The triton X-100-insoluble residue ('cytoskeleton') of aggregated platelets contains increased lipid phosphorus as well as125I-labeled glycoproteins

Abstract

When125I-labeled blood platelets were lysed with an equal volume of 2% Triton X-100, 10 mM EGTA, and 100 mM Tris, pH 7.4, the Triton-insoluble residue ('cytoskeleton') contained about 11% of the platelet protein, predominantly actin and actin-binding protein, 3.5% of the platelet radioactivity, and less than 2% of the platelet lipid phosphorus. Radioautographs of polyacrylamide gels of this residue (reduced samples) differed markedly from those of whole platelets. Glycoprotein IIb (mol wt 121,000) was not detected and GP III (mol wt 107,000) was seen in trace amounts if at all. The amounts of the other labeled glycoproteins in the Triton-insoluble residue were estimated by scanning the radioautographs and comparing the area for each band with that of the similar band in whole platelet samples. In 11 experiments, 38% of GP IV (mol wt 89,000), 8% of the GP at mol wt 132,000 and 56% of the GP at mol wt 158,000 were insoluble in Triton. Since no glycoproteins were evident between about mol wt 90,000 and 170,000 in unreduced samples of these residues, the identity of the most abundant Triton-insoluble glycoprotein was unclear. In platelets stimulated with thrombin, Triton-insoluble protein increased to 16.7%, due at least in part to the formation of fibrin from platelet fibrinogen: it did not increase in platelets stimulated with ADP. The amount of Triton-insoluble radioactivity and lipid phosphorus, and the gel pattern of Triton-insoluble radioactive proteins was unchanged in platelets that were stimulated but not aggregated with thrombin or ADP. However, if the platelets were shaken with thrombin to form large aggregates, or were centrifugal before they were lysed, Triton-insoluble radioactivity and lipid phosphorus increased markedly, and the radioautographs resembled those of whole platelets. Electron microscopic examination revealed fragments of membrane, and the increase in radioactivity was ascribed to inadequate membrane lysis. When small aggregates were produced by shaking the platelets with ADP in the presence of added fibrinogen or incubating them with thrombin for 30 min without shaking, Triton-insoluble radioactivity increased in some experiments and when it did, GP IIb and III were evident in radioautographs. Lipid phosphorus always increased slightly. In striking contrast, platelets stimulated with concanavalin A showed a 2-12-fold increase in Triton-insoluble lipid phosphorus. We conclude that additional radiolabeled surface proteins become associated with the Triton-insoluble 'cytoskeleton' of platelets stimulated with concanavalin A, but that the apparently similar increases seen in platelets aggregated by the physiologic stimuli ADP or thrombin result from inadequate membrane lysis.