Endothelial dysfunction is a common problem associated with hypertension and is considered a precursor to the development of micro- and macro-vascular complications. The present study investigated the involvement of nNOS (neuronal nitric oxide synthase) and H2O2 (hydrogen peroxide) in the impaired endothelium-dependent vasodilation of the mesenteric arteries of DOCA (deoxycorticosterone acetate)-salt-hypertensive mice. Myograph studies were used to investigate the endothelium-dependent vasodilator effect of ACh (acetylcholine). The expression and phosphorylation of nNOS and eNOS (endothelial nitric oxide synthase) were studied by Western blot analysis. Immunofluorescence was used to examine the localization of nNOS and eNOS in the endothelial layer of the mesenteric artery. The vasodilator effect of ACh is strongly impaired in mesenteric arteries of DOCA-salt-hypertensive mice. Non-selective inhibition of NOS sharply reduced the effect of ACh in both DOCA-salt-hypertensive and sham mice. Selective inhibition of nNOS and catalase led to a higher reduction in the effect of ACh in sham than in DOCA-salt-hypertensive mice. Production of H2O2 induced by ACh was significantly reduced in vessels from DOCA-salt-hypertensive mice, and it was blunted after nNOS inhibition. The expression of both eNOS and nNOS was considerably lower in DOCA-salt-hypertensive mice, whereas phosphorylation of their inhibitory sites was increased. The presence of nNOS was confirmed in the endothelial layer of mesenteric arteries from both sham and DOCA-salt-hypertensive mice. These results demonstrate that endothelial dysfunction in the mesenteric arteries of DOCA-salt-hypertensive mice is associated with reduced expression and functioning of nNOS and impaired production of nNOS-derived H2O2. Such findings offer a new perspective for the understanding of endothelial dysfunction in hypertension.
The endothelium plays a significant role in maintaining vascular homoeostasis by synthesizing and releasing several vasodilator factors, including prostacyclin, NO (nitric oxide) and EDHFs (endothelium-derived hyperpolarizing factors). H2O2 may be an EDHF for the mesenteric artery. NOS proteins are a key family of NO-producing enzymes, and nNOS-derived H2O2 has been shown recently to be present in conductance vessels. However, there is currently no information regarding the physiological role of nNOS/H2O2 in the mesenteric arteries of mice and their response to hypertension.
In the present study, novel information is provided regarding endothelial homoeostasis in hypertension. In particular, our findings suggest that a decrease in nNOS-derived H2O2 production contributes to endothelial dysfunction in hypertensive mice. In addition, inhibition of nNOS/H2O2 participates in the regulation of blood pressure, as it impairs ACh-induced hypotension in normotensive mice.
These observations provide a new perspective for understanding the mechanisms involved in vascular dysfunction in hypertension. Considering the large number of individuals with drug-resistant hypertension, future research is needed to explore the modulation of this pathway as a potential therapeutic target.
The endothelium plays a significant role in maintaining vascular homoeostasis by synthesizing and releasing several vasodilator factors, including prostacyclin, NO (nitric oxide) and EDHFs (endothelium-derived hyperpolarizing factors). Decreased endothelium-dependent vasodilation, which is associated with endothelial dysfunction, is a hallmark of most forms of hypertension [1,2]. DOCA (deoxycorticosterone acetate)-salt hypertension is a salt-sensitive model of hypertension with markedly depressed plasma renin activity and is characterized by low angiotensin II levels . The endothelial dysfunction found in DOCA-salt hypertension is in general associated with a decreased production or bioavailability of NO, as well as with an increased production of O2− (superoxide anion) derived from NADPH oxidase or uncoupled eNOS (endothelial nitric oxide synthase) .
Many studies have revealed that endothelium-derived H2O2 (hydrogen peroxide) acts as an EDHF in animals and humans [5–7]. nNOS (neuronal nitric oxide synthase) is expressed in the endothelium of the mouse mesenteric artery and aorta [8,9]. Aside from NO, nNOS also produces H2O2 by self-dismutation of superoxides or by electron transport in the reductase and oxidase domains of nNOS . H2O2 was described as a major endothelium-dependent relaxing factor in mouse aorta and as an EDHF in mouse mesenteric artery [7–9]. Moreover, the presence of nNOS seems to be essential for endothelium-dependent hyperpolarization in mouse mesenteric artery . Whereas the involvement of H2O2 and nNOS in endothelial dysfunction in a model of atherosclerosis has been described , their involvement in hypertension remains unknown. The aim of the present study was to investigate the contribution of nNOS-derived H2O2 to impaired endothelium-dependent relaxation in the mesenteric arteries of DOCA-salt-hypertensive mice (DOCA-salt mice). We hypothesized that the nNOS/H2O2 axis contributes to the endothelium-dependent relaxation in the mesenteric artery and that its impairment may contribute to endothelium dysfunction in this model of hypertension.
All studies were performed in male Swiss mice (12–15 weeks of age) purchased from the Animal Facility Center of the Federal University of Minas Gerais (CEBIO, UFMG). Standard diet and tap water were supplied ad libitum. All mice were housed six per cage at a constant temperature (24°C) with a 12-h light/12-h dark cycle. During all experiments, animal welfare was monitored at least twice daily, in order to certify that they had access to food and water, as well as to verify the successful recovery of the animals after surgery.
Induction of hypertension
Mice (25–30 g) were unilaterally nephrectomized under anaesthesia using a solution containing ketamine (80 mg/ml) and xylazine (10 mg/ml), administered once intraperitoneally. The skin on the left flank was shaved, and a 1.5-cm incision was made through the skin and abdominal muscle caudal to the rib cage. The left kidney was removed after being externalized, and the renal artery was ligated with a vein with 4-0 silk sutures (Ethicon). The skin and muscle layers were closed separately with 4-0 silk sutures. A small area between the shoulder blades was shaved, and a 1-cm incision was made for implanting DOCA pellets. Implantation was performed as described by Rhaleb et al.  to provide a dose of 1 mg/kg. The DOCA-salt mice were given water containing 0.9% NaCl and 0.2% KCl. The sham mice were also unilaterally nephrectomized, received a pellet without DOCA and were given tap water. All mice were placed on standard pellet rodent chow. Upon recovery, the mice were housed under standard conditions for 4 weeks, after which their SBP (systolic blood pressure) was determined by the tail-cuff method .
Mice were killed by decapitation without use of anaesthetics in order to prevent interference with the sustained phase of the contractile response . The abdomen was immediately opened, and the mesenteric arcade was removed. Branch II of the mesenteric resistance arteries (157±2.4 μm and 155.7±2.4 μm for sham and DOCA-salt mice respectively) in the mice were cleaned of fat and connective tissue, and a segment approximately 1.6–2.0 mm in length was removed as described previously . The rings were then mounted in PSS (physiological salt solution) with the following composition: 119 mM NaCl, 4.7 mM KCl, 0.4 mM KH2PO4, 14.9 mM NaHCO3, 1.17 mM MgSO4, 2.5 mM CaCl2 and 5.5 mM glucose. Mechanical activity was recorded isometrically using a wire myograph (DMT). After mounting, the artery was stretched to a length that yielded a circumference equivalent to 90% of that given by an internal pressure of 100 mmHg; this required a load of approximately 200 mg. The vessel was maintained for an equilibration period of 60 min. Concentration–response curves of freshly prepared ACh (acetylcholine), SIN-1 (3-morpholinosydnonimine hydrochloride), a spontaneous NO donor or H2O2 (0.0001–100 μM) (Sigma) were conducted in vessels pre-contracted to the same tension level (approximately 2.8 mN/mm) with submaximal concentrations of phenylephrine (1.0 μM and 3.0 μM for DOCA-salt and sham mice respectively). Changes in the isometric tension were analysed using PowerLab software (ADInstruments). L-NAME (NG-nitro-L-arginine methyl ester) (300 μM), a non-selective inhibitor of NOS, and L-ArgNO2-L-Dbu (L-ArgNO2-L-Dbu-NH2 2TFA) (1 μM), a selective inhibitor of nNOS, were incubated 30 min before the ACh concentration–response curve measurements. A similar protocol was used with catalase (2400 units/ml) to verify the participation of H2O2. Lower concentrations of phenylephrine (0.3 μM and 1 μM for DOCA-salt and sham mice respectively) were used with L-NAME to compensate for the shift to the left in the concentration–response curve. For L-ArgNO2-L-Dbu and catalase, no reduction in the concentration of phenylephrine was necessary.
H2O2 production in the mesenteric artery was measured using carbon microsensors with an H2O2-permeable membrane (ISO-HPO100, World Precision Instruments). Mesenteric artery segments were removed as described above, incubated in tubes containing PSS and maintained at 37°C. Vessels were stimulated with ACh (10 μM), after which the preparations were incubated for 30 min with either L-NAME (300 μM) or L-ArgNO2-L-Dbu (1 μM) and stimulated again with ACh (10 μM). Carbon microsensors were stabilized for 1 h in PSS and then placed next to the lumen of the vessels before ACh (10 μM) treatment. Currents (nA) were measured by microsensors, and H2O2 concentrations were determined by calibration curves of known concentrations of H2O2 (1 nM–10 μM).
Analysis of nitrite was performed using DAN (2,3-diaminonaphthalene) according to the method of Misko et al.  with minor modifications. In brief, mesenteric artery branches were kept in PSS at 37°C in a 5% CO2 atmosphere. After collecting the baseline sample, the mesenteric arteries were immediately pre-incubated with L-NAME (300 μM), L-ArgNO2-L-DBU (1 μM) or PSS for 20 min and then stimulated with a single dose of ACh (10 μM). After 15 min, 100 μl samples were mixed with 10 μl of fresh DAN solution (0.05 mg/l in 0.62 M HCl) in 96-well opaque black plates (Costar®, Corning). The reaction was processed for 10 min at room temperature and protected from light. After this period, the reaction was finalized with 5 μl of NaOH (2.8 M), and the absorbance was determined using a spectrofluorimeter (excitation at 365 nm, emission at 440 nm; Cary Eclipse Microplate reader, Varian). Standard curves were generated using 0.01–10 μM sodium nitrite (NaNO3).
Western blot analysis
Western blotting was performed as described previously , with some modifications. Briefly, the frozen mesenteric arteries (branches II and III) were homogenized in lysis buffer comprising 150 mM NaCl, 50 mM Tris/HCl, 5 mM sodium EDTA and 1 mM MgCl2. The buffer also contained 0.5% SDS plus protease inhibitors (SigmaFAST®; Sigma) and 1% Triton X-100. Proteins were denatured and separated by denaturing SDS/PAGE (7.5% gel). Proteins (25 μg) were then transferred on to a PVDF membrane (Immobilon-P; Millipore). The blots were blocked at 18°C with 2.5% BSA in PBS containing 0.1% Tween 20 before incubation with rabbit polyclonal anti-nNOS antibody (diluted 1:1000; SC-5302, Santa Cruz Biotechnology), mouse monoclonal anti-nNOS Ser852 antibody (diluted 1:1000; SC-19826, Santa Cruz Biotechnology), rabbit polyclonal anti-eNOS antibody (diluted 1:1000; SC-654, Santa Cruz Biotechnology), goat polyclonal anti-eNOS Ser1177 antibody (diluted 1:1000; SC-12972, Santa Cruz Biotechnology), goat polyclonal anti-eNOS Thr495 antibody (diluted 1:1000; SC-19827, Santa Cruz Biotechnology) or rabbit polyclonal anti-tubulin antibody (diluted 1:4000; SC-55529, Santa Cruz Biotechnology) at room temperature. The antibodies were detected by chemiluminescent reaction (ECL+ kit; GE Healthcare) followed by densitometric analyses using ImageQuant software.
An immunofluorescence assay was performed to identify the localizations of nNOS, eNOS and CD31 in sections of the mesenteric artery (branches II and III). Cryosections (5 μm) of the mesenteric artery from sham and DOCA-salt mice were embedded in Tissue-Tek OCT compound, fixed in acetone (15 min) and rinsed in PBS wash buffer (1.5% BSA and 0.3% Triton X-100 in PBS). This was followed by blocking procedures (0.3% Triton X-100 and 3% BSA in PBS, 30 min) and incubation of secondary antibodies with the alternating primary antibodies removed. Sections were stained overnight with goat anti-CD31 antibody (diluted 1:100; SC-1506, Santa Cruz Biotechnology) or mouse anti-nNOS antibody (diluted 1:100; SC-5302, Santa Cruz Biotechnology). The secondary antibodies were Alexa Fluor® 633-conjugated donkey anti-goat IgG (diluted 1:500; A-21082, Invitrogen), Alexa Fluor® 488-conjugated chicken anti-goat IgG (diluted 1:500; A-21467, Invitrogen) and Alexa Fluor® 633-conjugated donkey anti-mouse IgG (diluted 1:500; A-21050, Invitrogen). Cell nuclei were counterstained with DAPI. The microscopic data from the present study were obtained using the Nikon epifluorescence microscope in the Image Acquisition and Processing Center (CAPI-ICB/UFMG). The intensity of fluorescence was analysed in the areas of interest using ImageJ 1.42q software (NIH) as described previously .
Direct measurement of SBP and DBP (diastolic blood pressure)
Mice were anaesthetized with ketamine (80 mg/ml) and xylazine (10 mg/ml) administered simultaneously and intraperitoneally. The mice were placed on a heating pad to maintain constant body temperature. The right carotid artery and the left jugular vein were cannulated for analysing blood pressure and for infusion of drugs respectively. The blood pressure was monitored with a pressure transducer (TSD 104A, Biopac Systems) connected to a data acquisition system (MP100; Biopac Systems). Data were acquired and analysed using AcqKnowledge Software (Biopac Systems). After a stabilization period of 10 min, SBP and DBP were recorded over 4 min (basal). They were then taken again following an infusion of drugs for another 4 min to assess the return to the basal level. The drugs infused were ACh (1 μg/kg), TRIM [1-(2-trifluoromethylphenyl)imidazole] (200 μg/kg) plus ACh (1 μg/kg) and catalase (6 units/kg) plus ACh (1 μg/kg). TRIM and catalase were infused 5 min before ACh.
Results are expressed as means±S.E.M. Two-way ANOVA was used to compare all concentration–response curves, as well as the experiments involving the measurement of H2O2 and nitrite. Bonferroni's post-hoc test was only used for analysis of H2O2 and nitrite values. Student's t test was used for all other experiments. The ΔAUC (area under the curve) was calculated as the difference between the concentration–response curves in the absence and the presence of different drugs in the mesenteric arteries from sham and DOCA-salt mice. All statistical analyses were calculated using Prism 4.2 software (GraphPad) and were considered to be significant when P<0.05.
All animal work was performed under the guidelines for the humane use of laboratory animals at our institute and with the NIH Guide for the Care and Use of Laboratory Animals. All procedures were approved by the ethics committee of the Federal University of Minas Gerais (UFMG; protocol #227/08).
The concentration-dependent vasodilator effect induced by ACh was strongly inhibited in the mesenteric rings from the DOCA-salt mice (SBP 180±5 mmHg; P< 0.001) compared with that in the normotensive sham mice (SBP 128±4 mmHg), whereas the concentration-dependent effect of SIN-1 was similar in DOCA-salt and sham mice (Figure 1). Non-selective inhibition of NOS with L-NAME strongly inhibited the ACh-induced vasodilator effect in both the DOCA-salt and sham mice. The residual vasodilation was approximately 15% of the respective controls for both groups (Figure 2A). Selective inhibition of nNOS with L-ArgNO2-L-Dbu produced a significant inhibitory effect on the vessels from sham and DOCA-salt mice (Figure 2B). The comparison between sham and DOCA-salt mice, represented by the ΔAUC of the effects of L-NAME (Figure 2C) and L-ArgNO2-L-Dbu (Figure 2D), demonstrates that both inhibitors had a reduced effect on the mesenteric arteries of DOCA-salt mice. Degradation of endogenous H2O2 with catalase inhibited the response to ACh in the mesenteric arteries of sham and DOCA-salt mice (Figure 3A), although this reached a higher level of statistical significance in sham (P<0.001) than in DOCA mice (P<0.05). However, catalase did not modify the vasodilator effect of SIN-1 in arteries from DOCA-salt or sham mice (Figure 3B). In addition, the concentration–response curves induced by exogenous H2O2 were similar in the mesenteric arteries of both sham and DOCA-salt mice (Figure 3C). Finally, comparison of the effect of catalase in the mesenteric arteries from sham and DOCA-salt mice, represented by the ΔAUC, shows that catalase had a significantly smaller inhibitory effect on the vasodilator effect of ACh in DOCA-salt mice (Figure 3D).
Concentration–response curves of (A) ACh and (B) SIN-1 in mesenteric resistance arteries with a functional endothelium from sham and DOCA-salt mice
Participation of nNOS in the vasodilator effect of ACh in mesenteric arteries from sham and DOCA-salt mice
Participation of endothelium-derived H2O2 in the vasodilator effect of ACh in mesenteric arteries from sham and DOCA-salt mice
Measurement of H2O2 levels with carbon microsensors showed that ACh increases the production of H2O2 in sham, but not in DOCA-salt, mice (Figure 4). Non-selective inhibition of NOS with L-NAME inhibited the production of H2O2 in both the sham and DOCA-salt (Figure 4) mouse mesenteric arteries. However, the selective inhibition of nNOS with L-ArgNO2-L-Dbu demonstrated a significant effect only in sham mice (Figure 4). Altogether, these results suggest that nNOS-derived H2O2 is involved in ACh-induced vasodilation in murine mesenteric resistance arteries, but its production is blunted in the arteries of hypertensive animals.
Effect of non-selective (
L-NAME) NOS inhibition and selective ( L-ArgNO2- L-Dbu) nNOS inhibition on the production of H2O2 induced by ACh in the mesenteric arteries from (A) sham and (B) DOCA-salt mice
In mesenteric arteries from sham and DOCA-salt mice, ACh increased the production of nitrite, but the concentrations of nitrite were significantly different in the two types of mice (Figure 5). Non-selective inhibition of NOS with L-NAME inhibited the production of nitrite in both the sham and DOCA-salt mesenteric arteries (Figure 5). A similar effect was observed after selective inhibition of nNOS with L-ArgNO2-L-Dbu (Figure 5).
Effect of non-selective (
L-NAME) NOS inhibition and selective ( L-ArgNO2- L-Dbu) nNOS inhibition on the production of nitrite induced by ACh in the mesenteric arteries from (A) sham and (B) DOCA-salt mice
Western blot analysis
The expressions of eNOS and nNOS were evaluated in the mesenteric arteries from sham and DOCA-salt mice. As illustrated in Figure 6(A), the expression level of eNOS was significantly reduced in the DOCA-salt mice compared with that in the sham mice. In addition, the phosphorylation levels at the activation and inactivation sites of eNOS were analysed. There was a reduction in phosphorylated Ser1177 (Figure 6B) and a substantial increase in phosphorylated Thr495 in the mesenteric arteries of DOCA-salt mice (Figure 6C). Interestingly, the level of nNOS expression was also significantly lower in the DOCA-salt mice than in the sham mice (Figure 7A). Furthermore, the phosphorylation levels at the nNOS inactivation site Ser852 were increased in the DOCA-salt mice (Figure 7B).
Western blot analysis of the total amount of (A) eNOS and the phosphorylation levels of (B) Ser1177 and (C) Thr495 in the mesenteric arteries of sham and DOCA-salt mice (upper panels) and corresponding cumulative data (lower panels)
Western blot analysis of the total amount of (A) nNOS and of the phosphorylation levels of (B) Ser852 in the mesenteric arteries of sham and DOCA-salt mice (upper panels) and corresponding cumulative data (lower panels)
Fluorescence microscopy was performed to assess the immunolocalization of nNOS in the endothelium of the mesenteric arteries from sham and DOCA-salt mice. As illustrated in Figure 8, immunostaining for nNOS is seen in the endothelium of mesenteric arteries, as the fluorescence signal for nNOS antibodies co-localized with CD31, a surface marker for endothelial cells. The fluorescence intensity for nNOS was lower in DOCA-salt mice (2.4±0.3) than in sham mice (3.9±0.4; P<0.05) when normalized by the fluorescence intensity of CD31.
Immunofluorescence detection of nNOS in endothelium-intact mesenteric arteries from sham (upper panels) and DOCA-salt (lower panels) mice
Blood pressure measurements
Figure 9 illustrates the effect of the intravenous infusion of ACh in the absence or presence of TRIM or catalase in sham and DOCA-salt mice. In sham and DOCA-salt mice, the respective basal SBP (89.8±4.1 and 154.5±3.3 mmHg; P<0.001) and DBP (68.8±2.7 and 97.5±4.8 mmHg; P<0.01) were significantly different. An intravenous infusion of ACh led to a significant reduction in SBP and DBP in sham and DOCA-salt mice. TRIM and catalase reduced the effect of ACh on both SBP and DBP of sham mice (Figures 9C and 9D). However, in DOCA-salt mice, TRIM did not appear to mediate the ACh effect, whereas catalase only mediated the effect on SBP (Figures 9C and 9D).
Effect of selective inhibition of nNOS and degradation of H2O2 on SBP and DBP induced by intravenous infusion of ACh in sham and DOCA-salt mice
The primary findings of the present study can be summarized as follows: (i) nNOS-derived H2O2 plays an important physiological role in controlling the vasodilation of mesenteric resistance arteries; and (ii) the reduction in the expression and function of nNOS results in decreased H2O2 production in the mesenteric arteries of DOCA-salt mice. These observations suggest that altered regulation of the nNOS/H2O2 axis contributes to endothelial dysfunction in hypertension.
In the vascular system, eNOS-derived NO plays a significant role in the endothelial regulation of vascular tone [17,18]. However, a physiologically relevant role for nNOS has been attributed to the modulation of myogenic tone , cerebral blood flow  and systemic arterial pressure . Recent reports have shown that nNOS is constitutively expressed in the endothelium of the mouse aorta and produces H2O2 as well as NO [7,9]. H2O2 has been considered to be an EDHF in mesenteric [5,6], cerebral  and coronary  arteries.
The present study gives evidence for the involvement of nNOS-derived H2O2 in ACh-mediated vasodilation in mesenteric resistance arteries and in controlling the blood pressure. The following results support this hypothesis: (i) ACh produces an increase in H2O2 production, the vasodilation of mesenteric vessels and a reduction in the blood pressure, all of which are decreased by selective pharmacological inhibition of nNOS; (ii) catalase, which specifically targets H2O2, induces a pronounced decline in ACh-mediated vasodilation; (iii) exogenous H2O2 causes vasodilation in mesenteric arteries; (iv) nNOS was shown to be expressed in the endothelium of mesenteric arteries; and (v) in vivo treatment with an nNOS inhibitor and catalase mediated the reduction in blood pressure induced by ACh in normotensive, but not hypertensive, mice. These findings provide a new perspective in the understanding of the mechanisms involved in endothelial regulation of vascular tonus. Moreover, we describe in the present paper a new mechanism for endothelial dysfunction during hypertension.
There are two important points to consider regarding the above hypothesis. First, although both nNOS and eNOS generate NO, the enzymatic pathway is isoenzyme-specific [24,25]. Unlike eNOS, nNOS produces O2− and H2O2 before NO formation under physiological conditions. H2O2 can be formed through the self-dismutation of superoxides [10,26,27] or directly by electron transport in the reductase and oxidase domains of nNOS. In addition, nNOS has two cycles, each with two mono-oxygenase reactions, to synthesize two additional molecules of H2O2 and one molecule of NO . Secondly, it is important to emphasize that L-ArgNO2-L-Dbu has more than 1500-fold selectivity to nNOS over eNOS  and that the concentration used in the present study is 200-fold lower than the Ki for eNOS. Moreover, in previous studies where nNOS was knocked down, L-ArgNO2-L-Dbu did not induce any additional inhibition of the remaining vasodilation or production of NO induced by ACh [8,9]. Therefore the results of the present study are compatible with a selective inhibition of nNOS as reported previously and with its participation in the ACh-induced vasodilation of the mesenteric arteries of mice [8,9,11].
Regardless of the aetiology of hypertension (essential, renovascular, malignant or pre-eclamptic), impairment of endothelial regulation of vascular functions is a result [29,30]. The most discernible phenomenon of this dysfunction is the attenuated endothelium-dependent vasodilation observed in hypertensive animals and humans [31,32]. Such dysfunction is also found in DOCA-salt mice [4,33,34], a fact confirmed in the present study, as endothelium-dependent vasodilation induced by ACh was strongly impaired.
Endothelial dysfunction in hypertension has been commonly associated with a reduction in eNOS-derived NO production and bioavailability [35,36]. The impaired eNOS function in hypertensive subjects has been associated with uncoupling of the enzyme, resulting in a reduction of NO production and increase in the formation of O2− [4,37]. Our measurements of nitrite indirectly demonstrated a reduction in NO production and our Western blot experiments showed that the expression of eNOS is reduced in hypertensive animals. In addition, the phosphorylation of the activation site residue Ser1177 was reduced in the mesenteric arteries of DOCA-salt mice compared with that in sham mice. Simultaneously, increased phosphorylation of the inactivation site residue Thr495 was also observed. These changes in phosphorylation of the enzyme are indicative of the reduced functioning of eNOS in the mesenteric arteries of DOCA-salt mice. As the uncoupling of eNOS is associated with a compensatory increase in its expression, the present results do not suggest the uncoupling of eNOS in DOCA-salt mice .
The role of nNOS/H2O2 in endothelial dysfunction in hypertension has not yet been reported. In the present study, we provide consistent evidence showing that an impairment in the nNOS-derived H2O2 pathway contributes to endothelial dysfunction in mesenteric resistance arteries from DOCA-salt mice. Consistent with this proposal, selective pharmacological inhibition of nNOS with L-ArgNO2-L-Dbu caused a reduction in ACh-induced vasodilation that was significantly dampened in hypertensive animals compared with that in control animals. In addition, our results showed decreased expression and function of nNOS in hypertensive animals. The activity level of nNOS is regulated by phosphorylation at several sites . Phosphorylation at Ser852 by CaMKII (Ca2+/calmodulin-dependent protein kinase II) reduces nNOS activity . To activate nNOS, Ser1412 has to be phosphorylated by protein kinase B (Akt) . Additionally, phosphatase 1 dephosphorylates nNOS at Ser852, increasing its activity . Our experiments clearly showed that phosphorylation at Ser852 was increased. It is unclear whether this reflects an increase in the expression of CaMKII or a decrease in the expression of phosphatase 1 in DOCA-salt mice. However, in DOCA-salt rats, Giachini et al.  demonstrated that decreased expression of phosphatase 1 is involved in the increased reactivity of mesenteric arteries. In addition, the amount of total nNOS protein was significantly reduced in the mesenteric arteries of DOCA-salt mice. Another indication of the decreased functionality of nNOS was the reduced amount of H2O2 produced upon stimulation with ACh, which is sensitive to nNOS inhibition, in DOCA-salt vessels compared with that produced in sham mice. The participation of H2O2 in the vasodilator effect of ACh was also investigated in the presence of catalase. This enzyme reduced the effect of ACh in sham mice, but had just a minor effect on the arteries of DOCA-salt mice; catalase had no impact on the vasodilator effect of SIN-1. Altogether, our results suggest that the nNOS/H2O2 axis is impaired in the mesenteric arteries of DOCA-salt mice.
Note that in the presence of L-ArgNO2-L-Dbu, the remaining vasodilator effect of ACh was 65.6±5.8% in sham mice and 32.6±5.9% in DOCA-salt mice. Considering the L-NAME-sensitivity of ACh-induced vasodilation, which is the result of simultaneous activation of nNOS and eNOS, as demonstrated previously in mice aorta [8,9], the vasodilation observed after inhibition of nNOS is probably a consequence of the activation of eNOS. As the ACh-induced vasodilation in nNOS-inhibited arteries is lower in DOCA-salt mice than in sham mice, these results confirm the dysfunction of the eNOS/NO pathway and its contribution to the impairment of the endothelium-dependent ACh-induced vasodilation of mesenteric arteries from DOCA-salt mice, as described previously .
In addition to eNOS and nNOS, iNOS (inducible nitric oxide synthase) also seems to be involved in the induction or progression of the vascular dysfunction associated with hypertension . The expression of iNOS is increased in hypertensive patients , and its inhibition reverses the endothelial dysfunction [44,45]. Once expressed, iNOS produces a large amount of NO, promoting nitrosative stress through the formation of peroxynitrite and an up-regulation of arginase activity [45–47]. Moreover, iNOS also produces O2− that is transformed to H2O2 by spontaneous formation and by the action of superoxide dismutase . The role of iNOS in the impaired endothelium-dependent response to ACh in mesenteric arteries from DOCA-salt mice was not investigated in the present study. However, the reduced quantities of NO and H2O2 observed in DOCA-salt mice do not support the direct involvement of iNOS in the impaired response to ACh. The role of iNOS in the development of endothelial dysfunction in DOCA-salt mice cannot be ruled out; however, it is certainly an important topic for future investigations.
Therefore, in addition to the well-established reduction in expression and activation of eNOS [4,49,50] confirmed in the present work, our results demonstrate the importance of the expression and functionality of nNOS to the endothelial dysfunction observed in the mesenteric arteries of DOCA-salt mice.
In summary, the present study indicates that nNOS-derived H2O2 plays a relevant role in the endothelium-dependent vasodilation of murine mesenteric arteries. More importantly, our results show clearly that a reduction in nNOS-derived H2O2 production partially contributes to endothelial dysfunction in DOCA-salt mice. These observations provide a new perspective for understanding the mechanisms involved in vascular dysfunction in hypertension.
Grazielle Silva performed the myograph studies, Western blot studies, measurements of nitrite and wrote the first draft of the paper. Josiane Silva performed the immunofluorescence. Thiago Diniz conducted the measurements of H2O2 in mesenteric arteries. Virginia Lemos and Steyner Cortes designed the study and contributed to the paper.
This study was supported by the Fundação de Apoio a Pesquisa do Estado de Minas Gerais (FAPEMIG) [grant numbers CBB-PPM-00551/11 (to S.F.C.) and PRONEX/CBB-APQ-00746-09 (to V.S.L.)] and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) [grant number 473235/2010-2 (to S.F.C.) and 305163/2010-8 (to V.S.L.)] and the Programa Nacional de Pós-doutorado (PNPD) [grant number 2841/2010] from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).
area under the curve
Ca2+/calmodulin-dependent protein kinase II
diastolic blood pressure
endothelium-derived hyperpolarizing factor
endothelial nitric oxide synthase
inducible nitric oxide synthase
NG-nitro-L-arginine methyl ester
neuronal nitric oxide synthase
physiological salt solution
systolic blood pressure