Delayed Platelet Adhesion/Aggregation at Sites of Endothelial Injury in Mouse Cerebral Arterioles After Transient Elevations of Blood Pressure and Shear
Background and Purpose Prior research showed that injection of angiotensin II (Ang II) produces a transient elevation of blood pressure (BP) and shear in pial arterioles. This inhibited platelet adhesion/aggregation at a site of subsequently injured endothelium. The present study attempted to confirm the Ang II finding with a different method of endothelial injury, to test the hypothesis that the effect on adhesion/aggregation was a consequence of prolonged release of “classic” endothelium-derived relaxing factor (released by acetylcholine [Ach]; EDRFACh) produced by the preceding transient elevation in shear, and to show with the use of norepinephrine rather than Ang II that the effect of a preceding elevation of BP was independent of the pressor agent used.
Methods Focal platelet adhesion/aggregation was elicited in cerebral surface (pial) arterioles by producing minimal endothelial damage with a helium-neon laser/Evans blue dye technique. Vessels were observed by intravital microscopy. We recorded the time required for the laser to elicit adhesion/aggregation in control mice and in mice given Ang II in a dose of 16 μg/25 g IP. This dose produces an abrupt and significant elevation of BP and shear, which return to baseline levels in less than 30 minutes. Laser/dye damage of endothelium and resultant adhesion/aggregation of platelets were not induced until after BP and shear returned to basal levels. The effect of topical Ang II on damage-induced adhesion/aggregation was also tested. In addition, mice injected with Ang II were treated with either topical indomethacin 40 μg/mL or topical NG-monomethyl l-arginine (L-NMMA; 10−6 mol/L) in an effort to prevent the preceding increase in shear from inhibiting subsequent adhesion/aggregation. Finally, norepinephrine instead of Ang II was used to transiently raise BP (and shear) in an effort to delay subsequently induced adhesion/aggregation.
Results Platelet adhesion/aggregation at the injured site was significantly delayed by a prior transient rise in shear produced by either Ang II or norepinephrine. Locally applied Ang II failed to influence adhesion/aggregation, although a previous study showed that such Ang II reaches the endothelium. Locally applied indomethacin had no effect on inhibition of platelet adhesion/aggregation, but locally applied L-NMMA prevented the prior transient elevation of shear from inhibiting adhesion/aggregation at a subsequently injured site.
Conclusions Elevation of BP with consequent elevation of shear inhibits local platelet adhesion/aggregation even when the latter is initiated by endothelial damage produced after return of shear to basal levels. The direct action of Ang II on endothelium is not responsible for the effect on adhesion/aggregation, and indeed the effect is independent of the pressor agent. The pharmacological data, together with the literature, support the hypothesis that increased shear causes an increased release of EDRFACh, which may continue for at least some minutes after return of shear to normal levels.
This laboratory has previously shown that injection of angiotensin II (Ang II) causes a delay in the appearance of an adhering platelet aggregate at a site of endothelial injury in brain arterioles.1 This effect on platelet adhesion/aggregation was preceded by a transient rise in blood pressure (BP) and shear produced by the injection of Ang II. The inhibition of adhesion/aggregation was not due to a direct effect of Ang II on the platelet.1 Moreover, the delay could not have been caused by increased shear wiping the platelets off the injured endothelium because the injury was not pro- duced until after BP and shear had returned to basal levels. Increased release of prostaglandin (PG) I2 as a direct effect of Ang II action on endothelium2 3 was the hypothesized cause of the inhibited adhesion/aggregation. However, in later unpublished studies from this laboratory, direct measurement of 6-keto-PGF2α failed to show increased levels in plasma at the time when adhesion/aggregation was impaired.
Since then, the endothelium has been shown to release an endothelium-dependent relaxing factor (EDRF) derived at least in part from l-arginine and containing nitric oxide.4 Because this EDRF is released by acetylcholine (ACh) and in the cerebral circulation is of uncertain chemical composition,5 this laboratory has called it EDRFACh. This mediator not only dilates arterioles but is also a potent inhibitor of platelet adhesion/aggregation.4 6 Responses caused by EDRFACh are inhibited by NG-monomethyl l-arginine (L-NMMA).4 7 8 The basal release of EDRFACh is increased by increasing shear.9 Consequently, we can now hypothesize that the effect of our Ang II injection was due to increased release of EDRFACh, which continued for some minutes after shear had returned to normal. An effect of shear on release of prostacyclin10 was ruled out by experiments with indomethacin.
To further test the hypothesis that transient shear results in prolonged antiplatelet activity due to EDRFACh, the experiments reported below were performed. First, we replicated the original Ang II study using a different means of injuring the endothelium and eliciting platelet adhesion/aggregation. Second, the effect of the Ang II injection on adhesion/aggregation was nullified by applying topical L-NMMA. Moreover, Ang II directly applied to the pial vessels failed to influence adhesion/aggregation at the damaged site. Third, we repeated the study using norepinephrine instead of Ang II to transiently raise BP and shear. This last study effectively rules out effects of Ang II on other tissues as an indirect cause of the delayed platelet adhesion/aggregation and shows that adhesion/aggregation can be inhibited by an immediately preceding rise in shear irrespective of the pressor used to cause that rise in shear.
In the original study from this laboratory,1 red blood cell velocity was actually measured. Shear was calculated from the measurement of red blood cell velocity and diameter and found to rise and to return to normal levels before injuring the endothelium. Shear parallels BP and will be at control levels when BP returns to normal provided that arterioles have also returned to normal diameter. Since injections of either Ang II or norepinephrine in mice produce stereotypic transient elevations in BP with stereotypic return of pial arterioles to normal diameters once BP has normalized, it was thought unnecessary to actually measure shear in the present studies. Only BP was measured.
Materials and Methods
The preparation has been described in many publications.11 12 13 All procedures were in accordance with the guidelines of our Institutional Animal Care and Use Committee. Briefly, 166 male ICR mice (Harlan Sprague-Dawley, Indianapolis, Ind) were anesthetized with urethane, and a tracheotomy and craniotomy were performed. The dura was stripped, exposing the transparent arachnoid and underlying pial vessels. The mice were maintained at 37°C. The craniotomy site was continuously suffused with mock cerebrospinal fluid (Elliott’s solution [Elliott and Jasper14 ]) at a pH of 7.35. All topically applied drugs were administered in the mock cerebrospinal fluid at the same pH.
The arterioles (inside diameter, 30 to 50 μm) were viewed directly through a microscope. In each mouse one arteriole was selected for observation, the choice depending only on the presence of a straight segment as close to 35 μm wide as possible. The endothelium was injured with a laser/dye technique7 12 15 16 17 18 with the use of a helium-neon laser and intravascular Evans blue dye. The damage produced by the laser/dye is dependent on laser duration. Initially, no damage can be detected by transmission electron microscopy, but endothelium-dependent responses are lost.12 No platelet adhesion or aggregation is seen over the injured area. More prolonged exposure results in local platelet adhesion/aggregation. When they occur, such aggregates are recognized as “white bodies”19 when the vessel is viewed directly through the microscope.
Several experimental protocols were used: First, platelet adhesion/aggregation was initiated by the laser/dye, and the time required for the laser to initiate the first recognizable adhering aggregate was recorded. This time was recorded in control (saline-injected) mice and in mice injected with Ang II (16 μg/25 g IP) 30 minutes before laser injury. In this protocol, the intraperitoneal dose of Ang II we used was based on preliminary studies showing that it increased systemic BP by 53±19 mm Hg (mean±SD) within 2 minutes of intraperitoneal injection and that within 5 to 10 minutes the pressure declined toward control levels, reaching them within 30 minutes. Once this was established, BP was not monitored in the studies of platelet aggregation, but the preliminary study was so consistent that we are confident that all platelet aggregates were induced after BP had returned to normal.
Another protocol used norepinephrine 1.28 mg/kg IP instead of Ang II, and pressures were monitored continuously via femoral artery cannulas. The norepinephrine injection elevated pressure 51±15 mm Hg (mean±SD) within 2 minutes, with rapid return to baseline within 5 to 27 minutes after injection (n=20). The latency of adhesion/aggregation in the norepinephrine-treated group was compared with that in saline-injected controls, whose pressure did not change.
Aggregation latency was also measured in mice given topical treatment with Ang II or given topical L-NMMA (10−6 mol/L) or indomethacin (40 μg/mL) alone or in combination with intraperitoneal Ang II, as described in “Results.”
In experiments with only two groups, Student’s t test was used to compare aggregation latencies. Comparisons in experiments with more than two groups were made with Duncan’s test for multiple comparisons. In each experiment the mice in any group were alternated with mice in each of the other groups so that a strict contemporary design was maintained; that is, the mice in one experiment were all studied during the same period of time. Groups in one experiment should not be compared with groups in another experiment examined at a different time.
The drugs used in the present study were Ang II amide (Sigma); norepinephrine bitartrate (Sanofi Winthrop); indomethacin trihydrate (Merck), an inhibitor of cyclooxygenase; and L-NMMA (Sigma).
In all protocols, arterial Pco2, Po2, and pH values were obtained from arterial blood in each mouse. These values attested to the overall condition of the mice and were similar from group to group. The mean±SD values for all mice were as follows: CO2, 34±4 mm Hg; O2, 106±14 mm Hg; and pH, 7.37±0.11.
Effect of Systemic Ang II on Adhesion/Aggregation
Table 1⇓ shows that 30 minutes after the intraperitoneal injection of Ang II, when BP and shear were once again at control levels, it took significantly longer to induce platelet adhesion/aggregation at the injured site. The table also shows that the effect of Ang II was nullified by the presence of L-NMMA 10−6 mol/L topically applied for 10 minutes.
In a study separate from (ie, not contemporary with) the study in Table 1⇑, Ang II was again found to significantly delay the onset of adhesion/aggregation (Table 2⇓). In this study the adhesion/aggregation latency in Ang II–treated mice was not affected by topical application of 40 μg/mL indomethacin. Indomethacin applied in mice not injected with Ang II also had no effect.
Effect of Topical Ang II on Adhesion/Aggregation
The preceding data showed that Ang II inhibited adhesion/aggregation at the damaged site via an L-NMMA–inhibitable mechanism. The question remained: Was this a direct effect of Ang II or the result of some other action of Ang II that in turn inhibited adhesion/aggregation? To test the direct effect of Ang II we applied it topically (10 μg/mL) for 5 minutes. In the treated group (n=8), aggregation at the injured site began 35±16 (mean±SD) seconds after the laser was turned on. This was identical to the mean latency in eight controls (35±16 seconds).
Effect of Systemic Norepinephrine on Adhesion/Aggregation
Since the direct effect of Ang II on the vessels did not seem to account for the delayed platelet adhesion/aggregation caused by Ang II injection, we further tested the hypothesis that the delay was caused by the preceding period of abruptly elevated pressure and shear. We injected 13 mice with norepinephrine instead of Ang II. Thirty minutes later, after pressure had returned to normal, we provoked local platelet adhesion/aggregation. The aggregation latency in the norepinephrine-injected group was 85±37 seconds (mean±SD) compared with 53±28 seconds in 13 contemporary controls (P=.03, Student’s t test).
The data confirm that platelet adhesion/aggregation is delayed at a site of endothelial injury when the injury is produced after a transient increase in BP. The original findings, which used Ang II as the pressor,1 are confirmed with the use of a totally different laser/dye model to produce the endothelial injury. Moreover, the new data show that even when norepinephrine is used instead of Ang II to produce transient elevations of shear, platelet adhesion/aggregation is delayed by this pretreatment. Thus, the effect of the pressors in this study is not likely to be due to remote actions of one pressor or the other on organs other than brain.
The new data also show that the effects of Ang II are unlikely to be due to direct effects of Ang II on endothelium. This seems unlikely because suffusion of Ang II over the vessels for 5 minutes failed to influence the latency of adhesion/aggregation elicited at a point of endothelial damage. One may ask whether such application of Ang II does in fact reach the endothelium. The answer is affimative, since this laboratory has shown that in the mouse topical Ang II in doses lower than that used here produces endothelium-dependent constriction of pial arterioles.20 If the endothelium is injured, the constriction is eliminated or inhibited. Therefore, topically applied Ang II reaches the endothelium and releases some constricting mediator from it.
Finally, the new data show that L-NMMA, an inhibitor of EDRFACh-dependent responses, prevents the inhibitory effect of elevated shear on subsequent platelet adhesion/aggregation. The concentration of L-NMMA was 10−6 mol/L, a concentration that does not cause constriction of these arterioles but that does inhibit dilation dependent on EDRFACh.7 Presumably, this is a reflection of the shape of the dose-response curve relating EDRFACh to tone. In the initial relatively flat part of typical dose-response curves, it takes a large change in dose (eg, L-NMMA and EDRFACh) to produce a change in diameter. This may be the situation in the basal state. But when more EDRFACh is available (eg, when ACh is applied or shear is increased), then one is on the ascending limb of the curve, and small changes in EDRFACh levels, such as those caused by low doses of L-NMMA, will have much more noticeable effects on the response.
In contrast, indomethacin, an inhibitor of prostanoid synthesis, had no effect. This supports the hypothesis that EDRFACh and not a prostanoid is the mediator of the effect on adhesion/aggregation. This is important because increased shear can increase release of both EDRFACh and prostacyclin.9 10
The inhibitory actions of Ang II and norepinephrine treatment are not due to a long-acting effect of either pressor on the mouse platelets because Ang II has no antiaggregant effect on these platelets,1 and norepinephrine also fails to inhibit platelet aggregation in vitro (unpublished data, this laboratory).
The interpretation of our data rests on the assertion that shear was in fact elevated when either Ang II or norepinephrine was injected and on the assertion that shear was back to baseline levels when the endothelial injury was induced. These assertions are supported by the following facts: (1) In the initial Ang II study in this series,1 shear was calculated from direct measurement of red blood cell velocity and from arteriolar diameter. Shear behaved as described above. (2) According to the laws of physics, shear increases whenever intraluminal pressure increases and diameter remains unchanged or narrows. Direct pressure measurements in pial arterioles in this laboratory have shown that changes of pressure in pial arterioles parallel changes in systemic pressure.1 Moreover, well-known autoregulatory responses result in either no change or constriction of pial arterioles as the pressure rises, with return of baseline diameters when pressure returns to baseline levels (unpublished data, this laboratory). (3) In the present study of norepinephrine, pressure was continuously monitored, rose substantially and abruptly as described in “Materials and Methods,” and returned to baseline before initiation of endothelial injury. Thus, in both the Ang II studies and the norepinephrine study, shear must have risen abruptly and then returned to basal levels before the induction of endothelial injury and local platelet adhesion/aggregation.
The data do not permit an assessment of the degree of transient shear elevation required to subsequently delay adhesion/aggregation, nor do the data permit an assessment of how long the effects of a prior transient shear increase will last. One assumes that the height and diameter of the former will affect the latter. Relevant parameters would have to be monitored in a very much larger series of animals to provide data adequate to address these quantitative relations. Others22 have reported that in larger arteries, increased release of an EDRF may continue for at least 8 to 12 minutes after return of shear to control levels.
The latency of adhesion/aggregation is a “lumped” parameter that is affected by factors that modify both adhesion and initial aggregation. That is because the in vivo microscopic technique used here detects the first recognizable adhering mass of platelets. This is a function not only of adhesion to damaged endothelium but also of initial platelet-to-platelet aggregation at the site of adhesion. EDRFACh inhibits both adhesion and aggregation and is a more potent inhibitor of adhesion than prostacyclin.4 21 This is consistent with the data showing that L-NMMA but not indomethacin inhibits the effect of transiently elevated shear.
The control values for adhesion/aggregation latency were different in each of the three studies (Tables 1⇑ and 2⇑ and the norepinephrine study in “Results”). The control values in Table 1⇑ are higher than those in Table 2⇑ and lower than those in the norepinephrine study. However, the value in Table 1⇑ is not significantly different from the controls in either Table 2⇑ or the norepinephrine study. Nevertheless, because control values can vary for unknown reasons, from day to day, week to week, or month to month, it is essential that all groups in any given study be examined in contemporary fashion. Groups in one study should not be compared with groups in another study. In each of the three studies conducted here, the animals in one group were alternated with animals in the other groups so that equal numbers in each group were observed during the same period of time.
The data are of particular interest because they were obtained in vivo in the cerebral microcirculation. Initially, all studies of the effects of shear-related increases in EDRFACh on platelet adhesion/aggregation have been in vitro or ex vivo and involved larger vessels or tissue culture of endothelial cells. The present data suggest that in pial arterioles, after an increase in shear, there may be a functionally significant increment in release of EDRFACh for at least some minutes after shear has returned to normal. This might be important after transient, significant increases in BP, which might otherwise be expected to injure endothelium with conse-quent platelet adhesion/aggregation in the injured vessels.
This study was supported by grants HL-35935 and HL-45617 from the National Heart, Lung, and Blood Institute.
Reprint requests to Dr William I. Rosenblum, Box 980017, MCV Station, Richmond, VA 23298-0017.
Review of this manuscript was directed by Guest Editor Michael S. Wolin, PhD.
- Received November 15, 1994.
- Revision received December 16, 1994.
- Accepted December 29, 1994.
- Copyright © 1995 by American Heart Association
Rosenblum WI, El-Sabban F, Hirsh PO. Angiotensin delays platelet aggregation after injury of cerebral arterioles. Stroke. 1986;17:1203-1205.
Gryglewski RJ, Korbut R, Splawinski J. Endogenous mechanisms which regulate prostacyclin release. Hemostasis. 1979;8:294-299.
Mullane KM, Moncada S. Prostacyclin release and the modulation of some vasoactive hormones. Prostaglandins. 1980;29:25-49.
Rosenblum WI. Endothelium-derived relaxing factor in brain blood vessels is not nitric oxide. Stroke. 1992;23:1527-1532.
Nishimura H, Rosenblum WI, Nelson GH, Boynton S. Agents that modify ‘EDRF’ formation alter antiplatelet properties of brain arteriolar endothelium in vivo. Am J Physiol. 1991;261:H15-H21.
Rosenblum WI, Nishimura H, Nelson GH. Endothelium dependent L-ARG and L-NMMA sensitive mechanisms regulate tone of cerebral microvessels. Am J Physiol. 1990;259:H1396-H1401.
McIntire LV, Frangos JA, Eskin SG, Hall ER. Effect of hemodynamic shear on arachidonic acid metabolism of vascular endothelium. In: Hartmann A, Kuschinsky W, eds. Cerebral Ischemia and Hemorrheology. Berlin, Germany: Springer-Verlag; 1987:280-289.
Rosenblum WI. Constriction of pial arterioles by prostaglandin F2α. Stroke. 1975;6:293-297.
Rosenblum WI, Nelson GH, Povlishock JT. Laser induced endothelial damage inhibits endothelium dependent relaxation in the cerebral microcirculation of the mouse. Circ Res. 1987;60:169-176.
Faraci FM, Baumbach GL, Heistad DD. Myogenic mechanisms in the cerebral microcirculation. J Hypertens. 1989;7:S61-S64.
Rosenblum WI, Nelson GH. Endothelium dependence of dilation of pial arterioles in mouse brain by calcium ionophore. Stroke. 1988;19:1379-1382.
Rosenblum WI, Nelson GH, Weinbrecht P. Histamine elicits competing endothelium-dependent constriction and endothelium-independent dilation in vivo in mouse cerebral arterioles. Stroke. 1990;21:305-309.
Rosenblum WI, Nelson GH, Shimizu T. Endothelium dependent constriction by angiotensin II demonstrated in vivo in mouse brain arterioles. Endothelium. In press.
Rubanyi GM, Romero JC, Vanhoutte PM. Flow-induced release of endothelium-derived relaxing factor. Am J Physiol.. 1986;250:H1145-H1149.