool to study the effect of agents potentially able to prevent or attenuate the vascular pathology associated with the MS and type 2 diabetes. Further, high fructose diets certainly promote multiple undesirable metabolic and vascular disorders.