The pericyte response to ischemic stroke

Pericytes are a previously understudied, but crucial cell type of the vessel wall. In the brain, pericytes are part of the neurovascular unit and are involved in the control and regulation of cerebral blood flow, formation and maintenance of the blood-brain-barrier, and initiating evolutionarily conserved inflammatory and wound healing responses in the context of injury and disease. Previous research suggests that pericytes are particularly susceptible to cerebral ischemia, dying almost immediately after reduction of blood flow and constricting the microcirculation thereby causing ‘no-reflow’ after ischemic stroke. If true, this would preclude any therapy to ameliorate stroke outcome. These previ-ous studies, however, lacked in vivo evidence, required deeper, more dynamic experimentation and pivotally, left many questions unanswered. Therefore, we used transgenic mice where pericytes ex-press enhanced green fluorescent protein (EGFP) under the control of the platelet derived growth factor receptor β promoter (PDGFRb), in tandem with 2-photon microscopy, laser speckle imaging, histological and transcriptomic analyses, to assess in detail the pericyte response to stroke. Firstly, we demonstrate a novel damaged pericyte phenotype, where half of all pericyte sub-types incur damage in the form of blebs during stroke in vivo, which persists acutely after reperfusion as a loss of cellular membrane integrity. 24 hours after stroke, we show that pericyte death occurs acutely, with a 25% loss in pericyte density in the ischemic territory, and 30% of remaining pericytes staining TUNEL+. Critically, this leaves half the pericyte population alive in the sub-acute phase (Day 3-7). Here, we demonstrate that pericyte survival is region dependent within the infarct core where neurons are eradicated. Despite this, we further show that pericytes respond to local reductions in population density by entering the cell cycle, increasing vessel coverage and upregulating transcriptional pro-files associated with the cell cycle, extracellular matrix deposition, and blood vessel morphogenesis. Functionally, we demonstrate during a transient one-hour filament middle cerebral artery occlusion, pericytes ubiquitously constrict the microvasculature in a spectrum of severity. 87% of pericytes con-strict their associated capillary by 25% on average in a sub-type and depth dependent manner. Spe-cifically, thin-strand pericytes constrict more than junctional or mesh pericytes and along with all other sub-types, continue to constrict their associated capillary after reperfusion. The consequences of pericyte constriction materialize in the form of entrapped non-flowing vascular elements, where during stroke, we show that one third of pericytes are associated with non-flowing vascular elements (capillary stalls), and this association persists acutely post-reperfusion. In doing so, we causally im-plicate acute pericyte dysfunction in the ‘no-reflow’ phenomenon after stroke. Importantly, 24 hours post-stroke, we no longer detect significant amounts of entrapped vascular elements at pericyte loca-tions, and concomitantly find that all cortical pericyte sub-type populations have dilated their associ-ated capillaries to pre-stroke levels, implying functional impairment of pericytes, not immediate peri-cyte death, causes constriction of the microvasculature in the ischemic cortex. Strikingly, in the sub-acute phase, we identify a previously unreported second phase of pericyte constriction, where thin-strand and junctional pericytes reconstrict their associated capillaries to a degree previously predict-ed by the severity of pericyte constriction during stroke. Finally, we demonstrate using laser speckle imaging that bi-phasic microvascular pericyte constriction post-stroke correlates with and contributes to large-scale reductions of blood flow within the ischemic cortex. Taken together, our research demonstrates that pericytes are more resistant to cerebral ischemia than previously believed and are causally implicated in the “no-reflow” phenomenon after ischemic stroke. Thus, pericytes may be functionally impaired by stroke in the ischemic cortex far earlier than they die, suggesting they represent a potential target for stroke therapy acutely after reperfusion of the occluded artery