Wide-field optical imaging of neurovascular coupling during stroke recovery
Abstract
Functional neuroimaging, which measure vascular responses to brain activity, are invaluable tools for monitoring and treating stroke patients both in the acute and chronic phases of recovery. However, vascular responses after stroke are almost always altered relative to vascular responses in healthy subjects and it is still unclear if these alterations reflect the underlying brain physiology or if the alterations are purely due to vascular injury. In other words, we do not know the effect of stroke on neurovascular coupling and are therefore limited in our ability to use functional neuroimaging to accurately interpret stroke pathophysiology. There is a need for animal models to investigate the effect of stroke on neurovascular coupling to aid in better interpreting the results from functional neuroimaging.
To that end, we first implemented a mouse photothrombotic stroke model that mimics the physiology of a human stroke and therefore has high clinical relevance. Mice were implanted with bilateral cranial windows to allow long term multimodal optical access. The occlusion procedure was performed in awake animals while simultaneously monitoring changes to cerebral blood flow. Our optimized photothrombotic stroke to the somatosensory forelimb region produced a sustained behavioral deficit in the contralateral forelimb that could be monitored longitudinally. Next, we implemented simultaneous imaging of neuronal activity, through fluorescent calcium imaging, and hemodynamics, through intrinsic optical signal imaging, to investigate neurovascular coupling during stroke recovery. Additionally, we identified a novel use for spatial frequency domain imaging to quantify the spatial extent of the stroke core.
Finally, we combined the mouse stroke model and imaging platforms to investigate the effect of stroke on neurovascular coupling. We found that acute stroke led to the abolishment of both calcium and hemodynamic responses to sensory stimulation. This elimination of response was associated with a loss of correlation between calcium and hemodynamic activity in the acute phase. To quantify neurovascular coupling, we modeled spatiotemporal hemodynamics by convolving neural activity and hemodynamic response functions obtained from deconvolution. Hemodynamic response functions from healthy animals were unable to model hemodynamics in the acute phase, suggesting neurovascular uncoupling. However, hemodynamics could be modeled in the chronic phase, indicating chronic recoupling. Acute stroke also resulted in increased global brain oscillations, which showed distinct patterns in calcium and hemodynamics, and the increase in contralesional calcium activity was associated with increased functional connectivity. We also show that early return of responses, neurovascular recoupling, and global oscillations were predictors of improved behavioral outcomes.