Reactive oxygen species (ROS) and endothelin-1 (ET-1) contribute to vascular pathophysiology in obesity. In this context, whether ET-1 modulates hydroxyl radical (•OH) formation and the function of ROS/•OH in obesity is not known. In the present study, formation and function of ROS, including •OH, were investigated in the aorta of lean and leptin-deficient obese ob/ob mice. Hydroxyl radical formation was detected ex vivo using terephthalic acid in intact aortic rings and the involvement of ROS in ET-1-mediated vasoreactivity was analyzed using the antioxidant EPC-K1, a combination of α-tocopherol and ascorbic acid. Generation of either •OH, •O2−, and H2O2 was strongly inhibited by EPC-K1 (all P < 0.05). In obese mice, basal vascular •OH formation and ROS activity were reduced by 3-fold and 5-fold, respectively (P < 0.05 vs. lean). ET-1 markedly enhanced •OH formation in lean (6-fold, P < 0.05 vs. untreated) but not in obese mice. Obesity increased ET-1-induced contractions (P < 0.05 vs. lean), and ROS scavenging further enhanced the response (P < 0.05 vs. untreated). Exogenous ROS, including •OH caused stronger vasodilation in obese animals (P < 0.05 vs. lean), whereas endothelium-dependent relaxation was similar between lean and obese animals. In conclusion, we present a sensitive method allowing ex vivo measurement of vascular •OH generation and provide evidence that ET-1 regulates vascular •OH formation. The data indicate that in obesity, vascular formation of ROS, including •OH is lower, whereas the sensitivity to ROS is increased, suggesting a novel and important role of ROS, including •OH in the regulation of vascular tone in disease status associated with increased body weight.
- oxidative stress
- terephthalic acid
- vitamin E
- vitamin C
endothelin-1 ( et-1) is a potent vasoconstrictor (57) and mitogen (30) that has also been implicated in the pathogenesis of metabolic diseases, including obesity and diabetes (4, 29, 31, 43). Endogenous ET-1 contributes to increased basal vascular tone in human obesity (8, 31), and we and others have shown that vascular responsiveness to ET-1 is enhanced in obesity (8, 50). Recent evidence also suggests that ET-1 stimulates the generation of certain reactive oxygen species (ROS) (26, 52).
ROS are short-lived, oxygen intermediates that participate in activation of intracellular signaling pathways (48) and regulate cell function (14), and they are known to contribute to abnormal vascular reactivity in obesity-related diseases such as diabetes, atherosclerosis, and hypertension (16, 41, 44, 46, 48, 55). Although vascular effects of ROS have been often regarded as deleterious, resulting in impaired vasodilation and/or enhanced vasoconstriction (45), ROS also have beneficial effects; indeed, superoxide (•O2−), hydrogen peroxide (H2O2), and the hydroxyl radical (•OH) may also act as vasodilators under certain conditions (40, 53).
Previous studies have investigated vascular interactions between ET-1 and •O2− or H2O2 (26, 52); however, little is known about whether ET-1 also interacts with •OH, a highly reactive ROS (33). Hydroxyl radical is mainly formed by H2O2 via either the Fenton (13) or the Haber-Weiss reaction (23):
At high concentrations, hydroxyl radicals can cause membrane damage, DNA and RNA fragmentation, and lipid peroxidation (9, 49). In the vasculature, •OH regulates vasoconstriction (49) and vasodilation (40) under normal and pathological conditions (39, 40).
The aim of the present study was to characterize the interactions between ET-1 and ROS/•OH in the vasculature of normal C57BL6/J laboratory mice and to determine the possible modulatory effects of obesity on ROS-mediated vascular tone and •OH production using a model of leptin-deficient obesity.
Healthy 40-wk-old male C57BL/6 mice (controls) and severely obese leptin-deficient mice (B6.V-Lepob/J; Charles River, Sulzfeld Germany; obese) were used. Body weight of obese mice was 58 ± 4 g compared with 29 ± 1 g for control mice. Animals were held at the institutional animal facilities and received standard rodent chow and tap water ad libitum. Housing facilities and experimental protocols were approved by the local authorities for animal research (Kommission für Tierversuche des Kantons Zürich, Switzerland) and conform to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85-23, revised 1996). Mice were anesthetized (xylazine: 100 mg/kg body wt; ketamine: 23 mg/kg body wt; and acepromazine: 3.0 mg/kg body wt ip), and exsanguinated via cardiac puncture.
Measurement of Vascular Hydroxyl Radical Formation
To quantify •OH generation in very small specimens of living vascular tissue ex vivo, we established a method using terephthalic acid (TPA), a specific probe for •OH detection, slightly modified from published in vitro protocols (3, 42). Hydroxylation of TPA yields a stable and highly specific isomer (2-hydroxyl terephthalic acid, TPA-OH) (42) that is not formed by reaction with either superoxide or hydroperoxides (3). The sensitivity and specificity of TPA to detect •OH are comparable to that of the spin trap 5,5 dimethyl-1-pyrroline N-oxide (27), as detected by electron spin resonance technique (see also supplemental information, which may be found online at the American Journal of Physiology—Regulatory, Integrative and Comparative Physiology website). Vascular •OH generation was measured in aortic rings, which were left unconstricted and placed in gassed Krebs solution (95% O2, 5% CO2, 37°C, pH 7.4) for 4 h containing 2.5 mM TPA in the absence (basal •OH formation) or in the presence of 10−6 mol/l ET-1 (ET-1-stimulated •OH formation). TPA-OH was quantified fluorometrically (excitation: 326 nm, emission: 432 nm) using a microplate reader (SpectraMax M2) (42). Hydroxyl radical formation was calculated as relative light units per milligram of protein after subtraction of background fluorescence. Protein was extracted from vascular tissue in a RIPA buffer containing orthovanadate, PMSF, and complete mini protease inhibitor (Roche, Rotkreuz, Switzerland) using a cell shredder (Qiagen, Hombrechtikon, Switzerland) and incubated for 30 min at 4°C. Insoluble material was separated by centrifugation and protein content of the soluble fraction was quantified using Bio-Rad Protein Assay solution (Bio-Rad, Reinach, Switzerland), according to the manufacturer's instructions.
Characterization of the Antioxidant Activities of EPC-K1
The effects of EPC-K1 on •O2− and •OH formation and on the stability of H2O2 were determined using specific assays (12). The radical scavenger EPC-K1, a water-soluble compound, is a combination of two known antioxidants, vitamin E (alpha-tocopherol) and vitamin C (ascorbic acid) (54) linked by a phosphate diester (See also supplemental information that is available online at the American Journal Physiology—Regulatory, Integrative and Comparative Physiology website.). Effects of EPC-K1 (10−1 mg/ml) on xanthine/xanthine oxidase-mediated •O2− generation were determined by L-012-dependent chemiluminescence, as described (10). The overall measuring time was 6 min, and a total of 24 measurements were taken during this period. Each measurement was for 10 s, and the interval between each reading was 5 s (Lumat LB 9507, Berthold Technologies, Regensdorf, Switzerland). Total superoxide anion formation was determined by cumulative measurements of chemiluminescence over 6 min and calculated after subtracting the background. Hydroxyl radical was generated from the reaction of Fe2+ and ascorbic acid (51) (both at 100 μmol/l) in warm, gassed Krebs solution (37°C, 95% O2, 5% CO2), in the presence or absence of EPC-K1 (10−1 mg/ml) (35), and quantified using TPA (2.5 mM) hydroxylation (42). The effect of EPC-K1 (10−1 mg/ml) on hydrogen peroxide (0.3 μmol/l H2O2) was assessed using the Amplex Red assay (Molecular Probes, Eugene, OR), as previously described (18). The reaction product resorufin (18) was detected fluorometrically (excitation and emission wavelengths: 571 nm and 585 nm, respectively) using a SpectraMax M2 (see also supplemental information online at the American Journal of Physiology—Regulatory, Integrative and Comparative Physiology website).
Vascular Function Studies
The aorta was excised and placed in cold (4°C) Krebs-Ringer bicarbonate solution [in mmol/l: NaCl 118.6; KCl 4.7; CaCl2 2.5; MgSO4 1.2; KH2PO4 1.2; NaHCO3 25.1; ethylenediaminetetraacetic acid calcium disodium salt (EDTANa2Ca) 0.026; glucose 10.1], dissected free of connective tissue under a microscope (Olympus SZX9, Volketswil, Switzerland), and cut into rings 3 mm in length. Special care was taken not to damage the endothelium during this procedure. Segments of thoracic aorta were mounted onto tungsten wires (100 μm in diameter) and transferred to water-jacketed organ baths containing Krebs solution (95% O2,-5% CO2 at 37°C, pH 7.4) and connected to force transducers (Hugo Sachs Elektronik, March-Hugstetten, Germany) (50). Resting tension was gradually increased to the optimal level as previously described (50); vessels were repeatedly exposed to 100 mM KCl until a stable contraction response was achieved.
Endothelium-dependent and -independent vascular function.
Endothelium-dependent relaxation to ACh (0.1–10 μmol/l) was determined in the presence of indomethacin (10 μmol/l, a nonselective cyclooxygenase inhibitor, 30 min preincubation) in rings preconstricted with phenylephrine to 80% of KCl-induced contraction, as previously described (5). Endothelium-independent vasodilation was investigated using the nitric oxide (NO) donor sodium nitroprusside (10 μmol/l).
Reactive oxygen species and basal vascular tone.
Rings were pretreated with l-nitro-arginine methyl ester (l-NAME, 300 μmol/l) for 15 min to block effects of endogenous NO (24). To determine whether endogenous ROS formation contributes to basal vascular tone, the ROS scavenger EPC-K1 (10−1 mg/ml equal to 141.46 μM) (35) was applied to quiescent vascular rings and the resultant contraction, reflecting basal vascular ROS generation (6, 20) was measured. The contractile responses of quiescent vessels to scavengers or enzyme inhibitors allow us to indirectly quantify basal generation of molecules such as NO or •O2− in a particular vascular bed (6, 20).
Vascular responses to ET-1.
Vascular rings were exposed to ET-1 (0.01–300 nmol/l) in the presence of l-NAME (300 μmol/l) for 30 min, as described (56). To investigate the role of ROS for ET-1-induced contraction, selected rings were also preincubated with the ROS scavenger EPC-K1 (10−1 mg/ml) for 15 min.
Gene Expression Measurements
Steady-state mRNA expression levels of vascular nitric oxide synthase (NOS) isoforms 2 and 3 were determined by real-time quantitative reverse transcriptase-PCR. Aortic total RNA was extracted using the RNeasy micro kit for RNA isolation from fibrous tissues (Qiagen, Hilden, Germany). Purity of RNA was controlled by RT(−) reactions (PCR with nontranscribed RNA). RNA was reverse-transcribed with the Omniscript RT kit (Qiagen, Hilden, Germany). Real-time quantitative PCR was used to determine relative expression levels of murine genes encoding for NOS2 (Nos2, locus: NM_010927), NOS3 (Nos3, locus: NM_008713) and the housekeeping gene tyrosine-3-monooxygenase (locus: NM_011740). For amplification of a tyrosine-3-monooxygenase-specific cDNA fragment 5′-CGA GCA GGC AGA GCG ATA TG-3′ was used as the forward and 5′-AGA CGA CCC TCC ACG ATG AC-3′ as the reverse primer, for a Nos2-specific fragment 5′-GCA CCG AGA TTG GAG TTC-3′ as the forward and 5′-AGC ACA GCC ACA TTG ATC-3′ as the reverse primer. 5′-CCT AGT CCT CGC CTC CTTC-3′ was used as the forward primer for a Nos3-specific fragment and 5′-ACC ACT TCC ATT CTT CGT AGC-3′ as the reverse primer. Two-step PCR was performed with iQ SYBR Supermix PCR kit (Bio-Rad) as follows: activation of the hot start Taq polymerase for 3 min (95°C), followed by 40 cycles of denaturation at 95°C for 15 s (step 1), and annealing and extension at 60°C for 1 min (step 2). Fluorescence was detected at the end of each extension step. Identity and specificity of amplicons were confirmed by agarose gel electrophoresis, melting curve analysis, and sequencing (Microsynth, Balgach, Switzerland). Steady-state mRNA expression levels were calculated by normalizing the relative amount of each mRNA to the housekeeping gene tyrosine-3-monooxygenase (28) and expressed as arbitrary units (AU).
ET-1 and l-NAME were from Alexis Corp (Lausanne, Switzerland). EPC-K1 was from Senju Pharmaceuticals (Osaka, Japan). L-012 was from Wako Chemicals (Neuss, Germany), and Amplex Red was from Molecular Probes (Eugene, OR). All other chemicals were from Sigma Chemicals (Buchs, Switzerland).
Data are given as means ± SE, and n denotes the number of animals used per experiment. Contraction is expressed as a percentage of 100 mM KCl-induced contraction, and dilations are given as a percentage of the maximal contraction. EC50 values (as negative logarithm pD2) were calculated by nonlinear regression analysis, and area under the curve was calculated for each individual curve using SigmaPlot (SPSS Chicago, IL). Data were analyzed using ANOVA for repeated measurements with Bonferroni correction, the Mann-Whitney U-test, or unpaired Student's t-test when appropriate. A P value <0.05 was considered significant.
Efficacy of •OH-Generating Systems in Krebs Solution
Hydroxyl radical generation was determined in vitro by TPA, a specific probe for •OH (27). The efficacy of •OH generated by ascorbic acid and Fe2+ was compared with another •OH-generating system consisting of H2O2 and Fe2+. The amount of •OH generated by vitamin C and Fe2+ was 2.6-fold higher compared with H2O2 and Fe2+ (P < 0.05 vs. H2O2 and Fe2+, Table 1). DMSO, a known •OH scavenger (1), blocked in either system •OH formation by ∼80%, as shown in Table 1 (See also supplemental data for this article, which is available online at the American Journal of Physiology—Regulatory, Integrative and Comparative Physiology website.).
Vascular Hydroxyl Radical Formation and Acute Effects of ET-1
TPA was used to quantify vascular •OH formation ex vivo. Formation of •OH was detected under basal conditions in uncontracted aortic rings from control mice [2.5 ± 0.1 relative fluorescence units (RFU)/mg protein, Fig. 1A] and markedly increased by endothelin-1 (15.0 ± 1.0 RFU/mg protein, sixfold, P < 0.0001, vs. untreated, Fig. 1B). Basal •OH formation was reduced approximately threefold in obese animals (P < 0.0001 vs. lean, Fig. 1, A and B) and ET-1 slightly increased this attenuated •OH formation (0.6 ± 0.1 to 2.3 ± 0.3 RFU/mg protein, P < 0.0001 vs. untreated, Fig. 1B). Interestingly, ET-1-stimulated •OH formation in obese mice was still below basal levels of •OH detected in lean mice (2.3 ± 0.3 vs. 2.5 ± 0.1 RFU/mg protein, respectively).
Effects of EPC-K1 on ROS Formation and Vascular Tone
The effects of EPC-K1 on in vitro generation of •O2−, •OH, and H2O2 were determined. EPC-K1 inhibited xanthine-oxidase-mediated •O2− generation by 74 ± 2% (P < 0.0001 vs. control), that of ascorbic acid and Fe2+-mediated •OH formation by 84 ± 1% (P < 0.0001 vs. control), and that of H2O2 by 61 ± 2% (P < 0.0001 vs. control). We next investigated basal ROS bioactivity in quiescent aortic rings by measuring changes in vascular tone after applying EPC-K1, which caused contraction in all rings investigated (P < 0.05, vs. untreated, Fig. 2). EPC-K1-induced contractions were markedly reduced in obese mice (5-fold, P < 0.05, vs. control).
Effects of Exogenous ROS on Vascular Tone
The addition of exogenous ROS/•OH to precontracted vascular rings caused a vasodilator response (P < 0.0001, vs. untreated, Fig. 3, ROS/•OH in Fig. 3A), that was enhanced threefold by obesity (P < 0.05 vs. control).
Function and Expression of Vascular NO Synthases
NO-mediated endothelium-dependent relaxation to ACh was similar in lean and obese mice [−94 ± 1.6 vs. −94 ± 3.3%, not significant (ns), Fig. 3B], as was relaxation to sodium nitroprusside (SNP) (−109 ± 1.3 vs. −107 ± 1.5%). Similarly, vascular expression of NOS3 mRNA was similar between control (504 ± 53 AU) and obese mice (471 ± 74 AU, ns), as were NOS2 mRNA levels (control, 1.2 ± 1 AU, n = 4; obese, 0.95 ± 0.35 AU, n = 7, ns).
Effect of Endogenous ROS on ET-1-Induced Contractility
Contractions to ET-1 were enhanced by obesity (20.4 ± 4.9% vs. 10.6 ± 1.5%, P < 0.05, Fig. 4, A and B and Table 2). In obese mice, scavenging of ROS by EPC-K1 further enhanced ET-1-induced vasoconstriction (45 ± 9%, P < 0.05 vs. untreated); this effect was absent in lean controls (Fig. 4).
This study demonstrates interactions between the vasoconstrictor ET-1 and ROS/•OH under normal conditions and in obesity. We present a sensitive method to quantify hydroxyl radical formation in small vascular specimens ex vivo, using TPA hydroxylation, and show that ET-1 acutely stimulates vascular •OH formation in the aorta of normal laboratory mice. The results provide evidence that basal ROS activity, as well as ET-1-stimulated •OH formation were markedly reduced in obesity. Scavenging of endogenous ROS with the nonselective radical scavenger EPC-K1 increased vasoconstrictor responses to ET-1 in aortic rings of obese animals, but not in controls, indicating that ROS antagonize ET-1-mediated vasoconstrictor responses in obesity.
•OH is a highly reactive ROS, and its role in pathophysiology has only been clarified in part (17, 25, 48). Investigations on the role of •OH in the pathophysiology of cardiovascular disorders have, to date, been hindered by the lack of sensitive and specific methods. Previously, salicylic acid has been used to detect •OH formation (11). However, because salicylic acid also detects peroxynitrite (19) and inhibits enzymes involved in ROS production (7, 21), salicylic acid cannot be considered a specific probe for •OH. Therefore, one of the goals of the present study was to establish a specific and sensitive method that would allow us to measure •OH generation in small vascular specimens ex vivo. We used TPA, which is a highly specific and sensitive probe for •OH (3), as shown by electron spin resonance experiments (27) (see also supplemental information online at the American Journal of Physiology—Regulatory, Integrative and Comparative Physiology website), and therefore, no other probes for •OH were used in our experiments. Using the TPA-based •OH detection method, the present study shows that basal •OH formation can indeed be detected in very small specimens of arterial tissue in both lean and obese animals.
Endothelin contributes to the pathogenesis of atherosclerosis, hypertension, and diabetes (30, 43) and has been previously shown to stimulate the generation of ROS such as •O2− and H2O2 (26, 52). Presently, there is no information on whether ET-1 modulates vascular formation of •OH. Using TPA as a probe for •OH, we found that ET-1 is a potent stimulus of vascular •OH formation in animals with normal body weight. An unexpected finding of the present study was that this stimulating effect of ET-1 was more or less abrogated in obese mice. Because of the limitations in tissue and animals, only a single concentration of •OH was used. To our knowledge, this is not only the first demonstration that ET-1 modulates vascular •OH formation, but also that both basal and stimulated •OH formation are markedly reduced in obesity. In line with these results, basal vascular activity of ROS, as measured indirectly by EPC-K1-induced contractions, was also reduced in obese animals. Our study did not identify the nature of ROS mediating the observed effects in response to EPC-K1. One of the potential candidates of ROS could be •O2− as in a most recent study using young obese db/db mice, which express a mutated leptin receptor and unlike the ob/ob animals used in our study show high circulating leptin levels, Griendling and colleagues (41) demonstrated that aortic membrane superoxide generation was enhanced compared with lean controls.
EPC-K1 has been previously proposed to be a scavenger that is selective for •OH (35). On the basis of its chemical structure that combines ascorbic acid and α-tocopherol linked by a phosphodiester, it is very unlikely to be specific for •OH scavenging (54). Indeed, our results now show that EPC-K1 not only scavenges •OH but also that it potently inactivates •O2− and H2O2. Therefore, in the experiments involving EPC-K1, scavenging of any of the ROS mentioned may contribute to the observed changes in vasoreactivity. According to a previous report, EPC-K1 as a ROS scavenger induces NO-independent relaxation in precontracted canine coronary arteries (47). In the present study, however, we demonstrate that EPC-K1 also causes contraction of quiescent murine aortic rings, reflecting basal ROS vasodilator bioactivity. These differences in the effects of EPC-K1 are likely to be explained by 1) the state of contraction of the vascular rings, 2) by the type of artery studied and/or 3) by interspecies differences.
To investigate the effects of exogenous ROS/•OH on vascular tone, we used ascorbic acid and iron (45, 49) as a ROS/•OH generating system. In vitro experiments confirmed that this system is indeed a strong inductor of •OH formation and that concentration-dependent relaxation can be achieved (see supplemental information for details). The results of the present study demonstrate that exogenous ROS/•OH cause vasodilation, an effect that is markedly enhanced by obesity. This suggests that obesity modulates the sensitivity of the vasculature to vasodilator activities of ROS. Also, it is likely that obesity per se and not leptin deficiency enhances the responsiveness to ROS/•OH, since enhanced vascular relaxation to ROS/•OH was also observed in a model of diet-induced obesity (36). An unexpected finding was that no differences in the NO-mediated endothelium-dependent relaxation in the presence of cyclooxygenase inhibitors were seen between obese and lean animals. The steady-state mRNA expression levels of endothelial nitric oxide synthase (eNOS/NOS3) was also similar between the groups. Of note, in these experiments, a cyclooxygenase inhibitor was used since obesity enhances cyclooxygenase-mediated production of vasoconstrictor prostanoids (36, 50), which attenuate endothelium-dependent vascular relaxation in response to ACh in obesity (50). In a recent study, endothelium-dependent relaxation was found to be impaired in the aorta of obese db/db mice (41), which is likely to be explained by the lack of a cyclooxygenase inhibitor in the organ chamber experiments (50). The results of the present study are in line with previous studies showing that in the presence of cyclooxygenase inhibition, NO-dependent relaxation actually remains unaffected in diet-induced, as well as in monogenetic forms of obesity in rodents (2, 22, 36).
Similar to our recent study using a model of diet-induced obesity (50), we found that contractions to ET-1 are also increased in obesity due to leptin deficiency. The role of ROS for ET-1-induced contractions was investigated using the ROS scavenger EPC-K1. While in lean animals, EPC-K1 had no effect on contractions to ET-1, the antioxidant markedly enhanced contractions in obese animals, suggesting that endogenous ROS antagonize ET-1-mediated vasoconstriction in obesity, but not under normal conditions. The present data also suggest that this indirect vasodilator activity of ROS in obesity is independent of NO, since all experiments were performed in the presence of a NO synthase inhibitor. Indeed, reactive oxygen species have been shown to block vasoconstriction and mediate vasorelaxant effects in the vasculature by 1) activation of endothelium-dependent hyperpolarizing factor (32, 37), 2) by activating cGMP pathway (15, 34), and/or 3) by inhibition of phosphorylation of myosin light chain protein (38). In the present study, because NOS was blocked, cyclic GMP is unlikely to play a role in the ROS-mediated responses observed. We would like to again emphasize that the findings of ET-1-induced •OH generation and the observed changes in vasoreactivity are two independent observations (tissue specimens vs. arterial rings on passive tension), which were made under different conditions. Therefore, ET-1-derived •OH generation is unlikely to be the sole mechanistic link for the observed changes in vasoreactivity.
The present study demonstrates novel vascular effects and interactions between ET-1 and ROS/•OH under normal conditions and in states of obesity, in particular. Furthermore, the data presented here suggest possible “protective” vasodilator activities of ROS/•OH, which are altered in obesity. The possible pathophysiologicial significance of these observations in human obesity and/or insulin resistance remains yet to be shown. However, the activation of direct and indirect ROS-mediated vasodilator mechanisms may explain the clinical observation that only some, but not all, obese patients develop hypertension.
This work was supported by the Swiss National Science Foundation (SCORE 3200-058426.99, 3232-058421.99, and 3200-108258/1), and the Hanne Liebermann Stiftung, Zürich.
The authors gratefully acknowledge the expert technical assistance of Emerita Ammann and thank Senju, Osaka, Japan for providing EPC-K1.
↵* These authors contributed equally to this article.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society