Tissue plasminogen activator (tPA) is an effective treatment for stroke, but its utility is limited by fear of cerebral hemorrhage. Tenecteplase (TNK), a genetically modified form of wild-type tPA, exhibits a longer biological half-life and greater fibrin specificity, features that could lead to fewer cerebral hemorrhages than wild-type tPA in stroke patients. We injected radiolabeled blood clots into the cerebral circulation of New Zealand White rabbits. One hour later, we administered tPA (n=57), 0.6 mg/kg TNK (n=43), 1.5 mg/kg TNK (n=27), or vehicle control (n=37). A blinded observer examined the brains for macroscopic hemorrhage using a semiquantitative score. We estimated thrombolysis by assessing the amount of radiolabel remaining in the cerebral vessels postmortem. Both wild-type tPA and TNK caused thrombolysis in most subjects. Hemorrhage was detected in 26% (6/23) of the control group, 66% (27/41) of the wild-type tPA group, 55% (16/29) in the 0.6-mg/kg TNK group, and 53% (9/17) in the 1.5-mg/kg TNK group (P:<0.05, chi(2) test). The tPA group was statistically significantly different from the control group, but the TNK and tPA groups did not differ from each other. Neither TNK nor tPA affected the size of the hemorrhages. TNK shows comparable rates of recanalization compared with wild-type tPA in a model of embolic stroke. While tPA increases hemorrhage rate, the hemorrhage associated with TNK treatment is not statistically different compared with controls or the tPA group. These findings suggest that TNK shows promise as an alternative thrombolytic treatment for stroke, but we could not demonstrate improved safety compared with wild-type tPA.