Vascular injuries induced by materials released from platelet-rich thrombus in vivo

The effects of thrombus, thrombectomy and thrombolysis on endothelial function

OBJECTIVE

this study was undertaken to examine and compare the effects of thrombus, thrombectomy, and thrombolysis on endothelial function as measured by endothelium-dependent vasorelaxation (EDR).

METHODS

adult, male New Zealand white rabbits underwent ligation of the left common iliac to femoral artery to induce thrombosis and were then randomly assigned to one of five groups, n=6 in each. Group A consisted of ligation and thrombosis for 4 h. Group B underwent similar ligation for 4 h, but without intraluminal thrombus present. Following 4 h of ligation and thrombosis, Group C underwent thrombectomy while group D was treated with urokinase (UK), 4000 U/min for 30 min. Group E underwent UK infusion alone. The right external iliac artery served as control vessel in each group. All arteries were removed and endothelial function was determined by measuring EDR.

RESULTS

the presence of thrombus reduced EDR by 50% (group A) compared to control. Vessels with interrupted flow, but not exposed to thrombus, retained normal EDR (group B). Thrombectomy decreased EDR significantly (group C) compared to thrombolysis (group D) and control. UK did not significantly alter EDR (groups D, E).

CONCLUSIONS

exposure of endothelium to thrombus significantly decreases EDR. EDR was not affected by interruption of blood flow in the absence of thrombus. Thrombectomy appeared to cause a further additive insult to the endothelium. In contrast, thrombolysis with UK preserved residual endothelial function. These data suggest that it is important to differentiate the effects of thrombus on endothelium from effects due to thrombectomy or thrombolysis when evaluating treatment modalities for arterial thrombosis.

Exposure to thrombus diminishes endothelial derived relaxation in the rabbit carotid artery

PURPOSE

Thrombus is believed to be deleterious to intimal function. However, few studies have directly examined this effect. This study examines the effect of thrombus on endothelial-dependent and -independent vasorelaxation in the rabbit carotid artery.

METHODS

Twelve male New Zealand white rabbits (3.5-4.5 kg) were divided into two groups of six. Thrombosis was induced in group I by segmental right carotid artery ligation. Group II underwent segmental right carotid ligation immediately followed by removal of thrombus with normal saline flush through an arteriotomy. The left carotid arteries were exposed in both groups and served as internal controls. After 4 h, left and right carotid arteries were harvested, sectioned into 6-mm rings, and mounted on isometric force transducers in a physiologic bath. Thrombus was removed from the arteries in group I during the ring preparation process. Neither group I nor group II had thrombus in contact with endothelium during ex vivo testing. The arterial rings were constricted with norepinephrine (1 x 10(-4) M). Endothelium-dependent and -independent vasorelaxation to acetylcholine (Ach) and s-nitrosoacetylpenicillamine, respectively, were measured in a dose-response manner. Results were expressed as a percentage of vasorelaxation. Statistical analysis was performed using an analysis of variance.

RESULTS

Endothelial-dependent vasorelaxation, which tests for endothelial cell function, was decreased in the thrombus and endothelial ischemia group (I) compared to control as noted by vasorelaxations of 22% vs 34% at 1 x 10(-4) molar concentration Ach, and 33% vs 48% at 1 x 10(-3) molar concentration Ach, respectively (P = 0.05). By comparison, there was no difference in the endothelial-dependent vasorelaxation of the endothelial ischemia group (II) versus control. Endothelial-independent vasorelaxation, which tests for smooth muscle function, was not affected by either the thrombus and endothelial ischemia group (I) or the endothelial ischemia group (II) compared to the control group. The controls in group I and group II were slightly different. When this difference was removed, the resulting comparison of treatments in group I and group II approached significance at molar concentrations of 1 x 10(-4), 1 x 10(-5), and 1 x 10(-6) (P = 0.07, 0.06, 0.06).

CONCLUSIONS

The presence of thrombus within the rabbit carotid artery for a period of 4 h decreases endothelial-dependent relaxation. Four hours of endothelial ischemia without thrombus did not change endothelial-dependent vasorelaxation. Neither thrombus nor ischemia alone had any effect on the endothelium-independent vasorelaxation. We conclude that thrombus is deleterious to endothelial function independent of smooth muscle function in the acute setting as measured by endothelial-dependent vasorelaxation.

Renal involvement of thrombotic thrombocytopenic purpura: special reference to the glomeruloid structures

We report the case of a 9-year-old girl with biopsy-proven renal thrombotic microangiopathy in thrombotic thrombocytopenic purpura (TTP), with particular reference to the glomeruloid structures. The renal biopsy sample from this TTP patient revealed platelet thrombus deposition, a glomeruloid structure and aneurysm with relative sparing of the glomeruli. The glomeruloid structure displayed a proliferation of mainly capillary-sized channels lined by Factor VIII-related, antigen-positive plump endothelial cells embedded in the edematous connective tissue. These glomeruloid vessels communicated with the aneurysmal segment at the end portion of the arteriolar branch. We believe that the glomeruloid structures in TTP represent not merely organization or recanalization of thrombus but rather active angiogenesis through aneurysmal dilation in the arteriolized vessel, probably initiated by platelet agglutination.

Changes in the proliferative activities of cells in experimental atherosclerotic plaques during remodeling

To investigate the relationship between cytologic alterations and cellular proliferation during atherosclerotic remodeling, we examined experimental atheromatous plaques by immunohistochemistry. Plaques were formed on rabbit aortas by cholesterol-enriched diets and mechanical stimulation over a period of 2 months. Plaques were examined 1 month and 6 months after induction. We used antibodies RAM-11, HHF-35, and monoclonal anti-proliferating cell nuclear antigen (PCNA) antibody for detection of macrophages (Mphi), smooth muscle cells (SMC), and PCNA, respectively. One month after induction, the plaques revealed a thickened intima with a fibrofatty histologic pattern or accumulation of foam cells. With either histologic pattern, foam cells were found to be Mphi and proliferative activity was mainly observed in Mphi. Six months after induction, calcification and organization were seen on the induced plaques, suggesting progression of remodeling. There were fewer Mphi and more SMC compared with plaques examined 1 month after induction. Proliferative activity was observed mainly in SMC. We have demonstrated that the proliferative activity of cell types changes during remodeling of atheromatous plaques. Our results suggest an important relationship between the proliferative activity of SMC and remodeling.

Impaired NO release from bovine aortic endothelial cells exposed to activated platelets

We have previously shown that aggregated human platelets elicited a decrease in intracellular adenosine triphosphate (ATP), enhanced adenosine egress and damage to mitochondria in bovine aortic endothelial cells (ECs). To test whether such metabolic and ultrastructural changes could be associated with functional impairment of ECs, we investigated the effects of activated platelets on nitric oxide (NO) and prostacyclin release, and on the antiaggregation property of ECs. Pretreatment of ECs with aggregated platelets transiently stimulated basal NO release while prolonged (> or = 30 min) exposure dose-dependently inhibited NO release, both basal and in response to ATP or serotonin, with NO synthase activity being attenuated in these cells. Supplementary L-arginine (L-A) restored NO release completely. Prostacyclin release was also stimulated transiently but not affected by prolonged pretreatment. The antiaggregation property of ECs was attenuated by pretreatment with activated platelets but restored with L-A supplement. Although the effects of activated platelets and 0.5 mM acetylsalicylic acid (ASA) to attenuate the antiaggregation property of ECs were additive, activated platelets had no effect on ECs treated with 0.2 mM N omega-nitro-L-arginine (L-NA), suggesting a common mechanism. We conclude that prolonged exposure to aggregated platelets may affect the antiaggregation property of ECs by directly inhibiting NO synthesis, which may be normalized by L-A supplementation.

A platelet activation-specific monoclonal antibody that recognizes a receptor-induced binding site on canine fibrinogen

An activation-specific monoclonal antibody (MoAb) termed "Canine Activated Platelet 1" (CAP1) has been developed and partially characterized. Flow cytometric studies of isolated canine platelets, using adenosine diphosphate (ADP) and platelet activating factor (PAF) as agonists, demonstrated that CAPI binding site number was proportional to agonist strength and agonist concentration. MoAb CAP1 binding was diminished by ethylenediamine-tetraacetic acid, suggesting that the antigen was either stabilized by calcium or antigen binding to the platelet surface was mediated by calcium. ADP-activated gel-filtered platelets also demonstrated reduced binding of MoAB CAP1 even in the presence of 1 mM CaCl2. Binding of MoAb CAP1 could be partially restored by activating gel-filtered platelets with PAF, suggesting that the antigen was either present within platelet granule membranes or was exposed after binding of released proteins(s) with a platelet receptor. A monoclonal antibody to human platelet glycoprotein IIIa (GPIIIa), which cross-reacts with canine platelet GPIIIa regardless of platelet activation status, did not inhibit binding of MoAb CAP1. MoAb CAP1 bound to isolated canine fibrinogen captured on polystyrene microtiter plates in the absence of platelet proteins. Immunoblots indicated that MoAb CAP1 recognizes nonreduced fibrinogen as well as a plasmin digest of isolated canine fibrinogen. Results of the present studies suggest that MoAb CAP1 recognizes a receptor-induced binding site on canine fibrinogen.

Hyperthermia induces ultrastructural changes in mouse pial microvessels

BACKGROUND

Pial microvessels' responses to local hyperthermia revealed the development of in vivo spontaneous thrombosis. The cellular and subcellular changes which contribute to such events remained unexplored. Therefore, the effect of regional hyperthermia (43 degrees C) on mouse pial microvessels was studied at the ultrastructural level.

METHODS

A simple cranial window assembly, including an artificial cerebrospinal fluid delivery and heating system to ensure a precise brain regional temperature, was used. The animal core body temperature was maintained at 37 degrees C. Topical and transvessel bimodal fixation of microvessels was done with a phosphate buffered mixture of glutaraldehyde and paraformaldehyde, followed by a standard electron microscopy procedure.

RESULTS

When the pial microvessels of control (37 degrees C) animals were examined, no evidence of cellular damage was discerned. Endothelial cells including luminal membrane were unchanged. Degranulated platelets or platelet aggregates were not seen. However, numerous platelets in association with scattered red blood cells and occasional white blood cells could be observed in a close proximity, but not adhered, to the endothelial wall of hyperthermic (43 degrees C) brains. Platelets displayed a variety of forms consistent with the onset of platelet activation. Discoid platelets containing granules and spheroid degranulated platelets and those with large pseudopodia were recognized. The venular endothelial surface revealed conspicuous endothelial change, with the presence of endothelial denudation. The site of platelet aggregation in both venules and arterioles was accompanied by focal endothelial lucency and denudation vacuole formation, luminal membrane rupture, and swelling of the nuclear envelope.

CONCLUSIONS

These findings demonstrate the extent of damage to the pial microvasculature in response to a local hyperthermic exposure. The results emphasize that changes in the endothelium may represent the earliest signs of oncoming vascular pathology.

The effect of thrombin and serine proteases on intracellular Ca2+ in rat aortic smooth muscle cells

Cultures of vascular smooth muscle cells (VSMC) are commonly used to study the events and defects found in hypertension and atherosclerosis. In particular Ca2+ homeostasis in cellular signalling has been the focus of extensive research. Since trypsin has been shown to mobilise Ca2+ in some cell types, we have investigated its effect on various aspects of Ca2+ homeostasis in rat aortic smooth muscle cells (RASMC). The effects of trypsin, alpha-chymotrypsin and elastase (other serine proteases) on intracellular Ca2+ in cultured aortic cells isolated from Wistar rats have been investigated. Trypsin (24 micrograms/ml) elicits intracellular Ca2+ mobilisation, after which cells become nonresponsive to thrombin Ca2+ mobilisation but retain responsiveness to Angiotensin II (AII). alpha-Chymotrypsin (24 micrograms/m) inhibits the thrombin Ca2+ mobilising response, without itself initiating a Ca2+ transient or affecting AII Ca2+ mobilisation. Elastase (24 micrograms/ml) was not effective in mobilising intracellular Ca2+ or inhibiting the thrombin response. We have also observed diminished thrombin Ca2+ mobilisation responses between cells in suspension and cell monolayers, which appeared to be unrelated to proteolysis but due to morphological changes of the cells. Our results suggest that trypsin acts on the thrombin receptor via a specific proteolysis mechanism to mobilise intracellular Ca2+ ([Ca2+]i) in RASMC. The amount of Ca2+ released by thrombin or trypsin is dependent on the morphology of the cell and the state of the tethered ligand of the thrombin receptor exposed by the protease.

Correlations between morphological changes in platelet aggregates and underlying endothelial damage in cerebral microcirculation of mice

BACKGROUND AND PURPOSE

The aim of this study was to test the hypothesis that, once formed, platelet aggregates may injure underlying cerebrovascular endothelium. Such injury could make the same site selectively attractive to the next wave of passing emboli or activated platelets. This vicious circle could account for repetitive, stereotypic symptoms in transient ischemic attacks.

METHODS

In pial arterioles, minor endothelial injury was produced by a laser/dye technique. After various periods of platelet aggregation at the damaged site, the vessels were fixed in situ for electron microscopic study. The degree of platelet activation (rounded and/or degranulated forms) was evaluated by counting these forms in the electron photomicrographs. These counts were related to the degree of endothelial damage ascertained in the micrographs. Other statistical relations were also examined.

RESULTS

Endothelial damage progressed in parallel with the duration of platelet aggregation and the degree of platelet degranulation at the site. Correlations were number of activated platelets versus degree of damage, r = .43, P < .03; duration of aggregation versus damage, r = .52, P < .01; and number of degranulated platelets versus the degree of endothelial damage, r = .83, P < .001. If an aggregate embolized, endothelial damage did not appear to progress. No correlation existed between the duration of exposure to the laser and the degree of injury.

CONCLUSIONS

The parallel between changes in platelets and endothelial damage could represent either an effect of endothelium on platelets or an effect of platelets on endothelium. Although the former alternative cannot be totally ruled out, the observations seem to fit best the hypothesis that progressive endothelial damage can result from increasing activation and degranulation of overlying platelets.

Transient platelet accumulation in the rat brain after common carotid artery thrombosis. An 111In-labeled platelet study

BACKGROUND AND PURPOSE

Thromboembolic events are a major cause of ischemic stroke. To obtain evidence for platelet embolization after cerebrovascular injury, the accumulation of indium-labeled platelets was documented after photothrombosis of the rat common carotid artery.

METHODS

Heterologous blood was collected from donor rats, and the isolated platelets were labeled with 111In-tropolone. Labeled platelets were then infused into Wistar rats 30 minutes before right carotid artery thrombosis. Nonocclusive common carotid artery thrombosis was induced by a laser-driven rose bengal-mediated photochemical insult to the vascular endothelium, and the rats were killed 15 minutes or 3 hours later. Carotid arteries and brains were immediately removed and dissected for regional radioactivity assessment or sectioned for the autoradiographic visualization of platelet emboli.

RESULTS

At 15 minutes after thrombosis, the ratio of right-to-left common carotid artery radioactivity was significantly elevated compared with control (33 +/- 12 [mean +/- SEM] versus 0.97 +/- 0.2). Within individual brain regions, including the frontal and frontoparietal cortices and hippocampus, significant elevations in right-to-left radioactivity ratios were also documented. Autoradiographic images revealed multiple foci of 111In-labeled platelets throughout the thrombosed hemisphere. At the level of the frontal cortex, bilateral platelet accumulation was seen. Regional counts demonstrated significantly increased platelet density within selective cortical and subcortical regions. In contrast to the 15-minute findings, right-to-left ratios of carotid arteries or brain regional radioactivities were not significantly elevated at 3 hours after injury. In addition, the areal densities of autoradiographically visualized platelets in the 3-hour group were not different from control except in the right frontal cortex.

CONCLUSIONS

These data demonstrate (1) the acute accumulation of labeled platelets in downstream vessels after nonocclusive common carotid artery thrombosis, (2) that platelet accumulation is widespread and also involves contralateral areas, and (3) that platelet accumulation within the thrombosed carotid artery and brain is largely transient.

Replication of endothelial cells and smooth muscle cells induced in vivo by hypercholesterolaemia and materials released from platelet-rich white thrombus

Endothelial cell injury is considered to be the primary event in atherogenesis. In this study, we investigated the effects of hypercholesterolaemia, and substances released from platelet-rich thrombi, individually and together on endothelial cells and the wall of the rabbit in vivo. We divided 24 rabbits into four groups: I was a control group on a normal diet; II was a tubing group on a normal diet, in which polyethylene tubing was inserted into the ascending aorta; III was a group being fed a cholesterol diet; and IV was a combined group being fed a cholesterol diet with polyethylene tubing in the ascending aorta. Segments from the descending thoracic and abdominal aortas which were not injured directly by tubing were examined morphologically and for [3H] thymidine incorporation into endothelial cells and smooth muscle cells. The descending aortas of groups II, III, and IV showed various degrees of endothelial cell damage. [3H]Thymidine incorporation into endothelial cells and aortic wall was increased in groups II, III, and IV (most in group IV) as compared with group I. These data indicate that hypercholesterolaemia and substances released from activated platelets and/or white mural thrombi can cause endothelial damage which may result in endothelial and smooth muscle cell proliferation. In addition, a combination of these two factors showed an additive effect on the endothelial injury and regeneration in vivo.

Vascular injuries induced by materials released from white mural thrombus and hypercholesterolemia in vivo

Materials released from platelet-rich white mural thrombi into the arterial circulation and dietary hypercholesterolemia could cause endothelial injury and regeneration, and also smooth muscle cell migration and proliferation in vivo in the intima. In addition, a combination of these two factors showed an additive effect on the endothelial injury in vivo. Further experiments are in progress to clarify how high a level of serum cholesterol and/or oxidized LDL may be sufficient for endothelial injury using this in vivo model for the better understanding of detailed pathogenesis of atherosclerosis.

Antiplatelet therapy for atherosclerotic disorders

Recent studies have revealed the important roles of platelets in atherogenesis via vascular injury. Our in vivo and in vitro studies clearly demonstrate that activated platelets directly inflict injury to vascular endothelial cells, which is associated with a decrease in intracellular cyclic AMP levels in vascular tissues. Antiplatelet therapy is clinically important not only for the prevention of thrombotic episodes but also for the prevention of vascular injury and atherosclerosis. A small dose of aspirin (80 mg) induces clinically hypoaggregativeness of platelets with concomitantly decreased levels of thromboxane A2 in plasma. Our clinical study involving more than 3 years of treatment with small doses of aspirin demonstrated favorable therapeutic effects characterized by hypoaggregation of platelets and increased levels of cAMP and 6-keto PGF1 alpha in plasma which will aid in the prevention of atherosclerosis.

Hypoxia- and endothelium-mediated changes in the pharmacological responsiveness of circumflex coronary artery rings from the sheep

1. The role(s) of the endothelium in modulating the responsiveness of isolated circumflex coronary artery rings (o.d. = 2.0-2.5 mm and o.d. = 0.6-1.3 mm) from sheep was investigated under oxygenated and hypoxic conditions. 2. Removal of the endothelium abolished the contraction produced by lowering the PO2 from 620 to 8 mmHg (either under optimal resting tension or precontracted by 40 mM KCl). In denuded artery rings sudden hypoxia caused relaxation. 3. Under oxygenated conditions, removal of the endothelium augmented the vasoconstrictor effects of U46619, 5-hydroxytryptamine (5-HT) and K+. In the denuded artery rings, hypoxia abolished the contractile effects of U46619 and reduced the contractile effects of 5-HT and K+. 4. Under oxygenated conditions, the vasorelaxant effect of adenosine was depressed by removal of the endothelium. In endothelium-denuded preparations, the small relaxant effect of adenosine remaining was greatly potentiated. 5. Haemolysate (1 microliter ml-1) caused an endothelium-dependent contraction under oxygenated conditions. The hypoxic contraction observed in the artery ring under resting tension was significantly potentiated by haemolysate (1 microliter ml-1). Haemolysate 1 microliter ml-1 had no effect on the denuded artery rings under hypoxic conditions. 6. Haemolysate (1 microliter ml-1) potentiated the vasoconstrictor effects of U46619 (0.5 microM), 5-HT (1 microM) and K+ (24 mM) under oxygenated conditions. 7. These results indicate that endothelium profoundly modifies the effect of hypoxia on the responsiveness of sheep isolated left circumflex coronary artery rings.

In vitro study of vascular endothelial injury by activated platelets and its prevention

We performed an in vitro study to assess injury to vascular endothelial cells by platelets. Cultured endothelial cells isolated from fetal bovine aorta were used. Addition of human platelets, activated by collagen or lysed by sonication, to the culture dish resulted in dose- and time-dependent damage to the cells as estimated by [3H]adenine release. Analysis of [3H]adenine nucleotides by thin-layer chromatography on PEI-cellulose revealed decreased intracellular ATP content in the cells treated with platelet lysate. The medium contained AMP and adenosine, the latter increasing following the treatment of the cells. Of the substances released by the activated platelets, thromboxane A2 (TxA2) and serotonin caused cell damage. Platelet-derived growth factor (PDGF), however, did not damage the endothelial cells up to a concentration of 200 ng/ml. Pretreatment of the cells with methysergide (10(-6) M) or ONO 3708 (10(-5) M), a TxA2 antagonist, only partially prevented the damage, while ZK 36374 (10(-6) M), a prostacyclin analog, and 3-isobutyl-1-methylxanthine (IBMX; 10(-3) M), a phosphodiesterase inhibitor, potently inhibited injury. We conclude that the substances released from activated platelets may injure endothelial cells in an additive or synergistic manner and that agents which produce effects that elevate cyclic AMP levels may protect the cells from damage induced by the platelets.