Comparison of mechanisms involved in impaired vascular reactivity between high sucrose and high fat diets in rats

Reactive oxygen species (ROS) have been implicated in the pathogenesis of many diseases including diabetes [1], cardiovascular disease, and ischemia reperfusion injury [2‐5]. ROS can be produced through uncoupling of the electron transport chain or nitric oxide synthase as well as by NADPH oxidases [6‐8]. Hyperglycemia as well as elevated plasma free fatty acids and triglycerides also contribute to increased ROS production in addition to initiating insulin resistance through down-regulation of insulin signaling pathway intermediates [9‐11]

Endogenous antioxidant pathways protect from increases in ROS, but are often compromised in disease states. An important constitutive antioxidant pathway is the breakdown of superoxide anions (O₂˙ˉ) by superoxide dismutase (SOD) to form hydrogen peroxide (H₂O₂), which is further converted by catalase or glutathione peroxidase to form water and oxygen [12‐14].

Increases in the vascular levels of O₂˙ˉ lead to endothelial dysfunction through scavenging of the endothelial vasodilator nitric oxide (NO) to form peroxynitrite (ONOO-) thereby decreasing the bioavailability of NO. In addition, diminished levels of the essential eNOS cofactor, tetrahydrobiopterin, results in uncoupling of eNOS and enhanced O₂˙ˉ formation leading to further decreases in NO levels [6‐8, 11]

Elevated plasma glucose and triglyceride levels, which can lead to increases in oxidative stress [15], are thought to be involved in the pathogenesis of impaired vasodilation observed with high caloric intake and resultant increases in body mass. Increased adiposity is also associated with chronic inflammation [16, 17] and may thus contribute to impaired vasodilation observed in overweight animals. The roles of inflammation and oxidative stress on responsiveness to the endothelium-dependent vasodilator acetylcholine (ACh) will be examined in the present study. The majority of research on the effects of high fat feeding on vascular function has been carried out using large conduit arteries such as the aorta from animal models of established obesity, including rats with high fat diet-induced obesity. Although these studies are informative in documenting the end effects of obesity, much less is known of the nature of vascular derangements in more modest settings of high caloric intake prior to overt obesity. Moreover, very little research has been conducted examining the effects of high sucrose feeding, in the absence of high fat intake, on vascular reactivity. In one recent study, inactive people who simultaneously ingested a candy bar and soft drink developed impaired endothelial function, as evidenced by lower flow-mediated vasodilation, compared to those who ingested the same meal but underwent a single bout of endurance exercise the day prior to blood flow measurements [18]. In studies of combined high fat and high sucrose feeding for 6 months, rats develop impaired aortic endothelial dependent vasorelaxation that is reversible following a low-fat complex carbohydrate diet [19]. Other studies have shown that feeding rats a combined high sucrose and moderately high fat diet for only 3 days similarly results in significantly impaired arterial vasorelaxation in the absence of weight gain [20]. Potential interactions between high sucrose and high fat in these prior studies remain unclear. Therefore, the present study was designed to examine the early effects of high fat or high sucrose feeding on vasodilation in resistance arterioles isolated from mildly overweight animals. Moreover, we examined the potential differential effects of high sucrose vs. high fat diets to elicit changes in endothelial function. We hypothesized that feeding rats diets containing high sucrose or high fat for 6 weeks would produce oxidative stress and inflammation leading to endothelial dysfunction and impaired vasodilation in small mesenteric resistance arterioles. Therefore, the goals of the present study were to 1) demonstrate that rats fed a high sucrose or high fat diet gain more adiposity compared to Chow fed controls, 2) show that ROS levels were increased in the plasma as well as in small mesenteric resistance arterioles of high sucrose as well as high fat fed rats, 3) determine if eNOS protein expression is reduced in rats fed either diet thereby contributing to impaired vasodilation, and 4) test whether ROS or inflammatory pathways are involved in the impaired endothelium-dependent vasodilation that occurs after 6 weeks of high sucrose or high fat intake.

All protocols and surgical procedures employed in this study were reviewed and approved by the Institutional Animal Care and Use Committees of the University of New Mexico School of Medicine and Arizona State University.

The major findings of this study are that 6 weeks of high sucrose or high fat feeding results in increased body mass and adiposity (Table 2) concomitant with increased oxidative stress and impaired vasodilation, although the mechanism of impaired vasodilation differed for each group. Vasodilation of arteries from Chow fed rats is dependent on NO as well as H₂O₂ (Figures 3 and 6). Feeding rats either HSD or HFD results in diminished vascular smooth muscle sensitivity to NO (Figure 7). Vasodilation is further impaired in rats fed HSD as a result of increases in H₂O₂ which acts to oppose vasodilation (Figures 1, 4 & 6). In contrast, the findings of this study implicate O₂˙ˉ-mediated scavenging of NO to form ONOO- in arteries from rats fed HFD (Figures 1 & 5).

Feeding rats a HSD diet alone elicited hypertriglyceridemia (Table 2), impaired glucose tolerance (Figure 1A), hyperleptinemia (Table 2), enhanced oxidative stress (Table 2 & Figure 1B) and impaired acetylcholine-mediated vasodilation (Figure 2), which are characteristics of metabolic syndrome and pre-diabetes. Animals on the HFD gained more adiposity compared to the HSD rats (Table 2) and demonstrated further characteristics of metabolic syndrome as they developed fasting hyperglycemia (Figure 1A) in addition to hyperleptinemia (Table 2), oxidative stress (Table 2 & Figure 1B), and impaired acetylcholine-induced vasodilation (Figure 2). These significant alterations in oxidative stress and vascular reactivity demonstrate that vascular dysfunction occurs well before the development of obesity and that different diets can have varying effects on biochemical parameters and vascular reactivity in rats.

Elevated plasma glucose, free fatty acids and triglycerides contribute to increased ROS production [9‐11]. Previous studies have shown that feeding obesity-prone Sprague-Dawley rats a moderately high fat diet (32% kcal% as fat) for 16 weeks results in elevated TBARS and O₂˙ˉ in aorta [27]. Similarly, our studies demonstrate increased fasting glucose in the HFD rats (Figure 1A) and plasma triglycerides in the HSD rats (Table 2) with both groups of rats developing elevated TBARS and DCF fluorescence, indicative of vascular ROS (Table 2 & Figure 1B).

Multiple oxidative stress pathways can potentially lead to decreases in NO bioavailability and therefore, a reduction in vasodilatory responses. A direct pathway occurs by O₂˙ˉ-induced scavenging of endothelial-derived NO to form peroxynitrite (ONOO-) [6, 7]. Another mechanism involves the uncoupling of eNOS as a result of diminished levels of the essential eNOS cofactor, tetrahydrobiopterin, resulting in enhanced O₂˙ˉ formation and further decreases in NO levels [6‐8, 11]. Furthermore, ONOO- itself can oxidize and deactivate tetrahydrobiopterin [28]. Moreover, H₂O₂, produced from the breakdown of O₂˙ˉ, is reported to exhibit vasodilatory or vasoconstrictor effects depending on the concentration or tissue examined [29‐31]. Since the observed increase in oxidative stress in HSD and HFD rats may directly affect blood vessel reactivity or impact NO bioavailability, the effects of oxidative stress on endothelium-mediated vasodilation were measured (Figures 3, 4, 5, 6).

Data from the present study demonstrate that high sucrose and high fat feeding result in significantly attenuated endothelium-dependent vasodilation compared to Chow-fed controls (Figure 2). In Chow-fed rats, ACh-mediated vasodilation appears to rely on NO since the NOS inhibitor LNNA nearly abolished the vasodilatory response (Figure 3). In contrast, inhibition of NOS caused no further impairment of vasodilation in HSD and HFD fed rats (Figures 4 & 5). These data suggest that ROS may be involved in the scavenging of NO in these animals as eNOS protein expression levels were not significantly different in the experimental groups compared to Chow fed rats (Figure 8). Further studies demonstrated that arteries from HSD and HFD rats have impaired vascular smooth muscle sensitivity to NO contributing to the impaired ACh-mediated vasodilation (Figure 7). Therefore, it is evident that the residual response to ACh following high calorie feeding is NO-independent and likely involves other endothelium-dependent vasodilatory pathways.

Since the observed increase in oxidative stress in HSD and HFD rats may impact NO bioavailability, the effects of oxidative stress on endothelium-mediated vasodilation were measured (Figures 3, 4, 5, 6). ACh-mediated vasodilation responses in arteries from HFD rats were greatly attenuated across a broad range of ACh doses and were normalized by the inhibition of ROS using either tiron and catalase or EUK-134 (Figure 5). In contrast, vasodilation of arteries from HSD rats was impaired at only high doses of ACh which likewise demonstrated improved vasodilation in the presence of tiron and catalase but not EUK-134 (Figure 4). These data illustrate that oxidative stress plays a role in the limitation of ACh-mediated vasodilation in HSD and HFD rats with only modest increases in adiposity. Oxidative stress is similarly involved in the impaired vasodilatory responses of aortic rings and renal arterioles from alloxan-induced diabetic rabbits [32]. Impaired endothelium-dependent vasodilation, as occurs in coronary artery diseases, has likewise been linked with increased oxidative stress since administration of the antioxidant vitamin C improved the response in humans [33]. Similarly, Zucker obese rats exhibit attenuated responses to ACh in isolated coronary as well as middle cerebral arteries [34, 35]. Moreover, diminished ACh-mediated vasorelaxation in thoracic aorta from insulin resistant rats has been observed [36].

To test whether impaired vasodilation in arteries from HSD and HFD is due to an interaction of ROS and NO, vessels from each group were incubated in the presence of both the NOS inhibitor LNNA and the ROS scavengers tiron and catalase or EUK-134. Data from these studies demonstrate that impaired vasodilatory responses recorded in arteries from HSD rats is at least in part mediated by reduced NO bioavailability as the combined treatment nearly abolished vasodilation (Figure 4). Similar results were observed in arteries from HFD rats treated with LNNA and tiron and catalase, suggesting that ROS may also reduce the bioavailability of NO in these vessels (Figure 5). However, superfusion of arteries with LNNA and EUK-134 did not normalize vasodilation in arteries from these rats (Figure 5). This apparent discrepancy may be due to the different mechanisms of action of each of these ROS scavengers.

Since H₂O₂ is known to exert both vasodilatory and vasoconstrictor properties, we examined the role of H₂O₂ by exposing arteries to catalase in the absence of superoxide dismutase mimetics. In arteries from HFD rats, this exposure did not improve vasodilation (Figure 6C) whereas the combined superoxide dismutase mimetics and catalase were successful at normalizing vasodilation (Figure 5). This suggests that in HFD vessels, elevated O₂˙ˉ may be responsible for scavenging of NO resulting in the production of ONOOˉ. In contrast, arteries from HSD fed rats demonstrate only a mild improvement in ACh-mediated vasodilation in the presence of tiron and catalase but no effect of EUK-134 (Figure 4). Since inhibition of H₂O₂ alone normalized vasodilation in this group (Figure 6B), this supports a role for H₂O₂ as a vasoconstrictor following HSD feeding. DCF can be oxidized by both ONOOˉ and H₂O₂ resulting in increased fluorescence as we observed in arteries from HSD and HFD rats (Figure 1B). Our data also demonstrate a physiological role of H₂O₂ in vasodilatory responses of arteries from Chow rats (Figure 6A), that has been described by others [29‐31].

We also observed a differential role of COX products between the different feeding regimens. In Chow rats, COX inhibition blunted ACh-mediated vasodilation (Figure 6A). In contrast, indomethacin normalized vasodilation in arteries from HSD rats suggesting a switch to vasoconstrictor COX products following high sucrose feeding (Figure 6B) that was only minimally present in the HFD group (Figure 6C). Thus, the profile of COX-derived vasoactive products may be altered by diet.

In summary, our data demonstrate that feeding rats either a high fat or high sucrose diet results in the development of oxidative stress as well as impaired vasodilation. The data highlight the importance of the type of diet as it can produce divergent effects on vascular reactivity pathways despite both groups developing increased body mass, adiposity and oxidative stress. Although rats in the HSD fed group develop similar levels of oxidative stress as observed in the HFD rats, the impaired vasodilation is not as severe and the mechanisms of impaired vasodilation are divergent. In the HFD group, the impaired vasodilation appears to be mediated in part by O₂˙ˉ scavenging of NO. In contrast, H₂O₂ is implicated in the impaired vasodilatory responses in vessels from HSD rats. In conclusion, the impaired vasodilatory responses to acetylcholine in rats fed either HSD or HFD are mediated by ROS scavenging of NO, impaired smooth muscle sensitivity to NO as well as by inflammatory factors.