Short-term and long-term role of platelet activating factor as a mediator of in vivo platelet aggregation.

BACKGROUND

Platelet activating factor (PAF) is a phospholipid released upon stimulation by a variety of cells and has been implicated in several pathophysiological events such as asthma and inflammatory diseases. However, although the ability to aggregate platelets in vitro was the first biological activity ascribed to PAF, its role in contributing to the in vivo formation of arterial thrombi has not been thoroughly clarified.

METHODS AND RESULTS

Intravascular platelet aggregation was initiated in two different animal models of arterial stenosis and endothelial injury. An external constrictor was positioned around rabbit carotid arteries and canine coronary arteries. After placement of the constrictor, a typical pattern of flow developed in the stenotic vessels. This pattern of flow, characterized by progressive reductions of carotid or coronary blood flow followed by spontaneous or induced restorations of flow (cyclic flow variations, CFVs), is related to recurrent platelet aggregation at the site of the stenosis followed by dislodgment of the thrombus. After observing CFVs for 30 minutes, BN52021 (up to 1.2 mg/kg), a potent and selective PAF antagonist, was given intravenously to rabbits (n = 12) and dogs (n = 10). BN52021 completely inhibited CFVs in 10 of 12 rabbits, whereas it was relatively ineffective in abolishing CFVs in dogs (only 2 of 10 animals inhibited). This different effect of BN52021 was not explained by too small a dose of the drug to achieve a complete blockade of PAF receptors in dogs, since ex vivo platelet aggregation was completely inhibited in both rabbits and dogs in response to exogenous PAF at concentrations up to 10(-5) mol/L. In a second group of 10 dogs, the hypothesis that PAF may become an important mediator of CFVs in dogs only several hours after endothelial injury was tested. After 30 minutes of baseline CFVs, these animals received a bolus of BN52021 up to 1.2 mg/kg. After this treatment, CFVs were completely abolished in 2 of 10 animals. The remaining 8 dogs were followed for an additional 8-hour period, at the end of which a second bolus of BN52021 was given. At this time, BN52021 was effective, as CFVs were abolished in 6 of 8 animals. These effects of BN52021 at 8 hours were not the consequence of a cumulative dose of the compound, since ex vivo platelet aggregation in response to PAF returned to baseline values immediately before administering the second dose. To identify possible sources of PAF other than aggregating platelets at the site of arterial stenosis, dogs in a third group were killed after 30 minutes (n = 7) and after 8 hours (n = 8) of CFVs. Histological sections of the stenotic coronary artery showed a marked leukocyte infiltration in these arterial segments after 8 hours of CFVs, whereas sections from dogs killed after 30 minutes showed only moderate or no infiltration.

CONCLUSIONS

These data demonstrate that PAF plays an important role as a mediator of platelet aggregation in vivo in rabbits and dogs. In the canine model, PAF appears to become more important after leukocyte infiltration of the arterial wall, as it may contribute to initiating enough platelet activation to lead to cyclic flow variations at sites of arterial stenosis and endothelial injury. Data from the present study suggest that PAF antagonists may be used as antiplatelet agents.