COMPARATIVE THROMBOLYTIC PROPERTIES OF TISSUE-TYPE PLASMINOGEN-ACTIVATOR AND OF A PLASMINOGEN-ACTIVATOR INHIBITOR-1-RESISTANT GLYCOSYLATIONVARIANT, IN A COMBINED ARTERIAL AND VENOUS THROMBOSIS MODEL IN THE DOG

rt-PA-K, a variant of recombinant tissue-type plasminogen activator (r t-PA) with substitution of amino acids 296 to 299 with alanine (KHRR29 6-299AAAA) has increased fibrin-specificity and reduced sensitivity to plasminogen activator inhibitor-1; rt-PA-T, with threonine 103 replac ed by asparagine has an additional glycosylation site and a reduced cl earance; and rt-PA-N, with asparagine 117 mutagenized to glutamine lac ks the high mannose carbohydrate side chain. We have investigated whet her combination of these properties in a single molecule might yield a n improved thrombolytic agent. The thrombolytic potency and fibrin-spe cificity of the combination mutant rt-PA-TNK was compared with that of rt-PA in a combined venous whole blood clot model and platelet-rich a rterial eversion graft thrombosis model in dogs given intravenous hepa rin and aspirin. Infusion of 0.125 to 1.0 mg/kg over 60 min in groups of 4 to 5 dogs produced dose-dependent fibrin-specific venous clot lys is. The thrombolytic potency (percent lysis per mg compound administer ed per kg body weight) of rt-PA-TNK was significantly higher than that of rt-PA as evidenced by a higher maximal rate of lysis of 480 +/- 10 0% versus 140 +/- 40% within the 2 h observation period per mg of comp ound administered per kg body weight (mean +/- SEM, p = 0.004) and a s ignificantly lower dose of 0.08 +/- 0.01 versus 0.21 +/- 0.04 mg/kg bo dy weight at which the maximal rate of lysis was obtained (p = 0.004). This higher thrombolytic potency was the result of a significantly re duced clearance (240 +/- 32 versus 540 +/- 49 ml/min, p = 0.002) and a similar specific thrombolytic activity (percent lysis per mu g/ml pla sma antigen level). Arterial reflow was obtained with 1 mg/kg rt-PA an d with 0.5 mg/kg rt-PA-TNK, but with each agent recanalization was con sistently associated with cyclic reocclusion and reflow. The frequency of arterial recanalization was somewhat higher with rt-PA-TNK (10/12) than with rt-PA (4/12) (p = 0.07) but the total patency times during a 2 h observation period were similar. Thus, rt-PA-TNK has an approxim ately 3-fold higher thrombolytic potency for whole blood clot lysis th an rt-PA. The activity ratio of rt-PA-TNK to rt-PA is, however, 2.5-fo ld higher in human in vitro plasma clot lysis systems than in canine s ystems, suggesting that the thrombolytic potency of rt-PA-TNK in man c ould be up to 8-fold higher than that of rt-PA.