Background Free radicals such as superoxide and nitric oxide (NO) play a key role in the pathophysiology of atherosclerosis. Mechanical forces such as pulsatile stretch may be involved in free radical production. We studied superoxide production by pulsatile stretch in human endothelial cells.
Methods and Results Human cultured aortic endothelial cells were exposed to pulsatile stretch up to 24 hours, and superoxide production was examined. Short-term stretch for 1 hour (10% average elongation, 50 cycles per minute) increased superoxide production 2.2-fold. This effect was reduced by diphenyleneiodonium chloride, an NADPH oxidase inhibitor, but not by the xanthine oxidase inhibitor oxypurinol or the cyclooxygenase inhibitor indomethacin. Prolonged stretch up to 6 hours increased superoxide production, but it returned to near the control level after 24 hours of stretch. However, after blockade of NO production, 24 hours of stretch did increase superoxide production 2.4-fold compared with 24 hours of stretch alone. Moreover, 24-hour stretch doubled NO synthase (NOS) (III) protein and mRNA expression. The tetrahydrobiopterin synthesis inhibitor 2,4-diamino-6-hydroxypyrimidine had no effect on unstretched cells but doubled superoxide production compared with 24-hour stretch alone; this increase was halved by cotreatment with 6-methyl-5,6,7,8-tetrahydropterine, a lipid-soluble form of tetrahydrobiopterine.
Conclusions Short-term stretch increased superoxide production from human aortic endothelial cells via NADPH oxidase and NOS (III), whereas prolonged stretch increased both superoxide and NO production. The increase in NOS (III) protein with prolonged stretch acts as a scavenger mechanism whereby NO inactivates superoxide. Tetrahydrobiopterin determines the balance of superoxide and NO production from NOS (III) after prolonged stretch in which NOS (III) level is upregulated.
Endothelial cells lie in a strategic anatomical position between the circulating blood and vascular smooth muscle. They are thus most exposed to physical forces such as shear stress,1 hydrostatic pressure,2 and pulsatile stretch, particularly in the arterial circulation. Changes in local flow acutely increase nitric oxide (NO) release in vitro1 and cause flow-dependent vasodilation in vivo.3 On the other hand, increases in hydrostatic pressure cause vasoconstriction,1 reduce NO production,4 and increase endothelin release.5 This may explain the greater endothelium-dependent relaxation in arteries than veins.6
Pulsatility and hence rhythmic stretching of vascular cells are equally important in the pathophysiology of atherosclerosis. In bovine endothelial cells, pulsatile stretch augments NO synthase (NOS) expression and NO production.7 The biological effects of NO, however, are determined not only by the amounts produced but also by the degree of inactivation by superoxide.8 9 Indeed, endothelial cells are a source of superoxide,10 11 and NOS is able to produce superoxide.12 Thus, to evaluate the biological effects of mechanical force, simultaneous characterization of both NO and superoxide production is essential to clarify the biological role of such stimuli in the cardiovascular system.
We investigated the effects and time course of pulsatile stretch on superoxide and NO production in human aortic endothelial cells in culture. Our results show for the first time that pulsatile stretch stimulates both superoxide and NO production from endothelial cells and that the concomitant expression of NOS and in turn NO release especially play an important role in regulating the balance of superoxide and NO.
Cytochrome c, SOD, DPI, oxypurinol, indomethacin, L-NMMA, L-NAME, DAHP, and 6-methyl-5,6,7,8-tetrahydropterine were purchased from Sigma Chemical Co. Ionomycin was purchased from Calbiochem-Novabiochem Corp. Flexercell strain unit system and Flex I culture plates were purchased from Flexcell Co. EGM was purchased from Clonetics.
HAECs were prepared from human aorta by enzyme digestion.4 13 HAECs were identified by their typical morphological and chemical characteristics, including growth of closely opposed polygonal confluent monolayers, maintenance of density-dependent inhibition after serial passage, and positive immunofluorescence for factor VIII antigen. HAECs were cultured on collagen type I–coated dishes or collagen type I–coated plates for stretch experiments (Flex I) in EGM (Clonetech) and 10% fetal calf serum (Hyclone). Cells between passages 2 and 6 were used for the experiments.
Stretch Device for Cultured Cells
Pulsatile stretch was given to HAECs as previously reported.14 15 Briefly, HAECs were seeded onto Flex I culture plates coated with type I collagen at an initial density of 105 cells/mL. Flex I culture plates were placed on a computerized Flexercell Strain Unit gasketed base plate in the incubator. The membranes were subjected to a vacuum of −5 and −20 kPa, respectively, at a frequency of 50 cycles/min for up to 24 hours. A vacuum of −20 kPa causes a deformation pattern of the membrane ranging from 0% at the center to 24% at the periphery (10% average).14 15 16 A vacuum of −5 kPa evokes a membrane deformation ranging from 0% at the center and 10% at the periphery (6% average). In parallel, other Flex I culture plates not subjected to stretch served as controls. The cell viability with and without stretch, as assessed with trypan blue exclusion test, was >90% throughout the experiments.
Measurement of O2− Production
O2− production was measured as the SOD-inhibitable reduction of cytochrome c.11 17 Briefly, HAECs were preincubated in DMEM without phenolred for 30 minutes at 37°C, and then cytochrome c (final concentration, 1 mg/mL) with or without SOD (final concentration, 500 U/mL) was added in a CO2 incubator. At the indicated time points, the medium was removed from the cells, and the absorbance was read at 550 nm against a distilled water blank. Reduction of cytochrome c in the presence of SOD was subtracted from the values without SOD. The portion of O2−-specific reduction of cytochrome c was between 20% and 35% according to the experiments. The optical density difference between comparable wells with or without SOD was converted to equivalent O2− production by use of the molar extinction coefficient for cytochrome c [21.0×103 (mol/L)−1 · cm−1].17
Western Blot for NOS Type III
NOS type III (ie, endothelial NOS) protein was analyzed by Western blot using an anti-human endothelial NOS antibody (Transduction Laboratory). The antibody was used at 250× dilution. This antibody is specific for human NOS III and does not cross-react with NOS I and II.18 After stretch experiments, HAECs were detached by trypsin-EDTA, and the cell number was determined by Coulter counter (Coulter Electronics). Then, 100 μL of lysis solution containing 10% glycerol, 2.3% SDS, Tris-HCl, pH 6.8, 62.5 mmol/L, 0.01% Bromophenol blue, and 5% mercaptoethanol was added to 105 cells. The lysate was then heated at 95°C to 100°C at 5 minutes. Next, 30 μL of cell lysates containing 3.3×104 cells was subjected to 7.5% single percentages gel (Ready Gel, BIO RAD). The separated proteins were electrophoretically transferred to Immunobilon-P membranes and then incubated with anti-human NOS III antibody for 1 hour as previously described.18 The membranes were finally visualized by the ECL kit (Amersham), and densitometric analysis was done by NIH image 1.54.
Northern Blot for NOS Type III
Total RNA was isolated by Trizol reagent (GIBCO BRL) according to manufacturer’s instruction. Then, 20 μg of total RNA was subjected to electrophoresis on 1% formaldehydeagarose gels and transferred to a nylon membrane (Highland-N, Amersham International). Blots were hybridized in QuickHyb (Strategene) with 32P-labeled cDNA probes prepared by random prime labeling. A 5′-fragment (1.6-kb EcoRI/Bg lII) of human NOS (III) cDNA clone19 (kindly provided by Dr T. Michel) was used for a NOS (III) probe. Membranes were exposed to Kodak Bio Max x-ray film at −70°C for 24 hours. Quantification of autoradiographs was obtained using the image analysis program NIH Image.
Data are presented as mean±SEM. Multiple comparisons were evaluated by ANOVA followed by Fisher’s protected least significant difference test. Student’s paired or unpaired t tests were used for comparisons between two experiments. A value of P<.05 was considered significant.
Time Course of O2− production in the Presence and Absence of Pulsatile Stretch
HAECs were incubated with reaction solution for the indicated time periods in the absence or presence of a physiological level of pulsatile stretch (10% average elongation, 50 cycles/min). Under both conditions, O2− production was increased in a time-dependent manner up to 60 minutes (Fig 1⇓). O2− production stimulated by stretch was significantly increased compared with that without stretch after 40 minutes (P<.05, n=6). The curves began to level off after 80 minutes (data not shown). Accordingly, a 60-minute interval was chosen for subsequent experiments.
Possible Pathway of O2− Production Induced by Short-term Stretch
To clarify the mechanism of stretch-induced O2− production, we studied the effects of several enzyme inhibitors. DPI, an NADPH oxidase inhibitor,23 significantly reduced O2− production induced by short-term stretch (Fig 2⇓; P<.05, n=6). On the other hand, OXY, a xanthine oxidase inhibitor, and INDO, a cyclooxygenase inhibitor, showed no effect on O2− production induced by short-term stretch (Fig 2⇓).
Contribution of NOS in O2− Production Induced by Short-term Stretch
Because NOS can act as an important source of O2− production,12 we further investigated the contribution of NOS by using NOS inhibitors and agonists. L-NMMA occupies the substrate binding site of NOS and thereby inhibits NO but not O2− production by NOS.12 20 In contrast to L-NMMA, L-NAME decreases electron flux from NOS and hence inhibits both NO and O2− production derived from the enzyme.12 20 To clarify the contribution of NOS in stretch-induced O2− production, we compared the effects of L-NMMA and L-NAME. Because the incubation medium is based on DMEM and contains ≈400 μmol/L of l-arginine, we choose a rather high concentration of both inhibitors (1 mmol/L). As Fig 2⇑ shows, L-NMMA significantly increased O2− production by short-term stretch (P<.05, n=6), whereas L-NAME had no significant effect. IONO, an agonist that increases intracellular calcium in endothelial cells, also showed no additional effect on O2− production by short-term stretch (Fig 2⇑; P=NS).
Effect of Prolonged Pulsatile Stretch on O2− Production
To examine the effect of prolonged stretch on O2− production, the time course of O2− production after exposure to stretch was evaluated at the indicated time points. Up to 6 hours, pulsatile stretch increased O2− production. On the other hand, O2− production gradually decreased after 12 hours of stretch. At 24 hours, there was no significant change compared with control conditions (Fig 3⇓; n=6).
Possible Pathway of O2− Production Induced by Prolonged Stretch
To clarify the pathway of O2− production induced by prolonged stretch, we used the same inhibitors as mentioned above. DPI significantly inhibited O2− production induced by prolonged stretch, but OXY or INDO showed no significant effect (Fig 4⇓; n=6). On the other hand, L-NMMA and L-NAME strikingly increased O2− production induced by prolonged stretch (2.4- and 1.5-fold; P<.05, n=6). After 24 hours of stretch, IONO significantly reduced O2− production induced by prolonged stretch (P<.05).
Contribution of NADPH Oxidase and NOS in O2− Production Induced by Prolonged Stretch
In the presence of L-NMMA, we evaluated the net amount of O2− production, including that which is normally scavenged by NO. O2− production by NADPH oxidase was calculated as the difference of O2− production measured in the presence of L-NMMA minus that measured in the presence of L-NMMA+DPI. Superoxide production by NOS (III) was calculated as the difference between that measured in the presence of L-NMMA minus L-NMMA+L-NAME, because L-NAME in contrast to NMMA inhibits not only NO production but also O2− from NOS (III).12 As the Table⇓ shows, short-term stretch increased net O2− production via NADPH oxidase 3-fold and via NOS (III) 1.8-fold compared with the basal production under static conditions (P<.05, n=6). Surprisingly, prolonged stretch increased net O2− production 6.5-fold via NADPH oxidase and 3.6-fold from NOS (III) compared with basal production (P<.05, n=6).
Effect of Prolonged Stretch on NOS (III) Protein
Western blot with a specific antibody against human NOS (III) was performed to clarify whether prolonged stretch increased NOS protein in human aortic endothelia cells. Fig 5A⇓ shows that prolonged stretch increased NOS (III) protein in a stretch-dependent manner. As indicated in Fig 5B⇓, NOS (III) protein was also increased by prolonged stretch over 12 hours in a time-dependent manner. Moreover, we did Western blot using a specific antibody against human NOS (II) (Santa Cruz Biotechnology), but the all results were negative (data not shown).
Effect of Prolonged Stretch on NOS (III) mRNA
To clarify the effect of stretch on NOS (III) mRNA, Northern hybridization analysis was performed. Fig 6⇓ shows that pulsatile stretch (10% average, 50 cycles/min) increased mRNA of NOS (III) compared with static control. Densitometric analysis clarified that 6, 12, and 24 hours of stretch increased NOS (III) mRNA compared with GAPDH 2.8±0.2-, 2.5±0.1-, and 2.2±0.2-fold, respectively (n=4).
Contribution of BH4 in O2− Production From NOS (III)
To clarify the possible contribution of BH4 to the regulation of O2− production from NOS, we examined the effects of the inhibitor of GTP cyclohydrolase I (DAHP)21 22 23 and the lipid-soluble analog of BH4 (MEBH4).21 Without stretch, neither DAHP nor MEBH4 exhibited a significant effect. MEBH4 alone exhibited no effect (data not shown); DAHP, however, significantly increased O2− production after 24 hours of stretch (Fig 7⇓; P<.05). Moreover, this increase was significantly reduced by cotreatment with MEBH4 (P<.05 versus DAHP alone).
The present experiments demonstrate for the first time in HAECs that (1) short-term and prolonged pulsatile stretch increases O2− production via NADPH oxidase and NOS (III), (2) prolonged pulsatile stretch increases both O2− and NOS (III) expression, and (3) NOS (III) upregulation by prolonged pulsatile stretch acts as a scavenger mechanism for O2−. Moreover, in contrast to static conditions, reduction of BH4 levels enhanced O2− production after prolonged pulsatile stretch.
DPI is a flavoprotein inhibitor, and it is also known to be a potent inhibitor of NADPH oxidase24 and NOS.25 In our experiments, both short-term and prolonged stretch–induced O2− production was significantly reduced by DPI. In bovine coronary endothelium26 and rat cultured aortic smooth muscle cell,27 NADH-dependent oxidase was reported to be a primary source of O2−. On the other hand and in line with our observations, NADPH oxidase was reported to be a major O2−-generating system in rabbit aorta.28 This suggest that O2−-generating systems differ among species and different kind of cells.
NOS is now considered to play a pivotal role not only in the production of NO but also as a source of O2−. Indeed, O2− production from purified NOS (I) has been reported.12 NO derived from NOS (III), on the other hand, can act as a scavenger that is able to capture O2− (6.7×109 mol/s) faster than SOD.29 Recently, NOS (III) was reported to be increased by stretch in bovine aortic endothelial cells.7 Our hypothesis, therefore, was that if NOS (III) can be upregulated by stretch in HAECs similar to bovine cells, NOS (III) upregulation may play a pivotal role in determining the balance between O2− and NO production by the endothelium.
Of particular interest was the time course of stretch-induced O2− production. Indeed, in the absence of L-NMMA, prolonged stretch exhibited weaker effects on O2− production than short-term stretch. Stretch has been reported to activate protein kinase C,30 and protein kinase C is known to activate NADPH oxidase to produce O2−.31 Therefore, it may be possible that prolonged stretch downregulates protein kinase C, and this in turn might reduce O2− production, but we considered another possibility. In this article, O2− production was measured with cytochrome c reduction.17 It would also be possible to evaluate O2− production with NBT reduction by histochemistry, but we choose cytochrome c reduction to obtain quantitative information. This method is quite specific for the O2− anion,32 but it detects only SOD-inhibitable or active O2−. Obviously, therefore, O2− that has been inactivated by NO is not capable of reducing of cytochrome c and hence is not detected by this assay. This, in turn, means that to obtain a measure of true O2− production, measurements must be performed in the presence of L-NMMA to inhibit NO. Our results using NMMA clearly showed that prolonged stretch increased net O2− production from NADPH oxidase by 6.5-fold and from NOS (III) by 3.6-fold. Moreover, prolonged stretch increased NOS (III) protein and NO production in HAECs. Thus, the biphasic time-production course of O2− production can be explained by the increased NO production with prolonged stretch. These results therefore suggest that the upregulation of NOS (III) by prolonged stretch plays a physiological role as a scavenger mechanism of O2−.
To get further insights into the regulation of the balance of O2− and NO production from NOS (III), we examined the role of BH4, an essential cofactor for NOS to produce NO.22 33 34 35 36 37 BH4 is produced via GTP cyclohydrolase I in endothelial cells,38 and we used DAHP as an inhibitor of this enzyme.38 39 In unstretched HAECs, DAHP had no effect on O2− production. However, in stretched (24 hours) HAECs, O2− production was markedly increased by DAHP. This increase in O2− production induced by DAHP in stretched HAECs could be partially reversed by 6-methyltetrahydropterine, a lipid-soluble analog of BH4.23 Hence, with prolonged stretch, the requirements of HAECs for BH4 are increased, and depletion of BH4 leads to an increased O2− rather than NO production by NOS (III). The increased requirement for BH4 under conditions of prolonged stretch can be explained by the increase in NOS (III) protein by stretch.
Upregulation of NOS (I) and (III) was reported by shear stress,40 estrogen,18 41 42 and stretch.7 This increase in NOS, at a glance, seems to play vasculoprotective and compensatory roles. Upregulated NOS may produce larger amounts of NO, but it may also produce more O2− to scavenge NO. Although 6-methyltetrahydropterine alone showed no significant effect on O2− production from static and prolonged stretched HAECs without DAHP treatment (data not shown), the reduction in O2− production by this analog of BH4 during DAHP treatment suggest a therapeutic potential of such compounds under the condition of BH4 depletion or increased demands for the cofactor.
In conclusion, our results demonstrate the importance of mechanical forces as determinants of the balance of O2− and NO in human endothelial cells. These mechanisms are likely to play a key role in the pathophysiology of hypertensive vascular disease43 and atherosclerosis,32 disease states in which both NO and O2− production in endothelial cells are upregulated.
Selected Abbreviations and Acronyms
|DMEM||=||Dulbecco’s modified Eagle’s medium|
|HAEC||=||human aortic endothelial cell|
|L-NAME||=||NG-nitro-l-arginine methyl ester|
|NOS||=||nitric oxide synthase|
This work was supported by the Swiss National Research Foundation (grant 32-32541.91/2) and a grant-in aid by Mobiliar Insurance, Bern, Switzerland, and the Karl Mayer Foundation, Liechtenstein.
- Received May 6, 1997.
- Revision received June 17, 1997.
- Accepted June 26, 1997.
- Copyright © 1997 by American Heart Association
Hishikawa K, Nakaki T, Marumo T, Hayashi M, Suzuki H, Kato R, Saruta T. Pressure promotes DNA synthesis in rat cultured vascular smooth muscle cells. J Clin Invest. 1994;93:1975-1980.
Pohl U, Holtz J, Busse R, Bassenge E. Crucial role of endothelium in the vasodilator response to increased flow in vivo. Hypertension. 1986;8:37-44.
Hishikawa K, Nakaki T, Marumo T, Suzuki H, Kato R, Saruta T. Pressure enhances endothelin-1 release from cultured human endothelial cells. Hypertension. 1995;25:449-452.
Awolesi MA, Sessa WC, Sumpio BE. Cyclic strain upregulates nitric oxide synthase in cultured bovine aortic endothelial cells. J Clin Invest. 1995;96:1449-1454.
Rosen GM, Freeman BA. Detection of superoxide generated by endothelial cells. Proc Natl Acad Sci U S A. 1984;81:7269-7273.
Matsubara T, Ziff M. Increased superoxide anion release from human endothelial cells in response to cytokines. J Immunol. 1986;137:3295-3298.
Pou S, Pou WS, Bread DS, Snyder SH, Rosen GM. Generation of superoxide by purified brain nitric oxide synthase. J Biol Chem. 1992;267:24173-24176.
Jaffe EA, Nachman R, Becker C, Minik D. Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest. 1973;52:2754-2758.
Yang Z, Noll G, Lüscher TF. Calcium antagonists differently inhibit proliferation of human coronary smooth muscle cells in response to pulsatile stretch and platelet-derived growth factor. Circulation. 1993;88:832-836.
Sampio BE. Mechanical stress and cell growth. J Vasc Surg. 1989;10:570-571.
Abusoud HM, Feldman PL, Clark P, Stuehr DJ. Electron transfer in the nitric-oxide synthases: characterization of l-arginine analogs that block heme iron reduction. J Biol Chem. 1994;269:32318-32326.
Cosentino F, Katusic ZS. Tetrahydrobiopterine and dysfunction of endothelial nitric oxide synthase in coronary arteries. Circulation. 1995;91:139-144.
Stuehr DJ, Fasehun OA, Kwon NS, Gross SS, Gonzalez JA, Levi R, Nathan CF. Inhibition of macrophage and endothelial cell nitric oxide synthase by diphenyleneiodonium and its analogs. FASEB J. 1991;5:98-103.
Griendling KK, Minieri CA, Ollerenshaw D, Alexander RW. Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells. Cir Res. 1994;74:1141-1148.
Mohazzab-H KM, Kaminski PM, Wolin MS. NADH oxidoreductase is a major source of superoxide anion in bovine coronary artery endothelium. Am J Physiol. 1994;266:H2568-H2572.
Pagano PJ, Ito Y, Tornheim K, Gallop PM, Tauber AI, Cohen RA. An NADPH oxidase superoxide-generating system in rabbit aorta. Am J Physiol. 1995;268:H2274-H2280.
Watson F, Robinson J, Edwards SW. Protein kinase C-dependent and -independent activation of the NADPH oxidase of human neutrophils. J Biol Chem. 1991;266:7432-7439.
Ohara Y, Peterson TE, Harrison DG. Hypercholesterolemia increases endothelial superoxide anion production. J Clin Invest. 1993;91:2546-2551.
Tayeh MA, Marletta MA. Macrophage oxidation of l-arginine to nitric oxide, nitrite, and nitrate: tetrahydrobiopterin is required as a cofactor. J Biol Chem. 1989;264:19654-19658.
Kwon NS, Nathan CF, Stuehr DJ. Reduced biopterin as a cofactor in the generation of nitrogen oxides by murine macrophages. J Biol Chem. 1989;264:20496-20501.
Giovnelli J, Campos KL, Kaufman S. Tetrahydrobiopterine, a cofactor for rat cerebellar nitric oxide synthase, does not function as a reactant in the oxygeneration of arginine. Proc Natl Acad Sci U S A. 1991;88:7091-7095.
Baek KJ, Thiel BA, Lucas S, Stuehr DJ. Macrophage nitric oxide synthase subunits: purification, characterization, and role of prosthetic groups and substrate in regulating their association into a dimeric enzyme. J Biol Chem. 1993;268:21120-21129.
Gross SS, Levi R. Tetrahydrobiopterin synthesis: an absolute requirement for cytokine-induced nitric oxide generation by vascular smooth muscle. J Biol Chem. 1992;267:25722-25729.
Wernerfelmayer G, Werner ER, Fuchs D, Hausen A, Reibnegger G, Schmidt K, Weiss G, Wachter H. Pteridine biosynthesis in human endothelial cells: impact on nitric oxide-mediated formation of cyclic GMP. J Biol Chem. 1993;268:1842-1846.
Scottburden T, Elizondo E, Ge T, Boulanger CM, Vanhoutte PM. Simultaneous activation of adenylyl cyclase and protein kinase C induces production of nitric oxide by vascular smooth muscle cells. Mol Pharmacol. 1994;46:274-282.
Nishida K, Harrison DG, Navas JP, Fisher AA, Dockery SP, Uematsu M, Nerem RM, Alexander RW, Murphy TJ. Molecular cloning and characterization of the constitutive bovine aortic endothelial cell nitric oxide synthase. J Clin Invest. 1992;90:2092-2096.
Tschudi MR, Meason S, Lüscher TF, Malinski T. Direct in situ measurement of nitric oxide in mesentric resistance arteries: Increased decomposition by superoxide in hypertension. Hypertension. 1996;27:32-35.