Stroke is currently the leading cause of long-term disability and the fifth leading cause of death in the United States, and prevalence is on the rise due to increasing aging populations. Nearly 90% of strokes are ischemic, where there is a blockage of blood flow to the brain. Despite many clinical and pre-clinical attempts to develop novel treatments, there remains only one FDA-approved treatment for stroke, rtPA, which breaks down the clot, however only about 5% of patients are eligible to receive it. Numerous stage III clinical trials have failed, partially due to not taking major stroke risk factors into account during pre-clinical research. Given the poor prognosis and the fact that ‘time is brain’ when it comes to stroke damage prevention, there is a clear need to develop a rapid and long-lasting treatment to protect from impending ischemic stroke damage. Harnessing the brains endogenous mechanisms for protection via collateral therapeutics has become an important area of research and a potentially viable solution to prevent ischemic stroke damage. Our lab has discovered that when using a rat model of ischemic stroke (permanent middle cerebral artery occlusion), intermittent sensory stimulation treatment, when delivered within two hours of ischemic onset, completely protects the cortex from impending ischemic stroke damage. This protection is due in part to retrograde blood flow through collateral vessels and into the occluded middle cerebral artery. This collateral-based treatment is appealing as it is non-invasive and non-pharmacological, and has the potential to be delivered rapidly. This dissertation will present evidence of the translational capabilities of this treatment, and evidence demonstrating potential clinical limitations by testing it under conditions in which the brains vasculature has been altered, either due to anesthesia, stroke risk factors or the species used to model ischemic stroke. We found that the protective effects of this treatment are not dependent on the type of anesthesia utilized, and are long-lasting. Additionally, this treatment was tested in the presence of hypertension, the number one risk factor for ischemic stroke, and also in mice. The results suggest that this collateral-based sensory stimulation treatment does not prevent ischemic stroke damage in hypertensive conditions, in which the vasculature, including collateral vessels, are known to be impaired. Additionally, when a normotensive mouse model was tested, subjects were not protected. Although this strain reportedly has collaterals, they appear to be impaired under the ischemic conditions utilized in our lab. These mice may represent a human subpopulation that do not have fully developed or fully functional collateral vessels. If translational, this work serves to further characterize this treatment and potentially identifies groups of patients that may not exhibit complete protection from ischemic stroke, thus highlighting the importance of the collateral vasculature in the protection from ischemic stroke damage and emphasizing the need for further research on the collateral system to identify variables that can enhance its function.

ABSTRACT OF THE DISSERTATION

Large-scale spatiotemporal neuronal activity dynamics predict cortical viability in a rodent model of ischemic stroke

By

Ellen G. Wann

Doctor of Philosophy in Biological Sciences

University of California, Irvine, 2017

Professor Ron Frostig, Chair

Stroke is the fifth leading cause of death in the United States and frequently results in long-term disability (Benjamin et al., 2017). Improved stroke treatments are necessary as current therapeutic strategies are only effective for a subset of stroke patients and only reduce damage or its functional consequences. Our previous research demonstrated a novel ischemic stroke treatment in rodents in which intermittent sensory (whisker) stimulation delivered within 2 h (early stimulation) after distal permanent Middle Cerebral Artery occlusion (pMCAo) evokes neuronal activity that induces retrograde reperfusion via collateral blood vessels and protects the ischemic cortex from infarct. However, the same intermittent sensory stimulation results in exacerbated damage if delivered 3 or more hours after pMCAo (late stimulation) (Davis et al., 2011; Lay et al. 2010; Lay et al. 2011). The function of evoked neuronal activity in this model highlights the relevance of neuronal activity in the outcome of ischemic stroke, a role that has not received much attention in pre-clinical and clinical stroke research. Characterizing acute ischemic neuronal activity dynamics is additionally important for understanding the temporal and spatial development of ischemic pathophysiology. The main hypothesis of the dissertation was that particular features of post-ischemic neuronal activity critically predict the fate of the ischemic cortex. Using a 32 microelectrode array spanning depths of primary somatosensory cortex (S1) and neighboring cortical regions, electrophysiological recordings generated for the first time a continuous spatiotemporal profile of local field potentials (synaptic potentials recorded as local field potentials, LFP) and multi-unit activity (action potentials, MUA) from the MCA territory before (baseline) and directly after (0-5 hours) distal MCA occlusion in early stimulation, late stimulation, pMCAo alone, and surgical sham animals. Although evoked activity persisted for hours after pMCAo, spatiotemporal analyses revealed that large-scale spontaneous neuronal activity was disrupted after ischemic onset. In infarcted (pMCAo alone and late stimulation) animals, abnormal temporal coordination (synchrony) of large-scale neuronal activity and its underlying oscillations continued throughout the acute ischemic period. Conversely, the effects of ischemia on spatiotemporal synchrony were reversed over the course of the protective early sensory stimulation treatment. If translatable to human EEG, insights gained from evaluating neuronal activity dynamics and identifying electrophysiological signatures of acute ischemia may be potentially useful for supplementing diagnostic strategies and facilitating stroke treatment administration.

A distinguishing feature of animals is their ability to detect, internalize, and respond to

their surrounding environment. Despite the seeming simplicity of our basic sensory functions, the ability to sense requires a surprisingly complex neural architecture. Mammalian sensory systems follow several organizing principles including topography and cortical columns. The two organizing principles of topography and columns are exquisitely demonstrated in the somatosensory (touch) system of rodents. Unlike humans, rats spend most of their time in dark tunnels underground and rely on the rich tactile information provided by their whiskers. The array of whiskers on the snout project to a discrete, barrelshaped regions in the primary somatosensory cortex that match the pattern and orientation of the facial whiskers. This remarkable region is referred to as the rodent barrel cortex, and is perhaps one of the beststudied sensory cortices in the mammalian brain. However, our understanding of the rodent barrel cortex comes mostly from the anatomy and function of individual neurons, despite a general agreement that sensory function arises from large populations of neurons not single neurons. This dissertation focuses on the functional organization of barrel cortex from a mesoscopic perspective that encompasses large populations of neurons. A primary focus is the functional consequence of large, intracortical activity spreads which are ubiquitous in sensory cortex but are still poorly understood. Experimental evidence focusing on the rodent barrel cortex is presented demonstrating how large, intracortical activity spreads (point preads) underlie sensory coding and integration, provide a robust substrate for invariant sensory coding, and establish a large area of protection from ischemic attack.