Fig. 1

Delayed administration of rt-PA in mouse stroke induces intracerebral haemorrhage which does not correlate with neurological deficit. Human recombinant t-PA (rt-PA; 10 mg/kg) or its HEPES vehicle were administered to mice after sham operation (n = 5 or 6, respectively) or post middle cerebral artery occlusion (MCAo) for 0.5 (n = 7 each group) or 4 h (n = 8 or 7, respectively). a, b No changes in brain albumin (a) and haemoglobin levels (b) were observed between rt-PA and vehicle in sham-operated mice at 24 h after surgery. c Brain albumin sharply increased in the ipsilateral hemisphere 24 post 4 h MCAo compared to 0.5 h MCAo, yet no differences were detected with rt-PA compared to vehicle treatment in both time points. d rt-PA increased brain haemoglobin in the ipsilateral hemisphere (indicating development of intracerebral haemorrhage by 24 h) only when injected at 4 h, but not 0.5 h, post MCAo. e Representative images of the ipsilateral hemisphere 24 h after vehicle or rt-PA treatment post 4 h MCAo. Accumulation of blood in the stroke-affected brain and formation of multiple petechial haemorrhagic infarctions are apparent with rt-PA. f Correlations between brain haemoglobin levels measured in the 4 h stroke cohort (black up-pointing triangle vehicle- and white circle rt-PA-treated) and functional tests performed, including (from left to right) neurological scoring, hanging wire test and ANY-Maze recordings over 5 min of distance travelled and immobility periods. Brain haemoglobin does not correlate with any functional parameter. Data in a–d is shown as individual animals with mean ± SEM. n = 5–8 (see detailed n numbers above). Statistical analysis in a and b by student t-test; in c and d *P < 0.05, ****P < 0.0001 by 2-way ANOVA with Sidak post hoc; in f by Spearman correlation for neurological score or Pearson correlation for all other tests. In panels a–d Outliers are denoted in black symbols and excluded from the analysis