Use of a spectrophotometric hemoglobin assay to objectively quantify intracerebral hemorrhage in mice

There is great interest in developing novel anticoagulant or thrombolytic strategies to treat ischemic stroke. However, at present there are limited means to accurately assess the hemorrhagic potential of these agents. The present studies were designed to develop and validate a method to accurately quantify the degree of intracerebral hemorrhage (ICH) in murine models. In a murine model, ICH was induced by stereotaxic intraparenchymal infusion of collagenase B alone (6 x 10(-6) U; n = 5) or collagenase B followed by intravenous recombinant tissue plasminogen activator (rt-PA) (0.1 mg/kg; n = 6). Controls consisted of either sham surgery with stereotaxic infusion of saline (n = 5) or untreated animals (n = 5). ICH was (1) graded by a scale based on maximal hemorrhage diameter on coronal sections and (2) quantified by a spectrophotometric assay measuring cyanomethemoglobin in chemically reduced extracts of homogenized murine brain. This spectrophotometric assay was validated with the use of known quantities of hemoglobin or autologous blood added to a separate cohort of homogenized brains. With this assay, the degree of hemorrhage after focal middle cerebral artery occlusion/reperfusion was quantified in mice treated with postocclusion high-dose intravenous rt-PA (10 mg/kg; n = 11) and control mice subjected to stroke but treated with physiological saline solution (n = 9). Known quantities of hemoglobin or autologous blood added to fresh whole brain tissue homogenates showed a linear relationship between the amount added and optical density (OD) at the absorbance peak of cyanomethemoglobin (r = 1.00 and .98, respectively). When in vivo studies were performed to quantify experimentally induced ICH, animals receiving intracerebral infusion of collagenase B had significantly higher ODs than saline-infused controls (2.1-fold, increase; P = .05). In a middle cerebral artery occlusion and reperfusion model of stroke, administration of rt-PA after reperfusion increased the OD by 1.8-fold compared with animals that received physiological saline solution (P < .001). When the two methods of measuring ICH (visual score and OD) were compared, there was a linear correlation (r = .88). Additional experiments demonstrated that triphenyltetrazolium staining, which is commonly used to stain viable brain tissue, does not interfere with the spectrophotometric quantification of ICH. These data demonstrate that the spectrophotometric assay accurately and reliably quantifies muri