Search

In the present paper, a standardized method for the induction and registration of platelet thrombi in arterioles (500 microns diameter) of small laboratory animals is described in full detail. Using an optoelectronic analogue computer device, different discriminating parameters characteristic for the thrombotic phenomenon are presented. As the topical application of exogenous PAF-acether induces the generation of endogenous PAF-acether according to previous investigations (Bourgain et al. (1985) Prostaglandins 30, 185) it was deemed interesting to investigate the effect of hydrogen peroxide using the described methodology. It was found that the latter substance not only primes the effect of PAF-acether-induced thrombosis, but also can trigger by itself PAF-acether modulated arterial thrombus formation. Experimental evidence is adduced that these thrombotic phenomena can be most efficiently down regulated by specific PAF-acether antagonists.

The in vitro effect of nifedipine and atenolol, either alone or in combination, on the proliferation and migration of rat aortic smooth muscle cells was investigated. Nifedipine inhibited the replication of arterial myocytes in concentrations ranging between 10 and 100 microM. The inhibition, evaluated as cell number, was dose- and time-dependent with an IC50 of 39 and 34 microM after 48 and 72 h, respectively; the cell doubling time increased with drug concentrations up to 118 h versus 28 h for controls. Atenolol alone failed to reduce arterial myocyte proliferation, and did not influence the effect of nifedipine on cell proliferation. Nifedipine and atenolol alone inhibited in a dose-dependent manner rat aortic myocytes migration induced by fibrinogen as chemotactic agent. When the combination nifedipine-atenolol was investigated, an additive inhibitory effect on cell migration was observed. These results provide in vitro support for a potential effect of this drug association on early steps of atherogenesis.

Platelet-vessel wall interactions and local thrombosis are investigated in vivo in a branch of the mesenteric artery of the guinea pig, using optoelectronic registration and ultrastructural control. Following an electrical challenge resulting in changes of cell membrane polarization, subsequent superfusion by PAF-acether or a stable analogue, (1-O-alkyl-2-N-methylcarbamyl-sn-glycero-3-phosphocholine, 10(-8) M focal concentration (f.c.)) for a restricted period results in endothelial cell retraction and bleb formation followed by platelet adhesion and the development of a thrombus which over time becomes invaded by leukocytes and eventually occludes the vascular lumen. It was demonstrated in a previous investigation that these phenomena are triggered by the generation of endogenous PAF-acether by the endothelial cells. Specific PAF-acether-antagonists, such as BN 52021 a ginkgolide, but also synthetic molecules, derivatives of the triazolo-pyridino-diazepine group (BN 50727, BN 50755 and BN 50789), significantly inhibit platelet-vessel wall interactions and thrombosis, but not the formation of blebs in the endothelial cells. Hydrogen peroxide (10(-5)M f.c.) not only primes the effect of PAF-acether, but is by itself capable of inducing thrombosis through a PAF-acether-mediated mechanism. Inhibition of acetyl hydrolase by PMSF (phenyl-methyl-sulfonyl-fluoride, 10(-5)M f.c.) invariably results in a significant enhancement of thrombosis, while conversely, inhibition of acetyl transferase by 27584 RP (4-(naphtylvinyl)pyridine hydrochloride, 10(-6)M f.c.) inhibits thromboformation indicating that the remodeling pathway is involved.

In rats treated with acetylsalicylic acid (100 mg/kg i.v., -3 h) to block platelet cyclo-oxygenase, a superfusion with tranylcypromine (TCP, 3 x 10(-3) M) plus arachidonic acid (AA; 1 x 10(-4) M) significantly enhances (+70%) the opto-electronically monitored formation of a thrombus, elicited by a superfusion with ADP (4 x 10(-4) M for 45 s) over a de-endothelialized site of a mesenteric artery (200-300 microns diameter). Treatment with ridogrel at doses blocking either singly thromboxane A2 (TxA2) synthetase (0.63 mg/kg i.v., -15 min, n = 6) or additionally TxA2/prostaglandin endoperoxide receptors (5 mg/kg i.v., -15 min, n = 6) reduces (-74%) or abolishes (-100%) respectively the enhancement by TCP + AA of the ADP-induced thrombus formation. These observations demonstrate (1) a transfer of prostaglandin endoperoxides from the vessel wall to the platelets to reinforce thrombus formation in vivo; (2) the antithrombotic potential of combined TxA2 synthetase inhibition plus TxA2/prostaglandin endoperoxide receptor blockade with ridogrel.

Using an optoelectronic device, we demonstrated that endotoxin modulates platelet-activating factor (PAF)-induced arterial thrombosis in a mesenteric artery of the guinea-pig. Indeed, topical superfusion by PAF induces the generation of platelet-vessel wall interactions, invariably followed by thrombus formation. A low dose of endotoxin markedly increased, while a high dose significantly reduced thrombus formation. The reducing effect of the high dose can be completely offset by indometacin, indicating the involvement of prostaglandin I2.

A method for "in vivo" induction and registration of arterial platelet thrombosis has been developed and standardized in a branch of the mesenteric artery of the white Wistar rat. It consists in local deendothelialization by electrical current. Thrombus induction is performed by topical superfusion with ADP; when the superfusion is discontinued the thrombus disappears but can again be induced reproducibly by renewal of the ADP superfusion after a time interval not exceeding 15 minutes. Registration of the thrombotic phenomenon is made possible by microprojection of the investigated arterial segment. An appropriate optoelectronic device allows the on-line derivation of discriminating parameters. Off-line storage and processing of the experimental data by computer is provided. The method is automatized and can easily be applied to other species of laboratory animals.

Cicletanine, a drug which affects membrane ion transport, induces a marked increase of the liberation of PGI2 as demonstrated by the increase of the stable metabolites in the plasma following intravenous administration of arachidonic acid. Furthermore, the inhibiting effect of tranylcypromine on prostacyclin synthetase is completely removed by this pharmacon. These observations are suggestive that this drug presents a scope for treatment of thrombotic disorders as well as hypertension.

A technique for induction and on-line quantification of local platelet thrombi in mesenteric arteries of small laboratory animals was developed and standardized in our laboratory. In the past, this model was used to study the nature of platelet-vessel wall interaction in the living animal. The ultrastructure of the experimental intimal lesion and the vessel wall regeneration were assessed by transmission electron microscopy (TEM), both in normal and pathologic conditions. Scanning electron microscopy (SEM) now shows the ultramorphology of platelet thrombi on the experimentally injured arterial segment following topical superfusion with ADP, mepacrine or platelet-activating factor (PAF). The application of these substances, each with proper bio-activity, leads to distinct types of platelet thrombi. Mepacrine or PAF superfusion causes large thrombotic masses, as compared to control, ADP induced thrombi, and seems toxic for the endothelial cells. Mepacrine thrombi differ significantly from PAF thrombi in their platelet density, degree of platelet activation and in their relation to the endothelium that surrounds the experimental lesion. Furthermore, PAF superfusion induces a phenomenon of spontaneous regeneration of the thrombus after its forced embolization. This is probably due to some unknown bio-action of PAF in the vessel wall.

Prostaglandins play an important role in the platelet-vessel wall interaction. Acetylsalicylic acid inhibits the cyclooxygenase activity within the vessel wall. Continuous superfusion with acetylsalicylic acid induces a rapid, but evanescent decrease in thromboformation. The trigger mechanism involved could be related to the ratio of cyclic endoperoxides (PGG2 and PGH2) versus prostacyclin (PGI2).

In order to obtain a continuous measurement of local blood flow in the cerebral cortex of a laboratory animal using chronically implanted sensors, we have developed a device based on the heat clearance principle. Flow information is obtained from temperature measurement by means of two thermistors one of them being heated at a defined level above ambient tissue temperature; as such, cooling of the heated thermistor caused by convection phenomena in its vicinity, can be related to local perfusion rate. In a first step "in vitro" measurements were performed in order to study the behaviour, sensitivity and reliability of the device; a physical model was established explaining the results. In this paper we describe "in vivo" tests in the rabbit's brain cortex with the miniature thermistors (0.5 mm diameter) chronically implanted (at the cortical surface). Results are correlated with oxygen tension measurements using (smaller) pO2 electrodes inserted into the cortical tissue. We have observed that all sensors are well tolerated by the animals who remain symptom free. Test experiments, inducing a well known physiological effect on local blood flow, such as arterial clamping, inhalation of CO2 gas mixtures, etc., are performed. The phenomena during induced anoxic anoxia are also shown. These preliminary investigations are essential in order to attempt by future experiments the establishment of a correlation between "in vivo" recorded flow signals and the "in vitro" measured characteristics.

By comparison with the previously described results of similar experiments performed on normal rats (Potvliege and Bourgain, 1976, 1980), it was observed that with a fat-rich diet the endothelium contained a greater number of Weibel-Palade bodies and that there was an increased tendency for blood monocytes to stick to the endothelium as well as for platelets to form, small, non-adherent aggregates particularly at sites of arterial branching opposite intimal cushions. The intimal cushions were larger and they reacted to local thrombogenic treatment by greater and more prolonged myointimal growth. In addition, although their endothelial lining did not become detached, individual cells showed signs of degeneration. These changes are interpreted as indicating that a fat-rich diet enhances atherogenesis at points of greater haemodynamic stress by locally increasing intraluminal platelet aggregation and endothelial-cell degeneration. These 2 ever-recurring events favour, in the former case, a slow but steady liberation of platelet factors, and, in the latter, an easier penetration into the artery wall of mitogenic substances which stimulate myointimal growth. The adhesion of monocytes to the endothelium may represent a first step towards their migration into the intima where, at a later stage, they may become transformed into foam cells, thus completing the atherogenic cycle.

The effect of cicletanine, an antihypertensive drug, was studied in two in vivo models. It was demonstrated that the drug markedly affected the response of intravenously administered arachidonic acid in the rabbit. In this model a marked increase of 6-oxo-PGF1 alpha was observed when challenged by the intravenous injection of 50 micrograms/kg b.w. of arachidonic acid. In the rat model used for the study of platelet-vessel wall interaction, it was demonstrated that the inhibition of prostacyclin synthetase could be offset by cicletanine. These results indicate that the drug modulates the generation of prostacyclin and as such is capable of affecting the peripheral resistances which determine the level of the blood pressure.

The continuous superfusion of arachidonic acid over a branch of the mesenteric artery in an animal in-vivo model results in an initial increase followed by a decrease to control values of ADP induced platelet thrombogenesis. This phenomenon is observed in normal rats as well as in rats submitted to the intravenous administration of ASA three hours prior to the investigation. In the latter group the cyclooxygenase of the platelets is permanently affected while the cyclooxygenase of the endothelial cells is regenerated at the time the investigation is started. The tachyphylactic-like phenomenon observed with arachidonic acid is probably related to the dynamics of the enzymatic reactions of the arachidonate cascade involving the platelets and the endothelial cells surrounding the local de-endothelialized area.

Superfusion of PAF-acether over a branch of the mesenteric artery in the guinea pig invariably results in local endothelial injury and thrombus formation within 3-10 minutes. The thrombotic phenomena do not disappear when PAF-acether superfusion is discontinued, and even when forced embolization is induced. Within a very short interval renewal of thrombosis occurs at the same site. Several data point to a mechanism involving generation and release of endogenic PAF-acether. Recents findings on PAF-acether release by cultured endothelial cells indicate that in the in vivo situation this phenomenon could well be responsible for maintaining the thrombotic status as demonstrated by ultrastructural analysis. In a later stadium polymorphonuclear leukocytes are also involved in total thromboformation.

In order to quantify the effect of specific drugs on the cerebral cortex an "in vivo" model has been developed for the induction and the observation of anoxic anoxia. Rabbits are used as test animals. Sensors for the assessment of local parameters are chronically implanted: ECoG electrodes are applied; pO2 electrodes are inserted into the cortical tissue. The derived somatosensory evoked potentials are used for evaluating the cerebral cortical function. Animals are cannulated with a tracheal tube, curarized and artificially ventilated. Anoxic anoxia, controlled by a special purpose microprocessor system, may then be induced and repeated in a reproducible way. During the experiments local (pO2, ECoG, SEP) as well as general parameters (ECG, heart rate, systemic blood pressure, rectal temperature) are recorded and stored on analog magnetic tape as well as digitized with the microprocessor system. The cortical pO2 is measured with a polarographic method, the SEP's are obtained on-line by time coherent averaging and the ECoG states (e.g. epilepsy during anoxia) are derived by using band-pass filters and rms detectors. Off-line the signals (pO2, heart rate, mean systemic blood pressure, temperature) are standardized and represented together with parameters derived from ECoG and SEP. SEP-waveform parameters indicating intensity (norm) and similarity with a reference SEP signal (correlation value) are used. The measuring and processing method is still being optimized; special attention is being paid towards the quality of the calculated SEP's which are to be used for the quantification of the cortical function during reference, anoxia and recovery period. As such, in order to improve the signal-to-noise ratio of the SEP's and, consequently, of the derived parameters, ECoG signals are digitized off-line and subjected to a preprocessing, implying filtering and spectral analysis procedures.

An in vivo animal model was developed in the rabbit for the study of the mechanisms involved in the generation and release of prostanoids. Following heparinization and, if required, further sensitization of the animals by intravenous administration of haemolysed blood, injection of doses of arachidonic acid not exceeding 180 micrograms/kg induced a marked fall in arterial blood pressure on condition that the plasma anti-inflammatory protein levels were within the normal range. Cicletanine, certain diuretics (furosemide and bumetanide), as well as calcium-entry blockers such as verapamil and the association of insulin and potassium ions, all markedly decreased the AA50 value and were accompanied by a significant increase in the plasma levels of 6-oxo-PGF1 alpha, enhancing as such the ratio of 6-oxo-PGF1 alpha versus TXB2 in plasma. The infusion of insulin in association with potassium ions induced a similar but less sustained effect. Drugs which affect membrane ion transport were investigated in relation to an enhancing effect on the generation and release of prostanoids following the administration of arachidonic acid.

Diabetes was induced in 24 adult male Wistar rats by a single intravenous injection of streptozotocin. The ultramorphology of the coronary arterial wall was studied by scanning and transmission electron microscopy up to 10 weeks following this diabetes induction. At that moment, several ultrastructural characteristics of primary atherogenesis could be observed, i.e. important myointimal thickening with extensive medial smooth muscle cell migration towards and into the intima, and synthesis of an abundant mass of intercellular ground substance in the media and in subendothelial areas. Lipid deposition or foam cell formation was not observed. These results confirm that experimental diabetes is a risk factor for the coronary arterial wall as it induces typical atherogenic phenomena.

The effect of intravenously administered defibrase was studied on the white intra-arterial thrombus. The administration resulted in all experimented animals in a statistically significant decrease of the thrombus formation.

Arteries taken from 1 to 3 days after local thrombosis had been induced by the passage of a weak electric current followed by the topical administration of ADP were studied by electronmicroscopy. The characteristic changes observed during this period included total disappearance of pre-existing thrombi, rapid reconstitution of the endothelium and important hypertrophy of smooth muscle cells in the media. Thrombocytopenia induced by the administration of antiplatelet serum produced no alterations in untreated arteries. When it was coupled to thrombus induction, healing of injured endothelial and smooth muscle cells was grossly impaired, hypertrophy of the media was absent and fibrin infiltrated the arterial wall at the site of thrombus induction. These results bring support to the view that blood platelets play an important role in the induction of the regeneration response of injured arterial cells.

The involvement of arterial smooth muscle cells in the development of atherogenic lesions following de-endothelialization and platelet-vessel wall interaction was described in detail by Ross et al. (1977). Bourgain & Six (1974) described a method for local de-endothelialization over a small area in a branch of the mesenteric artery of the male white Wistar rat. The vessel wall reaction to the endothelial cell loss was investigated in detail by Potvliege & Bourgain (1976). The reactive pattern following de-endothelialization includes both a marked hypertrophy of the smooth muscle cells, and, if induced at the site of bifurcation, is further accompanied by migration of smooth muscle cells into the subintimal layer (Potvliege & Bourgain 1980). Administration of a fat- and cholesteral-rich diet markedly increased these phenomena (Potvliege & Bourgain 1982).

Intimal cushions consist of small bundles of longitudinally oriented smooth-muscle cells (SMC) which are interposed between the endothelium and the lamina elastica interna (LEI) of arteries. They are consistently present on the mouths of artherial branchings, at sites of haemodynamic stress. In structure and location, they closely resemble early artherosclerotic lesions. In previous experiments, we studied the local consequences of a standardized microthrombosis induced in non-branching segments of mesenteric arteries of rats. In this experimental model, the endothelium gets reconstituted within 24 h while in the underlying media the SMC exhibit a marked growth reaction which reaches a peak on the third day and eventually terminates by the sixth day with full restitution to normal. In the present work, the experiment was repeated at sites of arterial branchings. It was seen that in this location the SMC growth reaction was greatly enhanced and led to a marked increase in size of the intimal cushions. In addition their endothelial lining, although reconstituted, showed a persistent tendency to segmental breakdown. These results support the view that intimal cushions develop as a consequence of local haemodynamic stress which, by causing endothelial-cell loss, triggers off the sequence of blood-platelet adhesion, aggregation and release of mitogenic factors, the latter being ultimately responsible for an ever-recurring stimulation of myointimal growth. The same mechanisms are likely to operate in atherogenesis.

By appropriate experimental procedures it was possible to distinguish the components of the contra- and ipsilateral cerebral somatosensory potentials evoked by electrical stimulation of an extremity in the unrestrained rabbit. The components of the contralateral response occur in the following order; an early positive primary wave (P1), a positive associative wave (P2), and ultimately a late negative N wave. The components of the ipsilateral response occur in the following order; a small positive wave P, the latency of which is intermediate between those of the contralateral P1 and P2 waves, and a large negative wave N, similar to the contralateral N wave. The different topographical distributions of these waves were elucidated by the use of insulated, chronically implanted electrodes glued onto the cortical surface. The properties of the waveform components were studied by various methods such as varying the stimulation parameters, simultaneous application of somesthetic and acoustic stimuli, and administration of narcotic drugs. The properties of P1 were similar to those of P28 in humans; the properties of P2 can be compared to those of P45; and, finally, the N wave resembles the late negative components observed in humans. Inconstant small positive waves of shorter latencies, which will be discussed in a following paper, may also be seen. Interestingly enough, no early negative wave such as that observed in humans (N20) was ever found. If, as is presently thought, this wave is, in fact, due to the folding of the cortical surface (Broughton, 1969), its absence is to be expected in the rabbit because the cortical surface of this species is lissencephalic and thus devoid of gyri.

For several years, an in vivo model for the induction and on-line quantification of arterial platelet thrombosis in mesenteric arteries of a small laboratory animal species has been developed in our laboratory. In the present paper, we further document the intimal lesions and the ADP superfusion-induced local platelet thrombus as seen in the scanning electron microscope. The surface morphology of the intimal lesion, induced by electric current, shows a circular or slightly oval denuded area, affecting about 15-20 endothelial cells. The edge of this lesion is often occupied by partially disrupted and detached endothelial cells. The successive embolizations of several ADP thrombi clean this edge and augment the denuded area. The final lesion never exceeds the area of 30-40 endothelial cells. ADP-induced platelet thrombi in invariably appear as loose, sponge-like platelet aggregates, very bloodstream-lined, anchored on the denuded subendothelium. There is an excellent correlation between the in vivo light microscopic observations and the actual ultrastructure of this platelet mass.

Topical superfusion by PAF-acether over a branch of the mesenteric artery of the guinea pig invariably results in the formation of a platelet thrombus, adhering onto a denuded area of the vessel wall, or at sites where local retraction of the endothelial cells has occurred. It is demonstrated that the application of exogenous PAF-acether triggers a process of local autogeneration and release of endogenous PAF-acether (or PAF-like substances), maintaining a long phase of sequential regrowth and embolization of thrombi, eventually leading to a complete occlusion of the arterial lumen. Extracts of Ginkgo biloba, administered intravenously or in topical superfusion, inhibit both exogenous- and endogenous-induced thromboformation.

In the course of this study it was found that the functional state of the cortex during anoxic anoxia could be more accurately monitored by the cortical somatosensory evoked potentials (EPs) than by the electrocorticogram (ECoG). Compared with the ECoG, the EPs yield more detailed information. Five different steps may be distinguished in the evolution of EPs; the ECoG, however, exhibits only two (almost trivial) abnormal patterns (epileptic pattern and flat response). Of particular practical interest are the two steps preceding the disappearance of the EPs: (1) a rounding of the P wave accompanied by an increased voltage, followed later by (2) a decrease of the latter. If, at this stage, anoxia is maintained for 20 seconds, then permanent dysfunction is observed, at least during the time interval of the experimentation.

The administration of arachidonic acid to live rabbits if followed by the generation of prostacyclin and/or thromboxane. Cicletanine increased the production of prostacyclin in a first group of rabbits and amplified the prostacyclin/thromboxane ratio in a second group of the thromboxane type. The most probable mechanism for this action is activation of prostacyclin synthase by cicletanine. This was confirmed in the in vivo model by a study of the platelet-vascular wall interaction: tranylcyprominE, a prostacyclin synthase inhibitor, increased the interaction. Under these experimental conditions, cicletanine inhibited the effect of tranylcypromine and completely restored the enzymatic activity of prostacyclin synthase.

Prostaglandins play an important role in platelet-vessel wall interaction. The inhibition of PGI2 synthetase by 15-hydroxyarachidonic acid (15-HPAA) results in an enhancement of local thrombus formation.

Prostaglandins play an important role in the platelet-vessel wall interaction. The inhibition of PGI2 synthetase results in an increase of platelet thrombosis induced by adenosine diphosphate. Addition of exogenous arachidonic acid further increases the phenomenon. The ratio of cyclic endoperoxides (PGG2 and PGH2) to prostacyclin (PGI2) could well be the determining trigger in platelet-vessel wall interaction and local thrombosis.

The catastrophe theory evolved by Thom and Zeeman proposes a mathematical definition for the abrupt or 'catastrophic' changes that can suddenly occur in normally well-ordered and smooth-running systems. We have integrated this theory with our own PAF/cytokine feedback network hypothesis to explain the control and dysfunction of the inflammatory response. This process involves the activation of cells and factors such as proteases, and is coordinated by mediators such as PAF, cytokines and growth factors, minute amounts of which can prime cells to respond in an enhanced manner to subsequent agonistic stimuli. PAF and certain cytokines also possess the unique property of being able to induce the release of each other and their own generation in vivo. This 'singularity' may enable a self-generating feedback network to become established. The priming ability of these mediators indicates the extreme sensitivity of the inflammatory process and importance of a homeostatic equilibrium between the vectors involved in the priming and feedback processes and internal suppressive mechanisms. In pathological conditions, one can consider the phenomenon of PAF and cytokine autogeneration as a 'fold' in the feedback network and an expression of the singularity characteristic of the catastrophe hypothesis. This may lead to systemic toxicity and microcirculatory collapse, a characteristic feature of shock, sepsis, asthma, ischemia and graft rejection. A combination of drugs antagonizing the various feedback components may inhibit this catastrophic process and thus provide more successful therapy of these conditions.

Arterial platelet thrombosis was induced in vivo in a branch of the mesenteric artery of the white Wistar rat by topical superfusion by adenosine diphosphate following local de-endothelialization of the dissected segment by means of a small electrical current. Detection of the thrombotic phenomena was performed by projection of the arterial segment onto a set of light sensitive elements, which allows the registration of several discriminating parameters. The addition of arachidonic acid to the superfusing mixture results in an increase in thromboformation; this increasing effect can be blocked completely by prior inhibition of the vessel wall cyclooxygenase activity by inhibitors such as flurbiprofen or indomethacin. We therefore concluded that the arachidonate cascade, triggered within the endothelial cells, is involved in platelet-vessel wall interaction leading ultimately to local thrombosis.

As our experimental set-up for continuous recording of local blood flow in the cerebral cortex of a laboratory animal with chronically implanted miniature thermistors (based on the heat clearance principle) gave satisfactory results for routine tests of pharmacological agents during anoxia, hypoxia, hypercapnia, etc., experiments, we intended to standardize the apparatus, to increase accuracy, to facilitate calibration and to enhance flexibility with respect to the operator. The exponential aspect in the thermistor resistance/temperature characteristic is linearized by applying a logarithmic converter in the thermistor amplifier. Calibration to the centigrade temperature scale is performed by a three digit numerical adaptation of two thermistor constants determined in a thermostatic-cryostatic bath (zero and slope). A heating power measuring circuit is provided so that the dissipation constant of the thermistor implanted in tissue can be obtained and the thermal conductivity of the tissue can be estimated. Linearity of the relation between cooling of the heated thermistor and local flow, for small cooling values as they are registered in vivo, is still being investigated.

The effect of a topical paf-acether superfusion over an injured arterial segment was assessed in the guinea-pig, using an opto-electronic in vivo thrombosis model allowing on-line quantification of small platelet thrombus dynamics. As compared to control, ADP-induced, thromboformation and behaviour, exogenous paf-acether causes a large, dense platelet thrombus, invaded and surrounded by numerous leukocytes, spreading widely over the adjoining, vacuolized, endothelium. Its embolization has to be forced with prostanoids, mepacrine, EDTA, or with a specific paf-acether antagonist (BN 52021). A few minutes after such forced embolization, a new thrombus starts growing at the same site, without renewal of the paf-acether superfusion. This phenomenon of spontaneous reappearance after forced embolization can be followed during several hours. Experiments with labelled paf-acether and the paf-acether antagonist indicate a possible endogenous paf-acether (or paf-acether-like) production triggered by superfusion with exogenous paf-acether.

Flurbiprofen, a cyclooxygenase inhibitor, and tranylcypromine, a PGI2 synthetase inhibitor, in topical superfusion over an arterial segment respectively decrease and increase ADP-induced local white platelet thrombus formation. Superfusion of the arterial segment with PGI2 demonstrates a dose-response related inhibition of ADP-induced thrombosis.