Abstract Body: Introduction: Ischemic stroke is highly heterogeneous, with patient-to-patient differences in infarct location, severity, and degree of reperfusion, among other factors. Incorporating this variability in preclinical stroke models is useful to more comprehensively recapitulate the patient population. Prior studies employ Longa (complete reperfusion) or Koizumi (chronic hyporeperfusion) murine models of stroke interchangeably over a range of ischemic durations. These models represent distinct stroke phenotypes, but the unique features of each model warrant further definition. Therefore, we sought to differentiate the pathology of these models to more accurately model patient heterogeneity.
Methods: Transgenic mice expressing a fluorescent neutrophil marker (Ly6G-TdTomato) were subject to the Longa or Koizumi temporary middle cerebral artery occlusion (tMCAO) model of ischemic stroke. Ischemic duration was varied to model strokes of mild (30 min), moderate (60 min) and high (90 min) severity (n = 6-10 per experimental model). Laser speckle contrast imaging (LSCI) was performed at baseline, prior to and after reperfusion, and at endpoint to quantify cerebral blood flow (CBF) using a custom-made pixel assignment algorithm. Brains were collected at 24h or 72h and imaged via confocal microscopy to evaluate neutrophil infiltration in the ischemic hemisphere.
Results: Mortality significantly increased with ischemic duration in the Koizumi model at 24h, 48h, and 72h, but did not differ across the Longa models. LSCI demonstrated a 2-to-5-fold increase in the area of profound ischemia (lowest quintile of pixel values) in the Koizumi model compared to the Longa model and with increasing ischemic duration. Additionally, the Koizumi model exhibited less restoration of CBF following reperfusion and more variability in CBF within the ischemic hemisphere regardless of ischemic duration. The number of infiltrating neutrophils increased 10-fold between the 30- and 90-min Longa and Koizumi models, but the Koizumi model demonstrated increased variability in the number of neutrophils at 24h and 72h.
Conclusions: Our findings demonstrate differences in mortality, CBF, and immune response between stroke models of varying ischemic duration and reperfusion status. These unique pathological features can be linked to the physiological perturbations defining each model to establish a phenotypic spectrum that can be exploited to more thoroughly represent human stroke patient heterogeneity.