Multiscale modeling of blood flow control in the brain: How arteriolar- and capillary- mediated neurovascular coupling regulate cerebral blood flow

Introduction:: Blood is delivered to the brain via Penetrating Arterioles (PAs) that dive into the brain parenchyma before transforming into a vast capillary network. The cerebral microcirculation responds rapidly to neuronal activity and tightly regulates blood perfusion over a lifetime. Neurovascular communication (NVC) plays a critical role in normal brain function and is impaired in brain disorders. Due to their close proximity to every neuron, capillaries are ideally positioned to monitor neuronal activity and recent evidence suggests that are capable of sensing activity and initiating vasoactive signaling. How arteriolar- and capillary- mediated NVC intersect to coordinate blood flow distribution in the microvascular network has not been elucidated. The structural complexity and the inaccessibility to experimental observation makes modeling essential to gain insights into such mechanisms. In this study, we integrate detailed cell-level models into multicellular vessel segments to provide an in silico representation of capillaries and arterioles within a reconstructed large-scale vascular network. We examine the biophysical determinants of capillary- and arteriole- initiated vasodilatory signaling and their contribution to blood flow control in the brain.


Materials and Methods:: A cerebral microvascular network (Blinder et al. 2013) of the vibrissa primary mouse sensory cortex was reconstructed using graph theory principles. Models of capillary endothelial cells (cECs) and pericytes (PCs), and arteriolar endothelial (ECs) and smooth muscle cells (SMCs) capturing membrane potential and calcium dynamics are electrically coupled to form vascular segments (Fig. 1A) within the microvascular network that contains hundreds of thousands of vessel segments and millions of cells. Biomechanical models of arterioles and capillaries translate electrical and calcium signals into changes in vessel diameter and predict network hemodynamics in macroscale tissue volumes. Hemodynamic simulations predict pressure (P), flow (Q), and hematocrit (H_D) distributions along the dynamically changing network. The models incorporate rheological properties of the red blood cells and non-continuum effects via empirical formulas accounting for the Fahraeus-Lindqvist and plasma skimming effects. We use this theoretical framework to predict the contribution of capillaries and arterioles to blood flow distribution at the macroscale. We examine the effect of local stimulation (i.e. injected current or extracellular K⁺) of capillaries or PAs (Fig. 1B) on electrical conduction along the network and the resulting changes in diameters and blood flow distribution.


Results, Conclusions, and Discussions:: We simulate the net effect of NVC mediators by applying a hyperpolarizing current to an arteriole or a capillary segment, approximately 600 μm deep in the cortex (cortical layer IV). Representative simulation results (Fig. 2) depict changes in voltage and flow distributions as the local hyperpolarization spreads. At low levels of stimulatory current (1nA), the hyperpolarizing signal was confined to the vicinity of the stimulated region, inducing a modest increase in the local blood flow perfusion as the electrical signal dissipates at branching points. Higher levels of stimulation (3.5nA) were required for the vasodilatory signal to conduct over significant distances and reach upstream feeding vessels (Fig. 2).
Simulations also examined the role of K⁺ as a NVC mediator in arterioles and/or capillaries by increasing K⁺ concentration around SMCs and/or cECs within a 100μm³ tissue volume in cortical layer IV. The spread of the resulting electrical signal depends on the Kir channel density. This channel plays a dual role in the vasculature, as a K⁺ sensor and as an amplifier of propagating electrical signals. With sufficient Kir densities, stimulated cECs can initiate electrical signals that induce significant increases in blood flow. Depending on the Kir activity and the electrical coupling resistance of ECs, signals can remain localized or propagate over significant distances reaching upstream arterioles. This can yield confined increases in blood perfusion or significant macroscopic changes in blood supply, respectively.
The model explores the biophysical determinants of neuronally-induced electrical signaling in microvascular networks and explores the physiological relevance of proposed NVC mechanisms for regulating blood supply.


Acknowledgements (Optional): : This work is supported by the National Institute of Neurological Disorders and Stroke and the National Institute of Aging of the NIH under award number R01NS19971.

References (Optional): : Blinder, P., Tsai, P.S., Kaufhold, J.P., Knutsen, P.M., Suhl, H. and Kleinfeld, D., 2013. The cortical angiome: an interconnected vascular network with noncolumnar patterns of blood flow. Nature neuroscience, 16(7), pp.889-897.