Intracellular pH and Tyrosine Phosphorylation but Not Calcium Determine Shear Stress–Induced Nitric Oxide Production in Native Endothelial Cells
Abstract Signaling pathways determining the shear stress–induced production of NO from endothelial cells in situ were investigated using a bioassay system in which shear stress was increased by inducing vasoconstriction in an endothelium-intact donor segment (rabbit iliac artery) while maintaining a constant luminal perfusion rate. Shear stress–induced NO production, as assessed by changes in the tone of a preconstricted endothelium-denuded detector ring, was biphasic and consisted of an initial transient (20- to 25-minute) Ca2+-dependent phase followed by a Ca2+-independent plateau phase, which was maintained as long as the donor segment remained constricted. Stretching the donor segments to their in vivo length abolished the initial phase without affecting the plateau phase of NO release. Inhibition of the Na+-H+ exchanger using HOE 694 elicited an intracellular acidification, which attenuated shear stress–induced NO production. The specific protein kinase C inhibitor, Ro 31-8220, was without effect, whereas the unspecific inhibitors, staurosporine and calphostin C, abolished the shear stress–induced production of NO. Erbstatin A, a tyrosine kinase inhibitor, attenuated the shear stress–induced tyrosine phosphorylation of specific cellular proteins and abrogated the associated NO production. In summary, these data indicate that shear stress activates the NO synthase at basal levels of [Ca2+]i via a mechanotransduction cascade that involves tyrosine phosphorylation and can be modulated by changes in pHi. The apparent fundamental alteration of the endothelial NO synthase under shear stress that renders its maintained activation independent of an increase in [Ca2+]i is probably the consequence of a change in the enzyme microenvironment.
The shear stress (or viscous drag) exerted upon the luminal surface of the endothelium by the streaming blood is considered to be the physiologically most important stimulus for the release of NO from endothelial cells (for review see Reference 11 ). The level of shear stress and the release of NO elicited by altering either diameter or flow are positively correlated, suggesting that local changes in tone (eg, neurogenic or myogenic constriction) are as important as changes in flow for the regulation of endothelial NO release in vivo.2 3 4 The main physiological consequence of this relationship is that any decrease in vessel diameter, at constant flow, increases the shear stress to which the endothelial cell layer is exposed and elicits the release of NO, which in turn feeds back to inhibit the original vasoconstriction.
The mechanism by which the endothelium is able to sense changes in shear stress on its luminal surface remains obscure, although there have been reports that perturbation/disruption of the endothelial cytoskeleton may be responsible for initiating some of the changes associated with exposure to shear stress. For example, exposure to elevated shear stress prompts reorganization of F-actin microfilaments in vivo5 and in vitro6 7 and alters the topography of endothelial cells such that they become streamlined in the direction of flow.8 Other shear-induced effects, such as the increase in endothelin-1 mRNA levels, can be mimicked by the actin-disrupting agents cytochalasin B and D and inhibited by stabilization of the cytoskeleton using colchicine.9 Such findings imply that shear stress–induced cytoskeletal disturbances and/or rearrangement form an essential and integrative part in endothelial mechanotransduction (for review see Reference 1010 ). The intracellular signal transduction pathway that is initiated by increases in shear stress has been reported to involve activation of PLC,11 and a rapid increase in intracellular levels of inositol 1,4,5-tris-phosphate12 13 enhanced the release of NO14 15 and increased cellular levels of cGMP.16 Induction of some early-response genes can be detected shortly after application of shear stress (c-myc after several minutes, c-fos and c-jun within 2 hours)17 as well as activation of the transcription factors AP-1 and NFκB.18
The ability of shear stress to enhance the activity of the NO synthase in endothelial cells may be the consequence of an increase in [Ca2+]i (see references 19-23), although other possibilities include an indirect action involving one or more of the cytoskeletal proteins (likely candidates being the annexins and integrins) or a direct effect on the membrane-bound endothelial NO synthase itself.
The aim of the present study was to identify the determinants of the signaling pathway involved in shear stress–induced NO release from endothelial cells in situ. Since under a laminar steady state flow, shear stress is inversely related to the third power of the vessel radius but only proportional to changes in flow (Hagen-Poiseuille’s law), it follows that relatively small changes in vessel diameter at constant flow can markedly increase shear stress at the endothelial surface. Therefore, we used a bioassay system in which shear stress was increased by inducing vasoconstriction of an endothelium-intact donor segment while maintaining a constant luminal perfusion rate. The shear stress–induced release of NO was assessed by monitoring the change in tone of a preconstricted endothelium-denuded detector ring.
Materials and Methods
Diclofenac (Voltaren) was obtained from CIBA-GEIGY; L-NNA (free acid) and HEPES were from Serva; ACh, PE, and atropine were from Sigma; staurosporine was from Boehringer; ionomycin, fura 2-AM, BCECF, erbstatin A, genistein, and calphostin C were from Calbiochem; and endothelin-1 was from Bachem. M-119 medium was from GIBCO BRL, and penicillin, streptomycin, l-glutamine, glutathione, and l(+)-ascorbic acid (Biotect protection medium) were from Biochrom. Glyceryl trinitrate was kindly provided by Pohl-Boskamp; U46619, by UpJohn; bovine recombinant superoxide dismutase (Peroxinorm), by Grünenthal; HOE 694, by Hoechst; and Ro 31-8220, by Roche Products Ltd.
Bioassay System for the Detection of Luminal NO Release
New Zealand White rabbits of either sex (2.0 to 2.5 kg body weight) were anesthetized with sodium pentobarbital (Nembutal, 60 mg/kg IV, Sanofi) and exsanguinated by cutting through both the aorta and vena cava. The abdominal aorta (detector) and iliac arteries (donor) were dissected and cleaned of adventitial adipose and connective tissue. A 4-mm-wide ring was cut from the middle of the aorta, denuded of its endothelium, and mounted between a GM2/GM3 force transducer (Scaime) and a rigid support for measurement of isometric force. The ring was superfused at a flow rate of 1.0 mL/min with warmed (37°C), oxygenated (95% O2/5% CO2) Krebs-Henseleit solution of the following composition (mmol/L): NaCl 119, NaHCO3 25, CaCl2 1.6, KCl 4.7, KH2PO4 1.2, MgSO4 1.2, glucose 11.1, and EDTA 0.026 (pH 7.4), along with 1 μmol/L diclofenac and 10 nmol/L superoxide dismutase. Passive tension was adjusted over a 30-minute equilibration period to 2 g; thereafter, the ring was submaximally constricted with 1 μmol/L PE, in the presence of atropine (0.1 μmol/L), to a tension of ≈8 g, and the absence of endothelium was confirmed by the lack of response to ACh (10 nmol). A 12-mm-long segment of the iliac artery (without visible branches) was cannulated at both ends and mounted in a thermostatted (37°C) organ chamber and had a resting outer diameter of ≈1.8 mm. The lumen of the segment was perfused with Krebs-Henseleit solution at a flow rate of 1.0 mL/min; the adventitial side, at a flow rate of 0.2 mL/min. After an equilibration period of 30 minutes, the system was oriented so that the luminal effluate from the endothelium-intact donor segment directly superfused the endothelium-denuded detector ring (delay, ≈2 seconds). The diameter of the donor segment was monitored using a micrometer scale (M3 microscope, Wild). NO release from the endothelium of the donor segment was elicited either by bolus application of ACh (10 nmol into the luminal perfusate) or by vasoconstriction (10 μmol/L PE). NO-mediated relaxant responses of the detector ring to the effluate from the donor segment were standardized by comparison with the response observed after the application of GTN (10 and 100 pmol) directly to the detector tissue. In the experiments in which the effects of the PKC inhibitors were studied, each segment served as its own control. Initially, all of the segments were subjected to shear stress in the presence of solvent. After the removal of shear stress and the return of the donor segment to resting diameter, the same segments were incubated with PKC inhibitors and subsequently subjected to shear stress.
Human umbilical vein endothelial cells or porcine aortic endothelial cells, isolated as previously described,24 25 were seeded either on quartz coverslips or in culture dishes (35 mm, Falcon) containing M-119 medium and 20% heat-inactivated FCS (Vitromex) supplemented with penicillin (50 U/mL), streptomycin (50 μg/mL), l-glutamine (1 mmol/L), glutathione (5 mg/mL), and l(+)-ascorbic acid (5 mg/mL). [Ca2+]i and pHi were estimated in cells grown on coverslips for 24 hours.
Measurement of pHi and [Ca2+]i
Endothelial cells were loaded with the pH-sensitive indicator BCECF by incubation with 3 μmol/L of its pentaacetoxymethyl ester (BCECF-AM) in M-119 medium for 40 minutes at 37°C. Cells were then washed twice, mounted in a parallel plate flow chamber superfused with Tyrode’s solution containing (mmol/L) NaCl 132, KCl 4, CaCl2 1.6, MgCl2 0.98, NaHCO3 11.9, NaH2PO4 0.36, and glucose 10, and placed on the stage of an inverted microscope (Diaphot-TMB, Nikon). pHi was determined fluorometrically using continuous rapid alternating excitation from dual monochromators set at 490 and 439 nm, respectively (Deltascan, Photon Technology). Incident light passed through a filter block (DM510, Nikon) containing a dichroic mirror (520 to 560 nm) and was focused onto the sample by means of a ×40 objective (Fluor 40, Nikon). Emitted light was collected by the objective, and fluorescence was detected by a photon-counting photomultiplier (D-104, Photon Technology) as described previously.26
For the measurement of [Ca2+]i, endothelial cells were loaded with the fluorescent Ca2+-sensitive dye fura 2 by incubation with 3 μmol/L fura 2-AM and 0.025% (wt/vol) Pluronic F-127 at 37°C for 90 minutes. Thereafter, the coverslips were washed in Tyrode’s solution, and [Ca2+]i was determined fluorometrically as described previously.26
Confluent primary cultures of porcine aortic endothelial cells were washed twice in M-119 medium containing 0.1% FCS and were exposed to a calculated shear stress of ≈15 dyne/cm2 in a cone-plate viscosimeter as described in “Results.” Thereafter, cells were washed with ice-cold HEPES buffer containing NaF (10 mmol/L), Na4P2O7 (15 mmol/L), Na3VO4 (2 mmol/L), leupeptin (2 μg/mL), pepstatin A (2 μg/mL), trypsin inhibitor (10 μg/mL), and phenylmethylsulfonyl fluoride (44 μg/mL) and harvested by scraping. The cell suspension was centrifuged at 13 000g for 60 seconds; cells contained in the pellet were then lysed in buffer containing 1% (vol/vol) Triton X-100, left on ice for 5 minutes, and centrifuged at 10 000g for 10 minutes. Approximately 30 μg protein from the resulting supernatant or from the Triton X-100–insoluble fraction was separated by 10% or 7% SDS-PAGE, respectively, as described previously.27 Tyrosine-phosphorylated proteins were detected with a mouse monoclonal anti-phosphotyrosine antibody (1 μg/mL) (Upstate Biotechnology Inc) and were visualized by enhanced chemiluminescence using a commercially available kit (Amersham). Prestained molecular weight marker proteins (BioRad) were used as standards for the SDS-PAGE.
Unless otherwise indicated, data are expressed as mean±SEM. Statistical evaluation was performed using the two-tailed Student’s t test for paired data. Values of P<.05 were considered statistically significant.
Shear Stress–Mediated NO Release
In the bioassay system, bolus administration of ACh (10 nmol) to the endothelium-intact donor segment elicited a prominent but transient relaxation of the endothelium-denuded detector ring (Fig 1⇓). Constriction of the donor segment at a constant perfusion rate increased fluid shear stress at the endothelial surface from 0.1±0.02 to 2.3±0.04 dyne/cm2 (calculated values) and evoked a pronounced vasodilatation in the detector ring. This response consisted of an initial transient peak followed by a stable plateau phase, which was maintained as long as the donor remained constricted (Fig 1⇓). Identical biphasic responses were observed in experiments in which donor segments were contracted with either the thromboxane mimetic U46619 or endothelin-1 (data not shown). The minimum decrease in the external diameter of the donor segment, which was associated with a shear stress–induced release of NO, was ≈0.3 mm (ie, a decrease in the resting external diameter of 18±2%). Once this threshold had been reached, a further 0.1-mm decrease in diameter of the donor segment was associated with a reduction in tension of the detector ring by ≈0.5 g, suggesting a graded release of NO in response to shear stress.
In experiments in which either the endothelium was removed from the donor segment (data not shown) or the segment was pretreated with the NO synthase inhibitor L-NNA (0.1 mmol/L, Fig 1⇑), neither ACh nor enhanced shear stress was able to elicit relaxation of the detector ring. However, the response observed after direct application of GTN (100 pmol) to the detector segment was unaffected by either of these treatments. These observations demonstrate that the relaxation of the detector is mediated exclusively by NO released from the endothelium of the donor segment.
Effects of Longitudinal Stretch and Extracellular Ca2+ Removal
Longitudinal stretching of the donor segment by 25% of its initial length ex vivo, thus reestablishing approximately the tethered length in vivo, decreased the external diameter by 10±2% but had no effect on the kinetics of the PE-induced contraction or on the release of NO elicited by ACh. However, this procedure completely abrogated the transient peak release of NO after vasoconstriction without affecting the plateau phase (Fig 2⇓).
Replacing the solution perfusing the donor segment with a nominally Ca2+-free solution containing 0.1 mmol/L EGTA (ie, lowering [Ca2+ ]o from 1.6 mmol/L to ≈10 nmol/L) completely abolished the ACh-induced release of NO and the peak component of NO release in response to vasoconstriction in the unstretched segment (Fig 3⇓). However, the plateau phase of NO release was unaffected in either the unstretched or the stretched segments (Fig 3⇓).
Subsequent experiments were performed in both stretched and unstretched segments in order to compare the effects of various inhibitors on Ca2+-dependent and -independent signaling processes. However, since the plateau phase of shear stress–induced NO release was affected identically by pharmacological intervention in both experimental conditions, only data obtained in unstretched donor segments have been presented.
Effect of Na+-H+ Exchange Inhibition on pHi and Shear Stress–Stimulated NO Release
In order to investigate the effect of changes in pHi on the shear stress–stimulated release of NO, experiments were performed using the specific Na+-H+ exchange inhibitor, HOE 694. The effect of this inhibitor on pHi, under conditions of continuous flow and in the presence of bicarbonate, was initially investigated in cultured porcine endothelial cells loaded with the pH-sensitive indicator BCECF. In these cells, HOE 694 (10 μmol/L) induced a rapid and prolonged intracellular acidification (Fig 4⇓), which was reversed upon washout of the inhibitor. Similar experiments in fura 2–loaded endothelial cells demonstrated that this acidification was not associated with changes in [Ca2+]i (Fig 4⇓). The effect of HOE 694 on pHi was not influenced by removal of extracellular Ca2+ (data not shown).
In the bioassay system, HOE 694 (10 μmol/L), when added to the luminal perfusate of the donor segment after establishment of the plateau phase, induced a rapid decrease in its amplitude (68.3±5.0%, n=7, P<.001). This effect was immediately reversed upon removal of the inhibitor from the luminal perfusate (Fig 5⇓). Superfusion of the detector tissue with Krebs-Henseleit solution containing HOE 694 (10 μmol/L) or treatment of the donor segment with the solvent alone was without effect. HOE 694 also failed to alter the response of the detector to bolus administration of GTN (the GTN-induced decrease in detector tone was 47.6±6.7% in control vessels compared with 45.7±4.6% in HOE 694–treated vessels; n=4). The inhibition of the shear stress–elicited release of NO by HOE 694 was also observed in nominally Ca2+-free conditions (Fig 5B⇓).
Effects of PKC Inhibitors
Pretreatment of the donor segment with the PKC inhibitor staurosporine (30 nmol/L, 60 minutes) followed by a 30-minute washout period had no effect on the ACh-induced release of NO but selectively attenuated the plateau phase of the NO released in response to vasoconstriction (Fig 6⇓). A second PKC inhibitor, calphostin C (30 nmol/L, 60 minutes; Fig 6⇓), also failed to affect the release of NO elicited by ACh but had the same effect as staurosporine on the release of NO elicited by shear stress. Neither staurosporine nor calphostin C influenced the PE-induced vasoconstriction of the donor segment. Pretreatment with the reportedly more selective PKC inhibitor, Ro 31-8220 (30 nmol/L, 60 minutes),28 enhanced the production of NO in response to bolus application of ACh (69.9±10.2% enhancement, n=4, P<.001) but failed to affect the shear stress–dependent NO release (Fig 6⇓).
Effects of Tyrosine Kinase Inhibitors on Shear Stress–Mediated NO Release
The tyrosine kinase inhibitor erbstatin A (3 μmol/L, 5 to 10 minutes) attenuated the ACh-induced release of NO by 91.4±5.3% (n=4, P<.001) and markedly inhibited both phases of the shear stress–induced NO release from the donor segment (Fig 7⇓). Removal of erbstatin A from the luminal perfusate (60 minutes) was associated with total recovery of the ACh response (Fig 7⇓, left), whereas both the peak and plateau phases of the shear stress–induced release of NO remained depressed (Fig 7⇓, middle and right).
Because of a direct attenuating effect on the vascular tone of the detector ring, the use of a second tyrosine kinase inhibitor, genistein (0.1 mmol/L), proved to be impracticable.
Effects of Erbstatin A on Shear Stress–Induced Tyrosine Phosphorylation
Exposure of cultured human endothelial cells to a calculated shear stress level of 15 dyne/cm2 in a cone-plate viscosimeter for 30 or 60 minutes enhanced the tyrosine phosphorylation of a series of proteins in both the Triton X-100–soluble and –insoluble (cytoskeletal) cell fractions. In the Triton-soluble fraction, shear stress was associated with enhanced tyrosine phosphorylation of a 42- and 44-kD protein doublet (Fig 8A⇓). When specific antibodies were used, these two proteins were identified as the 42- and 44-kD isoforms of the MAP kinase (data not shown). The shear stress–induced tyrosine phosphorylation of this protein doublet was not observed in cells pretreated with the tyrosine kinase inhibitor erbstatin A (30 μmol/L, Fig 8B⇓). Similarly, in the cytoskeletal fraction, the application of shear stress enhanced the tyrosine phosphorylation of at least four proteins corresponding to molecular masses of 88, 90, 103, and ≈114 kD. The tyrosine phosphorylation of these proteins was just apparent after 10 minutes, was markedly increased over 30 to 60 minutes, and was abrogated in erbstatin A–pretreated cells (Fig 8C⇓ and 8D⇓). Similar increases in tyrosine phosphorylation were also observed using immunofluorescence (I. Fleming, unpublished data, 1995).
In the present study, the putative signaling pathways involved in shear stress–induced NO release from endothelial cells in situ were investigated using a bioassay system in which shear stress was increased by inducing vasoconstriction while maintaining a constant luminal perfusion rate. Under these experimental conditions, the PE-induced decrease in the diameter of an endothelium-intact donor segment was associated with a biphasic relaxation of an endothelium-denuded detector ring. This relaxation was not observed in the presence of the NO synthase inhibitor L-NNA or after removal of endothelium from the donor, thus demonstrating that relaxation of the detector segment was solely mediated by the release of NO from the donor. Additional evidence that shear stress elicits NO release from the donor segment was obtained using electron spin-resonance spectroscopy (A. Mülsch, unpublished data, 1995).
The two distinct phases of shear stress–induced increase in NO production were differentially sensitive to extracellular Ca2+ in that the initial peak was abolished by the removal of Ca2+ from the luminal perfusate while the plateau phase remained unaffected. The involvement of extracellular Ca2+ in the signal transduction pathway that translates shear stress into an immediate increase in NO synthase activity has for some time been controversial.19 20 21 22 23 29 30 A biphasic increase in NO release, with kinetics similar to that observed in the present bioassay experiments, has been reported after the exposure of cultured endothelial cells to laminar flow.31 In the latter study, the initial peak production of NO was sensitive to the chelation of extracellular Ca2+ and the calmodulin antagonist calmidazolium, whereas the sustained phase of shear stress–induced NO release was independent of both interventions. Our finding that shear stress enhances NO synthase activity via Ca2+-dependent and Ca2+-independent mechanisms confirms and extends previous observations and may resolve the present controversy concerning the role of extracellular Ca2+ in shear stress–induced NO release. Since in the bioassay system the initial Ca2+-dependent phase of the shear stress–induced NO release was not observed in segments stretched to their in vivo lengths, it is likely that this phase reflects artificial in vitro conditions rather than a real physiological response. Endothelial cells, either in situ or in culture, after exposure to defined shear stress for a longer period of time adopt a characteristic spindle-shaped morphology and align in the direction of flow.5 6 7 32 Static culture conditions or the failure to stretch the excised vascular segments are associated with marked changes in endothelial cell morphology, such that they assume a more polygonal form. Since fluid shear stress on the endothelial surface is thought to be transduced through the cytoskeleton, it is more than likely that even minor changes in cell morphology have marked effects on mechanotransduction pathways. Thus, the shear stress–induced Ca2+ signal and the Ca2+ dependence of shear stress–induced NO release observed in unstretched endothelial cells may be directly related to such a subtle alteration in the cytoskeleton. This simple consideration might have fundamental implications with regard to the relevance of widely used experimental models.
The rapid and reversible effect of the Na+-H+ exchange inhibitor, HOE 694, on shear stress–induced NO production can be directly attributed to intracellular acidification, since this compound failed to affect [Ca2+]i. Moreover, it has been previously demonstrated that changes in pHi are able to modulate endothelial NO synthase activity at basal levels of [Ca2+]i.26 The inhibitory effect of HOE 694 on NO release also suggests that under conditions of continuous shear stress, the Na+-H+ exchanger is active in native endothelial cells. Although exposing cultured endothelial cells to an increase in flow has been reported to activate the Na+-independent Cl−-HCO3− exchanger, no evidence has been provided to suggest activation of the Na+-H+ exchanger in the presence of bicarbonate. In cultured endothelial cells, it seems that moderate increases in shear stress (0.5 to 2.0 dyne/cm2) induce a biphasic change in pHi consisting of a transient initial acidification followed by either a return to baseline values33 or a prolonged alkalinization above resting values (I. Fleming, unpublished data, 1995), whereas higher shear levels (>12 dyne/cm2) are associated exclusively with intracellular acidification.33 However, since intracellular alkalinization at moderate shear stress levels, such as those attained in the bioassay system, would tend to increase activity of the endothelial NO synthase,26 it is more than likely that changes in pHi contribute to the sustained NO production in response to shear stress.
Activation of PLC has been reported to be involved in the acute response of endothelial cells to shear stress; thus, it is plausible that the subsequent increase in diacylglycerol leads to the activation of PKC. However, little direct evidence exists to substantiate such a hypothesis or provide a link between PKC activation and shear stress–dependent NO formation. In the present study, the relatively unspecific PKC inhibitors, staurosporine and calphostin C, selectively impaired the plateau phase of the shear stress–dependent NO release without affecting either the initial peak or that elicited by bolus application of ACh. The reportedly more specific PKC inhibitor, Ro 31-8220,28 on the other hand, significantly prolonged ACh-induced NO release but failed to affect the sustained shear stress–induced NO release. These observations indicate that the effects of staurosporine and calphostin C are likely to be unrelated to the inhibition of PKC. One possible explanation for this unexpected finding could be related to staurosporine-mediated alterations in the actin cytoskeleton. Indeed, staurosporine and another unspecific PKC inhibitor, H-7, have been reported to elicit rapid changes in the organization of the actin cytoskeleton in rat astrocytes, a phenomenon not observed after application of Ro 31-8220.34 In the latter study, staurosporine was shown to decrease the thickness and linear appearance of actin microfilament bundles (stress fibers), attenuate phosphorylation, and alter the distribution of myosin light chain throughout the cytoplasm. These effects of staurosporine were unaffected by downregulation of PKC with phorbol 12-myristate 13-acetate and could be partially reproduced by inhibition of myosin light chain kinase.34 Therefore, it is possible that disruption of the actin cytoskeleton in endothelial cells may account for the effects of staurosporine and calphostin C on shear stress–dependent NO release observed in the present study. However, we observed no effect of cytoskeleton-disrupting agents such as cytochalasin B, taxol, and nocodazole on the plateau phase of shear stress–induced NO release (data not shown).
Over the last few years, evidence has accumulated to suggest that nonreceptor tyrosine kinases play a central role in endothelial cell signaling and may functionally link rearrangement of the cytoskeleton or focal adhesion points with more classical signal transduction pathways. Recent reports that an increase in tyrosine phosphorylation and activity of the focal adhesion kinase (pp125FAK) is associated with integrin clustering and integrin-mediated cell adhesion35 would also tend to support such a hypothesis. Results obtained in the present study also support a role for tyrosine kinases in mediating shear stress–induced NO production. Indeed, the tyrosine kinase inhibitor erbstatin A completely abolished the ACh-induced release of NO as well as both phases of the shear stress–dependent NO production. Although the effects of erbstatin A on both the ACh response and the initial phase of shear stress–induced NO production can be attributed to an attenuated influx of extracellular Ca2+,27 36 such a phenomenon cannot account for attenuation of the Ca2+-independent plateau phase of shear stress–induced NO release. Moreover, it is noteworthy that the effects of tyrosine kinase inhibition on the ACh-induced NO production and on the initial phase of the response to shear stress tended to be restored after removal of the inhibitor, whereas the plateau phase remained suppressed. Therefore, the inhibitory effect of erbstatin A may be related to the inhibition of “shear stress–activated” tyrosine kinases, which are distinct from those regulating Ca2+ influx. A more direct hint that shear stress stimulates the tyrosine phosphorylation of specific cellular proteins was provided by immunoblot analysis. A 42/44-kD doublet in the Triton-soluble cell fraction exhibited a rapid and marked increase in tyrosine phosphorylation after the application of shear stress. These proteins were identified as isoforms of the MAP kinase, which is known to be involved in the signal transduction pathway activated by growth factors and following the occupation of G protein–coupled receptors. It would also appear that MAP kinase activation may be differentially sensitive to Ca2+, since cell stimulation with agonists such as bradykinin require the presence of extracellular Ca2+,27 whereas mechanical stimuli have been reported to initiate MAP kinase activation in the absence of Ca2+.37 Moreover, the activation of the Ras-GTP/Raf/MAP kinase pathway in shear stress–activated endothelial cells may be a means of linking mechanical signals to alterations in gene transcription. Although the cytoskeletal proteins that were tyrosine-phosphorylated in response to shear stress in the present study have not yet been identified, a clear erbstatin A–sensitive increase in tyrosine phosphorylation of Triton-soluble and -insoluble proteins was observed.
In summary, the results of the present study demonstrate that shear stress induces NO production by a mechanism independent of extracellular Ca2+. This shear stress–induced NO release could be attenuated by decreasing pHi as well as by pretreatment with the unspecific PKC inhibitors, staurosporine and calphostin C. Data obtained using the tyrosine kinase inhibitor erbstatin A suggest that shear stress results in the activation of one or more tyrosine kinases and the tyrosine phosphorylation of a series of proteins. Moreover, tyrosine kinase inhibition abrogated shear stress–induced NO release, underlining the importance of tyrosine phosphorylation in the endothelial response to shear stress. The fundamental changes taking place within the cell that render the NO synthase essentially insensitive to Ca2+ remain to be fully elucidated. However, preliminary experiments suggest that the exposure of endothelial cells to shear stress markedly alters the physical characteristics of the particulate NO synthase (I. Fleming, unpublished data, 1995). Whether this shear stress–induced enhancement of membrane binding increases the activity of the enzyme at basal [Ca2+]i via an alteration in its microenvironment or via enhanced calmodulin binding is currently under investigation.
Selected Abbreviations and Acronyms
|HOE 694||=||3-methylsulfonyl-4-piperidinobenzoyl guanidine hydrochloride|
|MAP kinase||=||mitogen-activated protein kinase|
|PKC||=||protein kinase C|
|U46619||=||9,11-dideoxy-9α,11α- epoxymethanoprostaglandin F2α|
This study was supported by the Deutsche Forschungsgemeinschaft (Bu 436/4-3 and He 1587/5-1) and the Commission of the European Communities (BMH1-CT93-1893). Dr Hecker was supported by a Heisenberg fellowship from the Deutsche Forschungsgemeinschaft, and Dr Ayajiki was supported by a scholarship from the Japanese Ministry of Education, Science, and Culture. The authors are indebted to Isabel Winter and Annette Kirsch for expert technical assistance and to Dr Alexander Mülsch for measuring NO release from the bioassay system using electron spin resonance spectroscopy.
- Received August 28, 1995.
- Accepted January 16, 1996.
- © 1996 American Heart Association, Inc.
Bevan JA, Kaley G, Rubanyi GM. Flow-Dependent Regulation of Vascular Function. New York, NY: Oxford University Press; 1995.
Griffith TM, Edwards DH. Myogenic autoregulation of flow may be inversely related to endothelium-derived relaxing factor activity. Am J Physiol. 1990;258:H1171-H1180.
Pohl U, Herlan K, Huang A, Bassenge E. EDRF-mediated shear-induced dilation opposes myogenic vasoconstriction in small rabbit arteries. Am J Physiol. 1991;261:H2016-H2023.
Busse R, Mülsch A, Fleming I, Hecker M. Mechanisms of nitric oxide release from the vascular endothelium. Circulation. 1993;87(suppl V):V-18-V-25.
Langille BL, Graham JJK, Kim D, Gotlieb AI. Dynamics of shear-induced redistribution of F-actin in endothelial cells in vivo. Arterioscler Thromb. 1991;11:1814-1820.
Davies PF, Tripathi SC. Mechanical stress mechanisms and the cell: an endothelial paradigm. Circ Res. 1993;72:239-245.
Morita T, Kurihara H, Maemura K, Yoshizumi M, Yazaki Y. Disruption of cytoskeletal structures mediates shear stress-induced endothelin-1 gene expression in cultured porcine aortic endothelial cells. J Clin Invest. 1993;92:1706-1712.
Davies PF, Barbee KA. Endothelial cell surface imaging: insights into hemodynamic force transduction. News Physiol Sci. 1994;9:153-157.
Prasad ARS, Logan SA, Nerem RM, Schwartz CJ, Sprague EA. Flow-related responses of intracellular inositol phosphate levels in cultured aortic endothelial cells. Circ Res. 1993;72:827-836.
Rubanyi GM, Romero JC, Vanhoutte PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol. 1986;250:H1145-H1149.
Pohl U, Holtz J, Busse R, Bassenge E. Critical role of endothelium in the vasodilator response to increased flow in vivo. Hypertension. 1986;8:37-44.
Ohno M, Gibbons GH, Dzau VJ, Cooke JP. Shear stress elevates endothelial cGMP: role of a potassium channel and G protein coupling. Circulation. 1993;88:193-197.
Falcone JC, Kuo L, Meininger GA. Endothelial cell calcium increases during flow-induced dilation in isolated arteries. Am J Physiol. 1993;264:H653-H659.
Shen J, Luscinskas FW, Connolly A, Dewey CF Jr, Gimbrone MA Jr. Fluid shear stress modulates cytosolic free calcium in vascular endothelial cells. Am J Physiol. 1992;262:C384-C390.
Hecker M, Mülsch A, Bassenge E, Busse R. Vasoconstriction and increased flow: two principal mechanisms of shear stress-dependent endothelial autacoid release. Am J Physiol. 1993;265:H828-H833.
Fleming I, Hecker M, Busse R. Intracellular alkalinization induced by bradykinin sustains activation of the constitutive nitric oxide synthase in endothelial cells. Circ Res. 1994;74:1220-1226.
Fleming I, Fisslthaler B, Busse R. Calcium signaling in endothelial cells involves activation of tyrosine kinases and leads to activation of mitogen-activated protein kinases. Circ Res. 1995;76:522-529.
O’Neill WC. Flow-mediated NO release from endothelial cells is independent of K+ channel activation or intracellular Ca2+. Am J Physiol. 1995;269:C863-C869.
Kanai AJ, Strauss HC, Truskey GA, Crews AL, Grunfeld S, Malinski T. Shear stress induces ATP-independent transient nitric oxide release from vascular endothelial cells, measured directly with a porphyrinic microsensor. Circ Res. 1995;77:284-293.
Kuchan MJ, Frangos JA. Role of calcium and calmodulin in flow-induced nitric oxide production in endothelial cells. Am J Physiol. 1994;266:C628-C636.
Barbee KA, Davies PF, Lal R. Shear stress–induced reorganization of the surface topography of living endothelial cells imaged by atomic force microscopy. Circ Res. 1994;74:163-171.
Ziegelstein RC, Cheng L, Capogrossi MC. Flow-dependent cytosolic acidification of vascular endothelial cells. Science. 1992;258:656-659.
Fleming I, Bara AT, Busse R. Calcium signalling and autacoid production in endothelial cells are modulated by changes in tyrosine kinase and phosphatase activity. J Vasc Res. In press.
Tseng H, Peterson TE, Berk BC. Fluid shear stress stimulates mitogen-activated protein kinase in endothelial cells. Circ Res. 1995;77:869-878.