INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISUCUSSION REFERENCES

THE PHYSIOLOGICAL PRODUCTION OF reactive oxygen species (ROS) such as superoxide and H₂O₂ is necessary for maintenance of cell function and cell homeostasis. However, a surplus generation of ROS, termed oxidative stress, has been implicated in pathophysiological mechanisms. In the vasculature, oxidative stress has been recently found to play a role in the pathophysiology of hypertension and atherosclerosis. Recent studies have shown that an excess amount of oxidant induces profound vascular endothelial cell (EC) dysfunction such as vascular permeability aggravation via stimulation of platelet-activating factor production (27), abnormal coagulation due to inhibition of prostaglandin I₂ synthesis (38), and cell lysis (40).

ECs participate in the regulation of vascular tone and remodeling by modulating the secretion of vasoactive substances. Itoh et al. (18, 19) reported that vasoactive substances regulate not only vascular tone but also vascular growth; i.e., angiotensin II, a potent vasoconstrictor, promotes vascular growth. Furthermore, we elucidated that C-type natriuretic peptide (CNP) is secreted from ECs (24, 37) and regulates vascular remodeling (6, 23, 35, 36).

Endothelin (ET) is a 21-amino acid peptide isolated from the supernatant of porcine aortic ECs (39). ET exerts potent cardiovascular actions, the most prominent of which is vascular constriction. Increased ET-1 expression has been observed after treatment of cultured ECs with growth factors and cytokines such as thrombin (7), transforming growth factor- (25), tumor necrosis factor- (26), or insulin (32). Vasoactive substances, such as angiotensin II (5, 15) and antidiuretic hormone (15), also increase ET-1 mRNA expression in ECs. On the other hand, cGMP-elevating vasodilators, such as nitric oxide (NO) (14) and atrial natriuretic peptide (13), have been known to decrease endothelial ET secretion. Recently, Michael et al. (29) reported that oxidative stress elicited by glucose oxidase decreased the expression of ET in cultured rat pulmonary arterial ECs; however, the mechanisms by which oxidants regulate ET production remain unclear.

Adrenomedullin (Am), which was first isolated from human pheochromocytoma, is a potent vasorelaxant peptide that is secreted from vascular ECs (21). Am exerts a hypotensive effect and an antiproliferative effect through a cAMP-dependent process (20). Chini et al. (1) reported that Am acts as a cytoprotective autacoid to suppress oxidative stress through the cAMP signaling pathway. In vascular smooth muscle cells (VSMCs), exogeneously administered Am has been reported (22) to suppress ET-1 production.

The intracellular Ca²⁺ concentration ([Ca²⁺]_i) is an important significant signal transduction in the cardiovascular system. It was reported that oxidants increased [Ca²⁺]_i in ECs (9). Furthermore, treatment with Ca²⁺ ionophores inhibited ET production in ECs (31). These findings suggest that oxidative stress suppresses ET production by modulating [Ca²⁺]_i. The underlying regulation mechanisms between ET production and [Ca²⁺]_i during oxidative stress are not fully understood.

In the present study, to elucidate the significance of oxidative stress in the modulation of endothelial functions, we examined the effects of H₂O₂ on the expression of two endothelium-derived vasoactive peptides, ET and Am, and their interaction. Moreover, we investigated whether alteration of [Ca²⁺]_i due to oxidative stress contributes to ET production or not.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISUCUSSION REFERENCES

Materials. Human Am was obtained from the Peptide Institute (Osaka, Japan). Menadion sodium bisulfite (MSB), N^G-nitro-L-arginine methyl ester (L-NAME), catalase, 8-bromo-cAMP (8-Br-cAMP), 8-Br-cGMP, actinomycin D, forskolin, fura 2-acetoxymethyl ester (AM), ionomycin, A-23187, thapsigargin (TSG), EGTA, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA), and cremophor EL were purchased from Sigma (St. Louis, MO). H₂O₂ was purchased from Santoku (Tokyo, Japan). We purchased dl--tocopherol from Nacalai Tesque (Kyoto, Japan).

Cell culture. Bovine artery endothelial cells (BAECs) were isolated from the adult bovine carotid artery by digestion with collagenase and elastase (37). The BAECs were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Hazleton Biologics; Lenexa, KS) with 100 units of penicillin and 0.1 mg/ml streptomycin, as we previously reported. The cultures were maintained at 37° in a humidified atmosphere of 95% O₂-5% CO₂. BAECs harvested between passages 8 and 16 were used in the experiments.

Preparation of conditioned media. BAECs were grown to confluence in 12-well dishes in DMEM supplemented with 10% fetal bovine serum. After the medium was changed to DMEM with 0.5% fetal bovine serum, the BAECs were treated with 0.25-0.75 mM H₂O₂ for specific periods. MSB (0.25-1.0 mM), a redox cycling drug, was also administered for 3 h to examine the effect of oxidative stress. To investigate the effect of antioxidants, the cells were pretreated with catalase, a specific reductase of H₂O₂, and dl--tocopherol for 30 min before the addition of H₂O₂. To address the effect of cAMP and cAMP-elevating agents on ET secretion from BAECs, the cells were treated with 8-Br-cAMP, Am, and forskolin. To examine the effect of [Ca²⁺]_i elevation on ET production, ionomycin and TSG were added. The harvested supernatants were stored at 20°. The BAECs were enumerated after incubation with or without H₂O₂ to confirm cell viability under oxidative stress.

Radioimmunoassay for ET-1 and Am. After incubation, the concentrations of ET and Am in the media were measured by specific radioimmunoassays (RIAs) developed by us (30, 33). In the RIA for ET-1, the cross-reactivities with ET-2, ET-3, and Big ET-1 were 100%, 60%, and 100% on a molar basis, respectively. A concentration giving half-maximal response (EC₅₀) was achieved at a concentration of 2.4 fmol/tube and minimal inhibition was detected at 0.2 fmol/tube. H₂O₂ had no effect on ET-like immunoreactivity during the assay. We measured synthetic ET (22 pg/tube) and the same amount of ET mixed with 5 µM H₂O₂ by RIA and compared the ET-like immunoreactivity. The ET-like immunoreactivity in the tube containing synthetic ET only was 22.4 ± 0.4 pg/tube (n = 4) and that containing synthetic ET and 5 µM H₂O₂ was 22.9 ± 0.3 (n = 4). In the RIA for Am, the cross-reactivity with calcitonin gene-related peptide-1, which has structual homology with Am, was <0.01%. EC₅₀ was achieved at a concentration of 8.0 fmol/tube and minimal inhibition was detected at 1.0 fmol/tube.

RNA extraction and Northern blot analysis. BAECs grown to confluence in 10-cm dishes were incubated with several agents for specific periods. Total cellular RNA was extracted by a method previously reported (36). RNA samples (25 µg) were applied to 1.0% formamide-agarose gels and transferred to a nitrocellulose membrane. The samples were then hybridized using rat ET-1 full-length cDNA or bovine Am partial-length cDNA, prepared by polymerase chain reaction by using the sense (408-428 bases; 5'-GCAGCCCGCATCCGAGTCAAG-3') and antisense (602- 622 bases; 5'-GTAATAGAGTCCGACCCAAGC-3') primers as probes. The probes were labeled by random priming with [-³²P]-deoxycytidine 5'-triphosphate. The blot was washed finally with 0.1× saline-sodium citrate buffer at 65°C. Blots were exposed to an imaging plate (Fuji Film, Japan) for 24 h. The signals on the imaging plate were quantified with an image analyzer (BAS2000, Fuji Film) and expressed by the intensity of photo-stimulated luminescence (PSL). The background PSL values were subtracted. PSL is proportional to the adsorbed radiation energy on the imaging plate. The main signals, with a size of 2.3 Kb, were quantified. The density of 18S rRNA stained with ethidium bromide was used to monitor the amount of total RNA in each sample.

Half-life study of ET-1 mRNA. BAECs were incubated in the presence of 10 µg/ml actinomycin D for specific periods, at which time total RNA was extracted. In parallel, the cells were coincubated with actinomycin D and 0.5 mM H₂O₂ for the same period.

cAMP measurement. Basal and H₂O₂-stimulated cAMP production in the medium were measured by RIA as described (30). After incubation for 24 h with or without H₂O₂, the supernatants were collected, succinylated, and mixed with assay buffer. The treated samples were measured with the use of a commercial kit (Yamasa; Tokyo, Japan).

Effects of monoclonal anti-CNP and anti-Am antibodies. To examine the action of endogenous CNP and Am, we utilized a monoclonal anti-CNP antibody (KY-CNP-1) and a monoclonal anti-Am antibody (KY-Am-1), which we previously developed (12, 30). These antibodies in ascites were purified with an Affi-Gel Protein A MAPS II kit (Bio-Rad; Richmond, CA) (30). The neutralizing effects of KY-CNP-1 (10 µg/ml) and KY-Am-1 (50 µg/ml) were confirmed by blockade of CNP-stimulated cGMP production in cultured VSMCs and the Am-induced cAMP elevation in BAECs, respectively (7, 23). The same concentration of mouse immunoglobulin (50 µg/ml, Sigma) was used as a control.

Measurement of [Ca²+]_i. [Ca²⁺]_i was measured as described elsewhere (9). Briefly, the BAECs seeded and grown to confluence on the coverslips were loaded with 2 µM fura 2-AM and 0.01% cremophor EL for 30 min at room temperature in modified Hanks' balanced salt solution composed of (in mM) 137 NaCl, 3.5 KCl, 0.44 KH₂PO₄, 25 NaHCO₃, 0.33 Na₂HPO₄, and 0.5 CaCl₂ with a pH of 7.4. The cells were then washed three times with fresh Hanks' balanced salt solution and incubated for 20 min at 37°C. The coverslips were mounted on the stage of a microscope (Diaphot, Nikon; Tokyo, Japan) that was fitted with a charge-coupled device imaging system (Hamamatsu Photonics; Hamamatsu, Japan). The cells were continuously superfused at 3 ml/min with Hanks' balanced salt solution, to which H₂O₂ was added.

Statistics. Data are presented as means ± SE. The significance of differences between groups was evaluated by repeated-measure analysis of variance, followed by Fisher's protected least-significant difference test. A value of P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISUCUSSION REFERENCES

H₂O₂ decreased ET secretion and suppressed ET-1 mRNA expression. Figure 1A illustrates the effect of H₂O₂ on ET secretion from BAECs. A significant decrease in ET secretion was first detectable 3 h after exposure to 0.5 mM H₂O₂ and was maintained for up to 24 h (P < 0.01 vs. basal secretion, n = 6). At each time point, ~50% suppression compared with control was observed in response to treatment with H₂O₂. Three hours after the administration of H₂O₂ (0.25-0.75 mM), ET secretion was suppressed in a dose-dependent manner, and EC₅₀ was ~0.5 mM, as shown in Fig. 1C (P < 0.01 vs. basal secretion, n = 6). Twenty-four hours of incubation with 0.5 mM of H₂O₂ did not significantly affect the cell number.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   Immunoreactive (ir) endothelin (ir-ET) secretion (A) and ET-1 mRNA expression (B) in bovine arterial endothelial cells (BAECs) more than 24 h after incubation with or without H2O2 (0.5 mM) (n = 3); *P < 0.01 vs. control. Ir-ET secretion (C) and ET-1 mRNA expression (D) in BAECs incubated for 3 h with increasing doses of H2O2 (0.25-0.75 mM) (n = 3); *P < 0.05 and **P < 0.01 vs. control.

MSB decreased ET secretion. Treatment with MSB mimicked the suppressive effect of H₂O₂ on ET secretion. Administration of 0.25-1.0 mM MSB for 3 h remarkably suppressed ET secretion from BAECs in a concentration-dependent manner (Fig. 2, P < 0.01 vs. basal secretion, n = 3). MSB (0.25 mM) suppressed ET secretion by 40% compared with the control.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Ir-ET secretion from BAECs (n = 3) incubated for 3 h with increasing doses of menadion sodium bisulfate (MSB, 0.25-1.0 mM). *P < 0.01 vs. control.

Antioxidants prevented H₂O₂-induced decrease of ET secretion. Pretreatment with 300 U/ml of catalase completely abrogated the H₂O₂-induced decrease in ET secretion (Fig. 3, P < 0.01 vs. H₂O₂ treatment, n = 6). Catalase alone slightly augmented the basal secretion of ET (P < 0.05 vs. basal secretion, n = 6). Pretreatment with 50 µg/ml dl--tocopherol also significantly attenuated the H₂O₂-induced decrease of ET secretion (Fig. 3, P < 0.05 vs. H₂O₂ treatment, n = 4).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Ir-ET secretion from BAECs (n = 6) pretreated for 30 min with 300 µg/ml catalase (CAT) or 50 µg/ml dl--tocopherol (VitE), followed by a 3-h incubation with or without H2O2 (0.5 mM). *P < 0.05, **P < 0.01 vs. control; P < 0.05, P < 0.01 vs. H2O2 alone.

H₂O₂ increased Am secretion and Am mRNA expression with augmentation of cAMP production. Basal Am secretion from BAECs incubated for 24 h without any treatment was 8.1 ± 0.2 × 10² fmol/10⁵ cells (n = 4). Treatment with 0.5 mM of H₂O₂ augmented Am secretion to 14.2 ± 0.5 × 10² fmol/10⁵ cells, as shown in Fig. 4A (P < 0.05 vs. basal secretion, n = 4). Am mRNA expression was also augmented by ~30% with 24-h treatment of 0.5 mM H₂O₂ compared with the control (Fig. 4B). Figure 4C showed that the addition of 0.25-0.75 mM of H₂O₂ resulted in dose-dependent increases of cAMP production for 24 h. H₂O₂ (0.5 mM) significantly augmented the cAMP level in the medium by ~30% compared with the control.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   A: adrenomedullin (ir-Am) secretion from BAECs (n = 3) incubated for 24 h with or without H2O2 (0.25-0.5 mM). *P < 0.05 vs. control. B: Am mRNA expression in BAECs incubated for 24 h with or without H2O2 (0.5 mM). C: ir-cAMP secretion from BAECs incubated for 24 h with or without H2O2 (0.25-0.75 mM). *P < 0.05, **P < 0.01 vs. control.

Exogenously administered Am and cAMP suppressed ET secretion. We then investigated the influence of Am and cAMP on ET secretion. Three hours of treatment with human Am and forskolin, which are well known to increase the intracellular cAMP levels, attenuated ET production from BAECs in a dose-dependent manner (Fig. 5, A and B; P < 0.01 vs. basal secretion, n = 6) and this effect continued for up to 24 h (data not shown). As shown in Fig. 5C, addition of 8-Br-cAMP resulted in a dose-dependent decrease of ET secretion from BAECs (P < 0.01 vs. basal secretion, n = 6).

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Ir-ET secretion from BAECs (n = 3) incubated for 3 h with or without 0.01-1.0 µM Am (A), 10-100 µM forskolin (B), or 10-100 µM 8-bromo-cAMP (8-Br-cAMP) (C). *P < 0.05, **P < 0.01 vs. control.

Neutralizing Am monoclonal antibody attenuated H₂O₂-induced suppression of ET secretion. To examine the significance of Am secreted from BAECs on ET secretion, we performed neutralization experiments with KY-Am-1 on H₂O₂-induced suppression of ET secretion. Treatment with 50 µg/ml KY-Am-1 for 12 and 24 h, respectively, completely inhibited the suppressive effect of H₂O₂, whereas control mouse immunoglobulin did not affect ET levels (Fig. 6; P < 0.01 vs. basal secretion, n = 12). KY-Am-1 alone significantly increased ET secretion compared with the control.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Ir-ET secretion from BAECs pretreated with () or without (+) an anti-Am monoclonal antibody (A-Am-Ab) (50 µg/ml KY-Am-1) or mouse immunoglobulin (control) and incubated for 24 h with or without H2O2 (0.5 mM) (n = 12). *P < 0.05, **P < 0.01 vs. control; P < 0.05 vs. H2O2 alone.

Role of cGMP pathway in ET secretion under oxidative stress. A few studies (13) suggested that NO inhibited ET secretion by elevating cGMP production in VSMCs. We therefore investigated the influence of NO and cGMP on ET secretion in the presence of H₂O₂. Administration of 10 µM 8-Br-cGMP for 3 and 24 h, respectively, decreased ET secretion from BAECs to approximately one-half of that seen with the control (3 h, control, 2.41 ± 0.07, 1.57 ± 0.17; and 24 h, control, 10.79 ± 0.78; 5.82 ± 0.51, ×10² fmol/10⁵ cells, P < 0.01 vs. basal secretion, n = 3). On the other hand, pretreatment with 10 and 100 µM L-NAME did not affect basal ET secretion or H₂O₂-induced suppression of ET secretion (Fig. 7).

View larger version (15K): [in this window] [in a new window]   Fig. 7.   Ir-ET secretion from BAECs (n = 12) pretreated with or without NG-nitro-L-arginine methyl ester (L-NAME) (100 µM) or an anti-C-type natriuretic peptide monoclonal antibody (A-CNP-Ab) (10 µg/ml KY-CNP-1) and incubated for 24 h with or without H2O2 (0.5 mM). *P < 0.01 vs. control.

H₂O₂ transiently increased [Ca²+]_i and ionomycin and TSG suppressed ET production in BAEC. Treatment with Am monoclonal antibody abrogated the H₂O₂-induced alteration of ET secretion for 12 or 24 h; however, it had no effect on ET secretion at 3 h (data not shown). To examine whether an increase in [Ca²⁺]_i affects ET production, we measured [Ca²⁺]_i under oxidative stress. As shown in Fig. 8A, treatment with 1 mM of H₂O₂ markedly increased [Ca²⁺]_i. Furthermore, ionomycin, a Ca²⁺ ionophore, and TSG, an inhibitor of endoplasmic reticulum ATPase, suppressed ET secretion dose dependently (Fig. 8, B and C). On the other hand, EGTA and BAPTA had no effect on H₂O₂-induced suppresion of ET secretion and ET-1 mRNA expression (data not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Intracellular Ca2+ concentrations ([Ca2+]i) in BAECs after treatment with 1.0 mM H2O2 (A). Ir-ET secretion from BAECs incubated for 3 h with or without ionomycin (10-100 µM; B), and thapsigargin (TSG, 0.3-3.0 µM; C). *P < 0.05 and **P < 0.01 vs. control.

	DISUCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISUCUSSION REFERENCES

ROS have been suggested to act as intracellular second messengers affecting gene regulation in various cells (11). In the present study, we demonstrated that ET production was markedly downregulated by oxidative stress. H₂O₂ suppressed ET production both dose and time dependently. The suppressive effect on ET secretion by MSB is consistent with the effect of H₂O₂. These results indicate that oxidative stress is a negative regulator of ET secretion. Treatment with catalase, the specific reductase of H₂O₂, not only inhibited H₂O₂-induced suppression of ET secretion but also significantly augmented basal ET secretion, implying that ECs are constantly exposed to oxidative stress at a certain level.

ECs are continuously subjected to flow-induced shear stress. Some studies (2, 4) recently reported that shear stress caused a sustained activation of prooxidant processes, resulting in redox-sensitive gene expression in human ECs, and that this activation was significantly attenuated by pretreatment with antioxidants. We have already shown that shear stress at a physiological level suppressed the production of ET and mRNA expression of ET-converting enzyme-1 (28). From our preliminary results, shear stress-induced downregulation of ET and ET-converting enzyme-1 suppression is blocked on pretreatment with N-acetyl-cystein, an antioxidant. These findings and our results indicate that ET production may be modulated via a redox-sensitive mechanism in the physiological state in vivo.

In the present study, we assessed ET-1 mRNA stability under oxidative stress by using actinomycin D. Inoue et al. (16) reported that the AUUUA motif, considered to be an RNA-destabilizing element, was found in the 3' untranslated region of ET-1 mRNA. In the current study, the ET-1 mRNA half-life was ~20 min in both the presence and the absence of 0.5-mM H₂O₂. This indicates that H₂O₂ did not significantly alter the ET-1 mRNA half-life, suggesting that the downregulation of ET-1 mRNA under oxidative stress was modulated at the transcriptional level.

In previous studies (17), administration of interleukin-, tumor necrosis factor-, phorbol ester, oxidized low-density lipoprotein, and lysophosphatidyl choline, which are well recognized to generate intracellular ROS, elevated Am production in cultured rat ECs. In the current study, we obtained direct evidence that Am secretion and Am mRNA expression were augmented by oxidative stress. Am exerts several actions through augmentation of cAMP production. In the present study, H₂O₂ increased cAMP production in a dose-dependent fashion. In addition, both 8-Br-cAMP and forskolin, as well as Am, suppressed ET secretion from BAECs dose dependently. We further demonstrated that treatment with a neutralizing monoclonal antibody against Am inhibited the suppressive effect of H₂O₂ on ET secretion for 24 h. These results indicate that Am secretion enhanced by oxidative stress inhibited ET secretion via a cAMP-dependent process in BAECs.

However, the neutralizing monoclonal antibody against Am failed to prevent H₂O₂-induced suppression of ET secretion at 3 h. Because some reports demonstrated the suppressive effect of increase in [Ca²⁺]_i on ET production in ECs, we investigated the role of intracellular Ca²⁺ in ET secretion. As shown in Fig. 8, H₂O₂ markedly increased [Ca²⁺]_i. Furthermore, ionomycin and TSG significantly suppressed ET secretion dose dependently. This effect seems to be dependent on intracellular Ca²⁺ fluxes because EGTA and BAPTA had no effect on ET production with or without H₂O₂. These results suggest that H₂O₂ suppressed ET secretion also via an increase in [Ca²⁺]_i in BAECs.

NO is known to produce cGMP in ECs through activation of soluble guanylate cyclase. It is reported that elevation of the cGMP levels in ECs inhibits ET secretion. Moreover, it is well established that an increase in [Ca²⁺]_i stimulates NO synthesis in ECs. A previous report (34) indicated that treatment with 1.0 mM H₂O₂ increase [Ca²⁺]_i and enhanced NO synthase (NOS) activity in ECs, leading to cell injury. Another group demonstrated that treatment with 1.0 and 10.0 mM H₂O₂ increased endothelial NOS (eNOS) tyrosine-phosphorylation and its interaction with caveolin-1, the coat protein of caveolae (8). On the other hand, Gupta et al. (10) demonstrated that in intact ECs, 0.1-0.5 mM of H₂O₂ had little effect on NOS activity. To determine the possible involvement of the NO-cGMP pathway in the suppression of ET secretion by H₂O₂, we examined the effect of L-NAME. There was no difference in ET secretion between control and L-NAME-treated BAECs. In our previous study (3), inducible NOS and eNOS expression were not significantly altered by the treatment with 0.25-0.75 mM of H₂O₂. These findings suggest that, whereas higher concentrations of H₂O₂ induce NOS activation by extracellular Ca²⁺ influx, activation of NOS at a level high enough to cause suppression of ET secretion may not occur in ECs in response to lower concentrations of H₂O₂ as used in the present study.

We have already shown that CNP induces growth inhibition and redifferentiation of proliferating and dedifferentiated VSMCs (6). We also demonstrated that CNP has the ability to potently augment intracellular cGMP accumulation by binding to the atrial natriuretic peptide-B receptor, which has been shown (23) to be selectively activated by CNP in VSMCs. We have previously (3) revealed that H₂O₂ caused up to a ninefold increase of CNP secretion from BAECs. Therefore, we examined the effect of the CNP neutralizing monoclonal antibody to investigate the possible involvement of the CNP-cGMP pathway in H₂O₂-induced suppression of ET production. The addition of anti-CNP monoclonal antibody KY-CNP-1, however, failed to prevent H₂O₂-induced suppression of ET secretion. It could be speculated that atrial natriuretic peptide-B receptor expression is very scarce in ECs, which we previously reported (35).

Our findings suggest that ECs possess the coordinate gene regulation mechanism to reduce the vasoconstricting peptide (ET) secretion and to increase the vasorelaxant peptide (Am) production in response to oxidative stress. By this mechanism, ECs function to protect against oxidative stress by increasing blood flow via VSMC relaxation and growth inhibition.